<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Yoga Raditya Nala on Medium]]></title>
        <description><![CDATA[Stories by Yoga Raditya Nala on Medium]]></description>
        <link>https://medium.com/@radityanala?source=rss-26bf16164f14------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*NsKF7OU27RVXff1N-knzYg.jpeg</url>
            <title>Stories by Yoga Raditya Nala on Medium</title>
            <link>https://medium.com/@radityanala?source=rss-26bf16164f14------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 20:38:44 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@radityanala/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Operasi Seroja: A Case of Republican Hypocrisy?]]></title>
            <link>https://radityanala.medium.com/operasi-seroja-a-case-of-republican-hypocrisy-d72640556004?source=rss-26bf16164f14------2</link>
            <guid isPermaLink="false">https://medium.com/p/d72640556004</guid>
            <dc:creator><![CDATA[Yoga Raditya Nala]]></dc:creator>
            <pubDate>Sat, 31 Jan 2026 16:00:34 GMT</pubDate>
            <atom:updated>2026-02-01T02:44:03.179Z</atom:updated>
            <content:encoded><![CDATA[<h4>How Indonesia’s invasion of East Timor exposed the tension between constitutional ideals and Cold War security fears.</h4><p><strong><em>Disclaimer</em></strong><em><br>This essay represents the personal analysis and opinion of the author. It is not intended to incite hostility, misinformation, or political action. Reproduction, redistribution, or use outside this platform </em><strong><em>without explicit permission from the author is prohibited</em></strong><em>.</em></p><h3>Preamble</h3><p>I used to be the kind of person who was always interested in history, including my own nation’s history — Indonesia. I am actually amazed at how this republic managed to gain independence and maintain it for almost a century, especially with all its diversity. Many great nations with similar conditions, e.g., the republics of the Soviet Union and Yugoslavia — have collapsed. And I’m kind of proud to say that this republic has lasted longer than those great nations. Thanks to doctrine — or rather, civic education taught since the early years of schooling, this is certainly one of the very few reasons why this republic still stands today.</p><p>I can yap a lot about what I like about this republic, but this essay isn’t about that. It’s actually about an inconsistency I found during my civic classes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*AlSJaA1eyWAF8aTv5RwwYg.jpeg" /><figcaption><strong>Queen Juliana of the Netherlands signing the treaty that transferred sovereignty to Indonesia, 27 December 1949.</strong> Joop van Bilsen for Anefo, CC0, via <a href="https://commons.wikimedia.org/wiki/File:Soevereiniteitsoverdracht_Indonesi%C3%AB._Koningin_Juliana_tekent_de_akte,_Bestanddeelnr_934-5295.jpg">Wikimedia Commons</a>.</figcaption></figure><p>You see, this republic is a great example of how Third World countries struggled during their early years of independence, starting from internal power struggles such as ideological diversity to external threats from their former oppressors. The people eventually had enough of all that violence, which then manifested in the preamble of the republic’s constitution, called the Undang-Undang Dasar, first adopted in 1945. In its first alinea, it clearly states this line:</p><blockquote>Bahwa sesungguhnya <strong>kemerdekaan itu ialah hak segala bangsa </strong>dan oleh sebab itu, maka <strong>penjajahan di atas dunia harus dihapuskan</strong>, karena tidak sesuai dengan perikemanusiaan dan perikeadilan.</blockquote><p>Those words were the result of how people in this republic felt during the colonial era, forming a belief that any form of occupation was inherently inhumane and unjust. This republic had learned that occupying other states was a terrible thing to do — <em>or at least, that was what I thought at first.</em></p><p>Until decades later, <strong>history proved otherwise.</strong></p><p>In 1975, Indonesia launched a military operation to invade East Timor, which had declared itself an independent state. The irony is that this inhumane and unjust act — <strong>contradictory to the republic’s own constitutional principles</strong> — was backed by the United States and its allies, countries often labeled as <em>champions of freedom</em>.</p><p>I can’t help but ask a simple question:</p><blockquote>Was the military breaking the principle written in the preamble of the constitution? Or was this simply an example of hypocrisy at its finest?</blockquote><p>But before we rush to a conclusion and an answer, we need to understand <strong>what actually happened beforehand</strong>.</p><h3>Background</h3><p>Around the 1960s, this republic experienced internal power struggles. The state claimed that communists had attempted to overthrow the legitimate government through what later became known as <strong>the September 30th Movement</strong>. As a result, the republic declared that it had had enough of the Communist Party, locally known as Partai Komunis Indonesia (PKI), banned it, and issued a law limiting the spread of communism and left-wing ideology through MPR Decree No. XXV/1966.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IpX_KNNpVqxfOWnh1sbI5Q.png" /><figcaption><strong>Troops of the Indonesian Government in Jakarta just after 30 September Movement. </strong>Public domain via <a href="https://commons.wikimedia.org/wiki/File:Troops_of_the_Indonesian_Government_in_Jakarta_just_after_30_September_Movement_-_October_1965.png">Wikimedia Commons</a>.</figcaption></figure><p>Now moving to the eastern part of the archipelago, a new country was about to be born — East Timor. Under Portuguese colonial rule, <strong>numerous political factions </strong>began to emerge, such as the Timorese Popular Democratic Association (Apodeti), the Timorese Democratic Union (UDT), and the Revolutionary Front for an Independent East Timor (Fretilin).</p><p>At the time, <strong>President Soeharto denied any aspiration to incorporate the former Portuguese colony into the republic</strong>, instead arguing that the <strong>Timorese had the right to self-determination</strong>. In April 1972, Indonesia’s foreign minister stated that Indonesia would support an anti-Portuguese liberation movement if the Timorese people desired it. From my perspective, this position appeared <strong>consistent with the constitution’s anti-colonial principle.</strong></p><p>However, as the Cold War continued, the republic’s past experiences made <strong>Jakarta fear </strong>that the independence of East Timor could invite increased <strong>communist influence in the region</strong> — especially given its geographical proximity to the archipelago. This concern was later emphasized by President Soeharto during a recorded meeting with Australian Prime Minister Gough Whitlam in September 1974. He argued that since East Timor was not economically viable, it would inevitably have to rely on external support, potentially allowing communist powers such as China or the Soviet Union to intervene.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5wL_WSJu1iB__PkRH1GhoA.jpeg" /><figcaption><strong>Photo of Mao Zedong with various Soviet leaders, including Joseph Stalin, 1949.</strong> Helsingin Sanomat, Public domain, via <a href="https://commons.wikimedia.org/wiki/File:Mao,_Bulganin,_Stalin,_Ulbricht_Tsedenbal.jpeg">Wikimedia Commons</a>.</figcaption></figure><p>By mid-1974, <strong>Indonesia began supporting Apodeti </strong>through financial aid and radio broadcast assistance, in the expectation that the organization would gain popular support over other factions associated with communism. This action can be viewed as the republic’s concern for national security and regional stability. Moreover, if East Timor were to fall under a communist-led faction, it was <strong>perceived as a potential backdoor </strong>for the spread of insurgent movements.</p><p>However, things began to deteriorate in 1975.</p><p>In January 1975, UDT and Fretilin entered into a coalition — a move also supported by the Portuguese authorities. Two major political factions merged into one, something the republic clearly disliked. The situation quickly changed. Due to several Fretilin policies that made UDT feel increasingly threatened, <strong>UDT withdrew from the coalition </strong>by May 1975.</p><p>From March to July, Portuguese authorities organized local elections, in which <strong>Fretilin consistently won</strong>, securing approximately 50–55 percent of the vote. This outcome sent a clear signal to Jakarta that left-wing influence in the region was growing. Around the same time, in July 1975, <strong>President Soeharto publicly stated that East Timor could not sustain independence</strong> due to its economic inviability, suggesting that its future lay in integration with the republic.</p><p>Following the election results and the withdrawal of Portuguese colonial authorities <strong>in August 1975, the military finally acted</strong>. It covertly supported a coup initiated by UDT, which sparked a brief civil war. UDT forces seized key locations, including weapons from police arsenals, with covert Indonesian logistical and military assistance. However, by the end of August, Fretilin forces had regrouped, defeated UDT, and recaptured the areas previously seized.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Xr0diUWwQZFysBK6UiyUDA.jpeg" /><figcaption><a href="https://en.wikipedia.org/wiki/Jos%C3%A9_Ramos-Horta"><strong>José Ramos-Horta</strong></a><strong>, Foreign Minister of Democratic Republic of East Timor, 1976.</strong> Rob Bogaerts / Anefo, CC0, via <a href="https://commons.wikimedia.org/wiki/File:Persconferentie_met_Leonel_Andrade_(Fretilin)_en_Jose_Ramos_Horta_(Minister_Buit,_Bestanddeelnr_928-6310.jpg">Wikimedia Commons</a>.</figcaption></figure><p>In November 1975, the democratically elected leadership of East Timor formally proclaimed the Democratic Republic of East Timor. At that moment, <strong>Jakarta was left with two choices</strong>:<em> </em>to accept an independent East Timor under Fretilin control, or to <strong>intervene directly and integrate the territory into the republic</strong>. The latter option was chosen.</p><p><em>What began as anti-colonial support gradually transformed into an active anti-communist military operation.</em></p><h3>Operasi Seroja</h3><p>On December 7th, 1975, what was officially framed as a limited military action to secure the border between East Timor and the western part of the island escalated into a full-scale invasion. Under the codename “<em>Operasi Seroja”</em>, <strong>Indonesian forces launched a coordinated assault </strong>by land, sea, and air, involving both the military and police units. It became <strong>one of the largest military operations in the republic’s history</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0wSSHTXY8ZeMwcyyRyQk-Q.png" /><figcaption><strong>Indonesian Invasion of East Timor (1975–1979).</strong> J. Patrick Fischer and Borysk5, CC BY-SA 3.0, via <a href="https://commons.wikimedia.org/wiki/File:Timor_-_Indonesian_Invasion.png">Wikimedia Commons</a>.</figcaption></figure><p>The operation <strong>began with a naval bombardment of the city of Dili </strong>in the morning, followed by a <strong>land invasion conducted by air using paratroopers and by sea using marines</strong>. On December 10th, the invasion forces reached and captured the second-largest city, Baucau. <strong>By 1976, over 35,000 personnel had been deployed to East Timor.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/929/1*G4i6Fts-WiwrRDSEOnIG3Q.jpeg" /><figcaption><strong>President </strong><a href="https://commons.wikimedia.org/wiki/Gerald_Ford"><strong>Gerald Ford</strong></a><strong> greeting Indonesian Foreign Minister </strong><a href="https://en.wikipedia.org/wiki/Adam_Malik"><strong>Adam Malik</strong></a><strong> the day before the Indonesian </strong><a href="https://en.wikipedia.org/wiki/Indonesian_invasion_of_East_Timor"><strong>invasion of East Timor</strong></a><strong>. December 6th, 1975. </strong>David Hume Kennerly, Public domain, via <a href="https://commons.wikimedia.org/wiki/File:Ford_and_Malik.jpg">Wikimedia Commons</a>.</figcaption></figure><p>In the context of the Cold War, Western powers were determined to prevent another territory from falling under perceived communist influence, including East Timor — <em>they just could not afford another “Saigon”</em>. <strong>The republic did not act alone.</strong> The invasion was carried out with political and material <strong>backing from the United States and its allies</strong>. Numerous forms of assistance were provided, including the deployment of C-47 and C-130 aircraft, weaponry, training, and substantial financial aid, alongside diplomatic support.</p><p>The operation continued, and <strong>by 1978, the annexation was essentially completed.</strong></p><p>A <strong>large number of people died </strong>during the initial years of the invasion and occupation. <strong>More than 100,000 East Timorese are estimated to have died </strong>as direct casualties of the conflict — some killed by military operations, others dying in camps, from starvation, or while hiding. Activists and journalists were also among the victims, and many observers argue that widespread human rights violations occurred.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*rYdhpGEAI1C0CkbvsysqWA.jpeg" /><figcaption><strong>Indonesian soldiers pose in November 1975 in Batugade, East Timor with a captured Portuguese flag. </strong>Streitkräfte Indonesiens (Indonesian Armed Forces), Public domain, via <a href="https://commons.wikimedia.org/wiki/File:Capture_of_Portuguese_flag.jpg">Wikimedia Commons</a>.</figcaption></figure><p>Despite violations of self-determination and widespread civilian suffering, the invasion faced limited practical consequences internationally. While condemned in several United Nations resolutions, <strong>it was largely justified</strong> by its supporters as a matter of national and regional security — <em>at least until 1991.</em></p><h3>The Outcome</h3><p>Given the background of the conflict, we know that <strong>East Timor was an independent state at the time</strong>. Invading it under the banner of anti-colonialism <strong>simply does not make sense</strong>. However, when we take the republic’s past and its anti-communist doctrine into account, the <strong>action becomes understandable</strong> — as an attempt to <strong>secure national security </strong>and <strong>eliminate a perceived communist threat.</strong></p><p>The military may have violated what was written in the constitution’s preamble, but <strong>something far more dangerous was believed to be waiting</strong> for the republic and, in their view, needed to be eliminated through immediate action. Still, another question lingers in my mind.</p><blockquote>If the military had succeeded in eliminating the threat, why was East Timor occupied for decades until 1999? Why not install a new anti-communist government and allow the Timorese to decide their own fate?</blockquote><p>Beyond the president’s statement that East Timor could not stand on its own, another important factor must be considered.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6pSgmeARcJOHo_c8AYoPbA.jpeg" /><figcaption><strong>The inauguration of Arnaldo dos Reis Araújo and López da Cruz as Governor and Vice Governor of the Province of Timor-Timur.</strong> CC0 via <a href="https://commons.wikimedia.org/wiki/File:First_Governor_and_Vice_Gov_East_Timor_Innitiation.jpg">Wikimedia Commons</a>.</figcaption></figure><p>Indonesia has a <strong>long history of uprisings and rebellions</strong>. From DI/TII, PRRI/Permesta, and GAM in Aceh, to the ongoing conflict with OPM in Papua, <strong>Jakarta learned one crucial lesson</strong>: one wrong move could trigger another rebellion. Indonesia was never a nation bound by a single shared identity prior to independence — it was held together largely by the common experience of being oppressed under the Dutch East Indies. This made the republic inherently fragile.</p><p>After conquering East Timor, releasing it was no simple matter. <strong>Jakarta sought to avoid setting a precedent</strong>, especially amid ongoing conflicts and separatist movements. Releasing East Timor could serve as an example for other regions to seek separation from the republic, thereby threatening national stability.</p><p>But things changed in the 1990s. <strong>Following the Santa Cruz massacre, public opinion toward the republic started to shift</strong>. Many countries that previously supported the invasion, such as the <strong>United States and Australia, began to pressure Jakarta </strong>to withdraw its forces due to growing domestic and international criticism. <strong>The 1998 reform movement was also a catalyst</strong> for the eventual independence of East Timor.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*jC76qslxS6pmWlj3zKRCpw.jpeg" /><figcaption><a href="https://en.wikipedia.org/wiki/Bacharuddin_Jusuf_Habibie"><strong>Bacharuddin Jusuf Habibie</strong></a><strong> takes the oath of office as President of Indonesia.</strong> Office of the Vice President of the Republic of Indonesia, Public domain, via <a href="https://commons.wikimedia.org/wiki/File:Habibie_presidential_oath.jpg">Wikimedia Commons</a>.</figcaption></figure><p>Following President Soeharto’s resignation, under Habibie’s presidency, <strong>East Timor was finally given a referendum in 1999</strong>, which later resulted in independence, officially recognized in 2002. This decision was not made without reason. At the time, <strong>the republic was on the verge of collapse</strong> — its economy had fallen apart, political legitimacy was shattered, and the nation itself was in crisis.</p><p>However, in the years that followed, <strong>what Jakarta feared was, to some extent, proven true</strong>. Following East Timor’s independence, numerous rebellions and uprisings across the nation escalated — GAM in the west and OPM in the east. These rebellions did not begin because of East Timor, but <strong>East Timor provided them with a powerful example</strong> — <em>that the republic could, under certain circumstances, give up a territory.</em></p><h3>Conclusion</h3><p>You see, the answer to <strong>my initial question isn’t as simple as yes or no</strong> — it is a <strong>deep and complex issue</strong>. I believe some of these questions have already been addressed in the previous sections, but I will end this essay with another question.</p><blockquote>Was the republic <strong>wrong for invading East Timor?</strong></blockquote><p><strong>Maybe</strong>. Invading another independent state is, by definition, an inhumane and unjust act for any sovereign country to commit. However, when viewed from the perspective of national security, and considering the perceived threats at the time, <strong>it becomes understandable — a defensive act</strong>, <em>at least from Jakarta’s point of view.</em></p><blockquote>Does this pose an <strong>inherent hypocrisy toward the constitution?</strong></blockquote><p><strong>Perhaps it does as well</strong>. At that moment, the republic stood between two choices: to <strong>remain idealistic </strong>and adhere strictly to constitutional principles, or to <strong>confront </strong>what was believed to be a <strong>real and immediate threat to national stability</strong>. In this case, the republic chose the latter.</p><p><em>Because what would the constitution mean if the republic itself ceased to exist due to external threats?</em></p><p>I am not the kind of person who agrees with the idea that <em>the ends justify the means</em>. However, in this case, this republic has proven that it can endure longer than many great nations in the world — and that counts for something.</p><h3>Reference</h3><p>Abdulsalam, H. (2017, December 6). Upaya Indonesia Mencaplok Timor Lorosae lewat Operasi Seroja. <em>tirto.id</em>. <a href="https://tirto.id/upaya-indonesia-mencaplok-timor-lorosae-lewat-operasi-seroja-cBe1">https://tirto.id/upaya-indonesia-mencaplok-timor-lorosae-lewat-operasi-seroja-cBe1</a></p><p>HISTORY.com Editors. (2025, May 27). <em>Indonesia invades East Timor | December 7, 1975 | HISTORY</em>. HISTORY. <a href="https://www.history.com/this-day-in-history/december-7/indonesia-invades-east-timor">https://www.history.com/this-day-in-history/december-7/indonesia-invades-east-timor</a></p><p>Southgate, L. (2019). The Indonesian Invasion of East Timor. In <em>Policy Press eBooks</em> (pp. 25–70). <a href="https://doi.org/10.1332/policypress/9781529202205.003.0002">https://doi.org/10.1332/policypress/9781529202205.003.0002</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d72640556004" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From Wired to Wireless: How Humans Learned to Communicate Through the Unseen]]></title>
            <link>https://radityanala.medium.com/from-wired-to-wireless-how-humans-learned-to-communicate-through-the-unseen-4ca876b18c85?source=rss-26bf16164f14------2</link>
            <guid isPermaLink="false">https://medium.com/p/4ca876b18c85</guid>
            <category><![CDATA[internet]]></category>
            <category><![CDATA[networking]]></category>
            <category><![CDATA[wifi]]></category>
            <category><![CDATA[wireless]]></category>
            <dc:creator><![CDATA[Yoga Raditya Nala]]></dc:creator>
            <pubDate>Sun, 25 Jan 2026 12:31:42 GMT</pubDate>
            <atom:updated>2026-01-25T12:31:42.175Z</atom:updated>
            <content:encoded><![CDATA[<p>The internet has become an inseparable part of our modern life. For decades, we have relied on it daily — for work, for fun, and often both. It offers us a whole new world, a parallel reality. <em>But have you ever wondered </em><strong><em>what makes the internet so convenient for the rest of us?</em></strong></p><p>The answer lies in wireless technology. Not long ago, <strong>connecting meant plugging a cable </strong>into your laptop. With the arrival of wireless, everything changed. Now we simply sit down, connect, and the world is at our fingertips. Records show that wireless technology carries approximately 70.6% of all internet traffic, which is enormous.</p><p>And yet, it’s easy to forget how strange Wi-Fi really is. Every day, we stream movies and send messages through thin air. Unlike wired connections, where data obediently marches from one point to another, wireless — is just far more complicated. Information travels through the radio waves — <strong>things that humans can neither see nor hear</strong>, at all. That’s where it begins to feel like <strong>magic</strong>.</p><p>But this “magic”, it never come easily. Radio waves are chaotic, scarce, and insecure. They collide, interfere, and spill secrets into the open air — just like people talk. Worse, we never truly know what happens in the invisible medium. Building a reliable system on such a fragile foundation was almost absurd. And yet, through persistence and ingenuity, humans engineered protocols and security layers that turned impossibility into everyday reality.</p><h3>Before 1985: The Invisible Wall</h3><p>The foundation of the Wireless Local Area Network (WLAN) was laid prior to 1985. At that time, as the internet evolved, the need for connectivity in offices worldwide increased. <strong>One option was running cables through walls and ceilings</strong> — like what we always did before. But sometimes, the walls are simply too hard and the ceilings are often too high, that made wired installation a massive effort.</p><p>WLAN technology offered an alternative — a simpler solution. The question was simple,</p><blockquote>Instead of relying on cables to link devices, why not use another medium entirely, such as light or radio waves?</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/740/1*7a2OUbnLacA0uZY0hkdyfQ.jpeg" /><figcaption><strong>U.S. Navy signal lamp training aboard USS Mason (DE-529), 1944.</strong> <em>Official U.S. Navy photograph, National Archives. Public domain, via </em><a href="https://commons.wikimedia.org/wiki/File:Signal_lamp_training_color.jpg"><em>Wikimedia Commons</em></a><em>.</em></figcaption></figure><p>This concept wasn’t new to humanity. For centuries, people have harnessed light for communication — for example, naval signals transmitted through Morse code. Similarly, radio frequencies had long been employed to broadcast information through radio stations, or even for sending messages through wireless telegraph. WLAN simply built upon these established principles, transforming them into a modern, wireless way of connecting devices.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*EEZeUQojFejxk7vV" /><figcaption><strong>Wireless telegraph operator at a typewriter, c. 1930.</strong> George H. Clark Radioana Collection, Archives Center, National Museum of American History, <a href="https://www.si.edu/object/archives/components/sova-nmah-ac-0055-ref3703">Smithsonian Institution</a>.</figcaption></figure><p><strong>Around 1980</strong>, <strong>IBM’s Rueschlikon Lab </strong>in Zurich began <strong>research using infrared technology</strong>. The approach seemed promising — fast data transfer without the clutter of wires — exactly what we all wanted. However, infrared light couldn’t penetrate walls, which was a dealbreaker for most companies.</p><p>On the other side of the world, <strong>HP Palo Alto Lab </strong>in California built a prototype using <strong>Direct Sequence Spread Spectrum (DSSS). </strong>A related early experiment was conducted by Norman Abramson at the University of Hawaiʻi, known as <strong>the ALOHA system</strong>. Operating at a low data rate of about 9600 bps, it enabled long-distance wireless packet communication. While modern cellular networks use far more structured and efficient techniques, ALOHA’s random-access principles influenced later wireless communication systems.</p><p>The challenge was that a <strong>technology needed to reach at least 1 Mbps </strong>to be considered under the IEEE 802 standard, and non-optical wireless simply couldn’t meet that at the time. The limitation wasn’t due to a lack of human ingenuity — we already knew how to exploit higher frequencies and broader bands.</p><p>The real issue was that wireless signals lacked access to low‑cost, wideband spectrum. <em>And why was that?</em> Because the <strong>spectrum was tightly controlled by the Federal Communications Commission (FCC)</strong>, reserved for government and military use, and largely unavailable for commercial applications. In other words, the bottleneck wasn’t technological — it was regulatory and financial.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3Yzf8CrLX5kzwHrRy244hA.jpeg" /><figcaption><strong>FCC authorization of spread spectrum under Part 15 (Gen. Docket 81–413), 1985. </strong>National Archives and Records Administration. Public domain via <a href="http://archive.org/details/fcc-47-cfr-spread-spectrum">Internet Archive</a>.</figcaption></figure><p>But somehow, <strong>everything changed in the summer of 1985. </strong>Mike Marcus, the chief engineer at the FCC, finally released the unlicensed Industrial, Scientific, and Medical (ISM) bands, provided that users implemented interference management. It was a turning point — <em>the birth of modern wireless networking, and the moment when Wi-Fi’s story truly began.</em></p><h3>The Origin of 802.11 Standards</h3><p>With the FCC’s green light, innovation accelerated. <strong>In 1987</strong>, A small group at <strong>NCR in the Netherlands </strong>(later acquired by AT&amp;T) implemented <strong>DSSS in the 2.4 GHz band and achieved 2 Mbps</strong>. The technique spread the data signal across a wide bandwidth using a pseudo-noise code agreed upon by both transmitter and receiver. if decoded correctly, the signal was intended for that receiver — but if not, it appeared random and noisy. This mechanism, when used for multiple users, became Code Division Multiple Access (CDMA). It was robust, against interference and allowed error correction, since the carrier signal was wider than the actual data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/760/1*vpcQqpGSEiT8euNEui6Diw.jpeg" /><figcaption><strong>Direct-sequence CDMA spreading via XOR coding. </strong>Picture by Marcos Vicente, CC BY-SA 4.0, via <a href="https://commons.wikimedia.org/wiki/File:Generation_of_CDMA.svg">Wikimedia Commons</a>.</figcaption></figure><p>On the other side of the world, <strong>Proxim </strong>in <strong>California </strong>used <strong>Frequency Hopping Spread Spectrum (FHSS)</strong>, also <strong>achieving 2 Mbps. </strong>this approach used narrower bands and hopped between them according to a sequence agreed upon by transmitter and receiver. If synchronized, the signal could be decoded — but if not, it may appeared incomplete and noisy. Meanwhile, <strong>Apple </strong>pursued an <strong>optical solution using Diffused Infrared (DFIR)</strong>, which, though <strong>unable to penetrate walls</strong>, <strong>still achieved 2 Mbps</strong>.</p><p>These innovations laid the foundation for the IEEE 802.11 standard. DSSS and FHSS were standardized to operate under the 2.4 GHz ISM bands, while DFIR was also included. <strong>IEEE 802.11 thus became the first standard with three different physical layers and three different MAC mechanism</strong>: Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA), Request-to-Send/Clear-to-Send (RTS/CTS), and Point Coordination Function (PCF).</p><p><strong>Motorola </strong>experimented with <strong>licensed microwave </strong>bands around <strong>18 GHz</strong>, achieving <strong>data rates of approximately 10 Mbps</strong>. These early experiments explored high-frequency, line-of-sight digital links — <strong>precursors to the millimeter-wave </strong>frequencies <strong>later adopted in 5G networks</strong>. While not mobile cellular, these tests demonstrated the potential of higher-frequency bands for transmitting digital data, laying conceptual groundwork for modern high-speed wireless systems.</p><p>Academia contributed as well, <strong>Worcester Polytechnic Institute </strong>studied <strong>multipath propagation</strong>, where signals bounced off walls and arrived delayed at the receiver. <strong>Decision Feedback Equalization (DFE) </strong>was developed to predict and subtract delayed signals, mitigating multipath issues. Though promising, this approach wasn’t adopted into 802.11 due to market limitations and prediction errors — moreover, a new and better technology was emerged.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*E6gYSoWviHO_JqTiqgoFPw.jpeg" /><figcaption><strong>Waves on different paths interfere due to phase shifts. </strong>Picture by Peppergrower, CC BY-SA 3.0, via <a href="https://commons.wikimedia.org/wiki/File:Phase_shift.svg">Wikimedia Commons</a>.</figcaption></figure><p><em>Still remember CDMA? </em>It was quite robust against interference, but not very efficient. Every receiver had to constantly decode all the traffic, even though most of the time the data wasn’t meant for them.</p><blockquote>What if, instead of broadcasting signals to everyone and differentiating them by code, we simply assigned each user their own frequency?</blockquote><p>That approach was tried as well, and it became known as <strong>Frequency Division Multiple Access (FDMA). </strong>In FDMA, specific frequencies are allocated to specific users. This way, receivers don’t waste effort decoding irrelevant traffic — they only listen to the frequency assigned to them. Far more efficient, right?</p><p>But another problem quickly appeared. Frequency bandwidth is inherently limited. Once you start dividing it among many users, the available slices become too narrow. In practice, FDMA couldn’t scale to support large numbers of users simultaneously.</p><p>This exact problem laid the foundation for research into <strong>M‑ary orthogonal coding.</strong> The breakthrough was simple yet profound — <strong>orthogonal signals do not cancel each other out</strong>. This meant that multiple overlapping frequency bands could be used simultaneously without the risk of mutual interference.</p><p>From this realization, a bold idea emerged,</p><blockquote>What if we used orthogonal signals to transmit data at the same time, thereby increasing the data rate while conserving bandwidth?</blockquote><p>It turned out to be more than just an idea — it worked. <strong>In 1996</strong>, the <strong>Commonwealth Scientific and Industrial Research Organisation (CSIRO) </strong>in <strong>Australia </strong>developed and patented the approach that we now know as <strong>Orthogonal Frequency Division Multiplexing (OFDM).</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*T-3uLbt-b80dJYyl91JVMw.jpeg" /><figcaption><strong>Receiver unit used in CSIRO’s original WLAN prototype.</strong> Courtesy of the <a href="https://collectionsearch.nma.gov.au/icons/piction/kaui2/index.html#/home?usr=CE&amp;umo=56577859">National Museum of Australia</a>.</figcaption></figure><p>In OFDM, data is divided across numerous orthogonal signals, each occupying overlapping frequency bands. Because the signals are orthogonal, they can coexist without interference. The result is a system where data can be transmitted in parallel streams, saving both bandwidth and time, and dramatically increasing overall throughput.</p><h3>WLAN Technology Under 802.11 Standards</h3><p><strong>In 1997</strong>, after the recognition that DSSS, FHSS, and even early OFDM techniques could achieve speeds up to 2 Mbps, <strong>the first IEEE 802.11 technical standard was officially published</strong>. This marked the birth of Wi‑Fi as a standard, though adoption was limited due to its modest performance.</p><h4>IEEE 802.11a/b</h4><p><strong>The real breakthrough came in 1999</strong>. With <strong>OFDM </strong>introduced as a way to conserve bandwidth and increase efficiency, the first major amendment under 802.11 was released — <strong>IEEE 802.11a. </strong>Operating in the <strong>5 GHz unlicensed band</strong>, it offered <strong>data rates up to 54 Mbps</strong> — a dramatic leap forward. However, the higher frequency had its drawbacks — signals struggled to penetrate walls, and the use of multiple narrower channels made equipment costly to manufacture.</p><p>Around the same year, parallel research with a <strong>completely different approach was underway</strong>, leading to the release of <strong>IEEE 802.11b. </strong>It was <strong>neither faster nor technically superior </strong>to 802.11a. Running on the <strong>2.4 GHz band</strong>, it could only deliver <strong>up to 11 Mbps</strong> using <strong>DSSS</strong>, which was far inferior compared to OFDM.</p><p>But at the same time, it brought an advantage its predecessor couldn’t offer. With the lower frequency, <strong>signal coverage improved</strong> and became <strong>more resilient</strong> when penetrating walls. The use of DSSS instead of the brand‑new OFDM also made <strong>production costs far cheaper</strong>, which in turn allowed <strong>802.11b to enter the market easily</strong> and gain traction.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/0*GrhHr_0QzaTJoFdX.jpg" /><figcaption><strong>Apple AirPort Extreme installed in an iBook G4.</strong> Picture by Sgroey, CC BY-SA 4.0, via <a href="https://en.wikipedia.org/wiki/File:Apple_Airport_Extreme_802.11g_card.jpg">Wikimedia Commons</a></figcaption></figure><p>This was later proven when <strong>Apple released the first mass‑produced devices with Wi‑Fi connectivity</strong> — the AirPort wireless base station and the iBook. That bold move pushed Wi‑Fi adoption into the mainstream market. 802.11b wasn’t really a step backward in technology, but rather a deliberate trade‑off between speed and coverage.</p><h4>IEEE 802.11g</h4><p>We realized that OFDM was technically superior when implemented under the 802.11a standard. However, that standard had a hard time breaking into the market, which raised another question,</p><blockquote>What if, instead of implementing OFDM on the expensive 5 GHz band, we used it on 2.4 GHz instead?</blockquote><p>An effort to implement OFDM under the 2.4 GHz band to achieve higher data rates did indeed happen. Researchers planned to use more and more orthogonal frequencies to reach similar speeds to what 802.11a had achieved years earlier — and it worked. <strong>By 2003, a new amendment was released — IEEE 802.11g</strong>, capable of <strong>speeds up to 54 Mbps </strong>while still remaining <strong>affordable for the market</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/0*uD9S8187h1g127sY" /><figcaption><strong>Wireless 802.11g router with ADSL2+ modem. </strong>Picture by Macic7, CC BY-SA 3.0, via <a href="https://commons.wikimedia.org/wiki/File:Wirelessadsl2%2Brouter.dlink.dslg684t.JPG">Wikimedia Commons</a>.</figcaption></figure><h4>IEEE 802.11n: Wi-Fi 4</h4><p>Another breakthrough came with <strong>Multiple Input Multiple Output (MIMO) technology</strong>, which enables the simultaneous transmission and reception of multiple data streams. The principle behind this innovation is straightforward — <strong>by adding more antennas</strong>, we are theoretically able to send and receive multiple streams of data concurrently. One of the key advantages is the significant increase in speed and throughput. When different data streams are transmitted through multiple antennas at the same time, the overall throughput scales almost linearly. This technique is also known as <strong>spatial multiplexing</strong>.</p><blockquote>But how does the stream from one antenna not interfere with another at the same time?</blockquote><p>Well, here’s where the magic happens. Each antenna transmits a different data stream, and because signals bounce off walls, objects, and even people, they often arrive at the receiver with unique signatures — different delays, angles, and strengths. The real magic isn’t in the air, but in the mathematical foundation behind it, which separates one stream from another based on those parameters. And yes, orthogonal signals are also used to ensure that one stream doesn’t interfere with the other.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/400/1*Xc_tqp4wX6kuKFrUU2mhnA.png" /><figcaption><strong>MIMO channel matrix with multiple transmitters and receivers.</strong> Picture by Benbaum, Public domain, via <a href="https://commons.wikimedia.org/wiki/File:Kanalmatrix_MIMO.png">Wikimedia Commons</a>.</figcaption></figure><p>In practice, <strong>MIMO wasn’t only used to stream more and more data </strong>at the same time. It was also applied through <strong>Space‑Time Coding (STC), </strong>where multiple antennas transmit the same data with the hope that at least one version will be received correctly despite fading or interference. I’d call that redundancy at its finest. <strong>Combined with OFDM</strong>, the result was even better — <strong>delivering both robustness and high throughput</strong>.</p><p><strong>The year 2009 </strong>was another big milestone for the 802.11 standards. After six years of development since the previous release, <strong>the 802.11n standard</strong> finally emerged, combining MIMO with OFDM. It was capable of achieving <strong>data rates up to 600 Mbps </strong>across <strong>both the 2.4 GHz and 5 GHz bands</strong>. And for the first time, it <strong>introduced wider bandwidth — 40 MHz</strong>, doubling the regular 20 MHz channels.</p><h4>IEEE 802.11ad</h4><p><strong>Around 2012</strong>, under <strong>IEEE 802.11ad</strong>, the first attempt to reach data rates on the order of gigabits began. The technology experimentally used Extremely High Frequency (EHF), more specifically the <strong>60 GHz band</strong>, with <strong>bandwidth exceeding 2 GHz</strong>, and was <strong>able to achieve up to 4.6 Gbps </strong>(<em>some even claimed up to 7 Gbps</em>). Yet it never managed to attract a large market — almost like a secret technology the public never embraced. Today, <strong>similar millimeter waves have become an important part of 5G network </strong>development.</p><h4>IEEE 802.11ac: Wi-Fi 5</h4><p>Not long after the release of the previous Wi-Fi standard, <strong>IEEE 802.11ac was introduced in 2013</strong>. You may recall the RTS/CTS mechanism, originally designed to mitigate the hidden terminal problem. With 802.11ac, this mechanism gained renewed relevance with <strong>the adoption of beamforming</strong>. Unlike traditional directional antennas that radiate in a fixed direction, beamforming dynamically focuses the signal toward a specific receiver. While this improves performance, it also increases the likelihood that other transmitters outside the focused beam may fail to detect an ongoing transmission — leading to potential collisions. RTS/CTS can help mitigate this effect by coordinating access before data is sent, allowing nearby stations to defer even if they cannot sense the beamformed signal.</p><p>It was also a major breakthrough at the time. 802.11ac was <strong>the first standard to introduce MU‑MIMO (MIMO for multiple users at once),</strong> allowing multiple antennas to send and receive data simultaneously for different users. Combined with improved modulation techniques, it was able to reach <strong>data rates of up to 3.5 Gbps</strong>.</p><h4>IEEE 802.11ax: Wi-Fi 6</h4><p><strong>It was the year 2021 </strong>when the latest technology entered the market — <strong>IEEE 802.11ax. </strong>Unlike previous developments that focused mainly on increasing data rates and making signals more robust, <strong>this one was different</strong>. It acknowledged the reality that Wi‑Fi had become ubiquitous, while the demand for connectivity was rising at the same time. That growing demand created challenges in dense networks, often leading to signal and performance degradation, especially in large spaces such as stadiums and halls.</p><p>To overcome that challenge, 802.11ax introduced several innovations to <strong>improve both efficiency and reliability</strong>, especially <strong>in dense environments</strong>. Features such as <strong>OFDMA (Orthogonal Frequency Division Multiple Access),</strong> which allows multiple devices to share channels simultaneously, were central to this progress.</p><p>It also acknowledged the fact that the IoT world was evolving, and the need for connectivity among IoT nodes was increasing. Since IoT nodes rely heavily on battery life, 802.11ax introduced <strong>Target Wake Time (TWT),</strong> which <strong>helps conserve energy for IoT and mobile devices </strong>by scheduling when they wake to transmit or receive data. These were among the key breakthroughs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Eq_Z903zRTqWROk8V50s2w.jpeg" /><figcaption><strong>Setting up Ruckus R550 802.11ax controller-based access points.</strong> Personal picture, taken during Wi-Fi lab setup.</figcaption></figure><p>Collectively, alongside earlier innovations, this standard was able to reach <strong>data rates of up to 9.6 Gbps </strong>and operate across both the <strong>2.4 GHz and 5 GHz bands</strong> with consistent performance and smarter resource management. It paved the way for greater spectrum availability and reduced interference, setting the stage for the next generation of Wi‑Fi advancements.</p><h4>IEEE 802.11be: Wi-Fi 7</h4><p><strong>In July 2025</strong>, a new amendment was released — <strong>IEEE 802.11be</strong>, better known as Wi‑Fi 7. This standard <strong>represents a major evolution </strong>in Wi‑Fi technology, marked by <strong>the adoption of the 6 GHz band</strong>. It supports <strong>channel bandwidths up to 320 MHz</strong>, doubling the maximum available in previous generations. Alongside <strong>enhancements in MU‑MIMO, OFDMA, and improved modulation schemes</strong>, Wi‑Fi 7 aims for <strong>data rates of up to 40 Gbps.</strong></p><p>Another important innovation is the adoption of <strong>Multi‑Link Operation (MLO), </strong>which allows devices to <strong>connect across multiple bands </strong>(2.4, 5, and 6 GHz) <strong>simultaneously</strong>. This capability improves reliability, reduces latency, and maximizes throughput by intelligently distributing traffic across available channels. Since the technology is still new, it has yet to reach the mass market — but slowly and surely, it is beginning to make its way into everyday use.</p><h4>Conclusion</h4><p>And with that, we’ve arrived at the latest development in Wi‑Fi technology. Decades of progress have truly paid off. With the arrival of 802.11be, <strong>Wi‑Fi is now theoretically faster than CAT6e cables</strong>, and perhaps <strong>even</strong> <strong>comparable to CAT7</strong>. I never imagined that wireless would one day surpass copper — yet here we are.</p><p>From just 1 Mbps at the start of its journey to tens of gigabits under the 802.11be standard, Wi‑Fi’s evolution is a story of persistence. Each decade brought a new amendment, each solving the problems of its time, and each standing as a testament to human ingenuity — <strong>addressing yesterday’s challenges while anticipating tomorrow’s needs.</strong></p><p>And Wi‑Fi 7 is not the end. <strong>Research is already underway on IEEE 802.11bn</strong> — Wi‑Fi 8, planned for <strong>release around 2030</strong>. If you’re curious, it’s worth keeping an eye on what the next chapter of this remarkable story will bring.</p><h3>References</h3><p>Cisco. <em>What Is Wi‑Fi 7? </em>Accessed January 24, 2026. Available at: <a href="https://www.cisco.com/site/us/en/learn/topics/networking/what-is-wi-fi-7.html">https://www.cisco.com/site/us/en/learn/topics/networking/what-is-wi-fi-7.html</a></p><p>IEEE Standards Association. <em>The Evolution of Wi-Fi Technology and Standards</em>. Published May 16, 2023. Available at: <a href="https://standards.ieee.org/beyond-standards/the-evolution-of-wi-fi-technology-and-standards/">https://standards.ieee.org/beyond-standards/the-evolution-of-wi-fi-technology-and-standards/</a></p><p>Pahlavan, K., &amp; Krishnamurthy, P. (2021). <em>Evolution and impact of Wi‑Fi technology and applications: A historical perspective. International Journal of Wireless Information Networks,</em> 28(1), 3–19. <a href="https://doi.org/10.1007/s10776-020-00501-8">https://doi.org/10.1007/s10776-020-00501-8</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4ca876b18c85" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Another Round of APIE Camp, But as a Teaching Assistant]]></title>
            <link>https://radityanala.medium.com/another-round-of-apie-camp-but-as-a-teaching-assistant-c3a8b2de343d?source=rss-26bf16164f14------2</link>
            <guid isPermaLink="false">https://medium.com/p/c3a8b2de343d</guid>
            <dc:creator><![CDATA[Yoga Raditya Nala]]></dc:creator>
            <pubDate>Mon, 19 Jan 2026 05:31:27 GMT</pubDate>
            <atom:updated>2026-01-20T09:48:37.406Z</atom:updated>
            <content:encoded><![CDATA[<p>It’s the start of the year, and APIE just launched another camp — more specifically, the experimental one. Unlike the regular camp that’s often held abroad with participants from many nationalities, this time it was regional, with local participants. The goal was simple: to test the new curriculum and the new regional system before rolling them out in the regular camp. Ladies and gentlemen, I present to you the APIE Pilot Project on Wi-Fi.</p><p>Just a quick disclaimer before I continue yapping: the project itself actually consists of two courses, BGP and Wi-Fi. However, this article will only cover Wi-Fi, since I was the teaching assistant for that course. The project had three stages: the online course, the e‑workshop, and finally the camp — just like regular APIE stages.</p><h3>Stage 1: The E-Workshop</h3><p>The e‑workshop was held on January 8, 2026. It mainly focused on scanning nearby wireless networks using Wi-Fi Scanner by LizardSystem. Participants learned how to gather information about their environment and understand basic wireless terms such as RSSI, channel, band, and mode. Honestly, this was pretty useful. I discovered that my own home network was interfering with itself due to adjacent channel interference after using the scanner.</p><p>The next hands‑on was about Wi-Fi planning. Participants were divided into groups and guided directly by their teaching assistants. They were given a floor plan of the university meeting hall and tasked with placing access points to maximize coverage. But it wasn’t just about coverage — they also had to consider user density, since the hall can hold at least 3,000 people at once. They needed to adjust transmit power, channel, and band to prevent access points from interfering with each other.</p><p>Participants did well overall. Frankly, the real challenge in Wi-Fi planning was that Omada Design Hub doesn’t support project sharing, so only one participant per group could access it. Another issue was the floor plan itself. Without a proper building survey, we didn’t know the actual materials used, which made predictive design less accurate. And of course, each tool is tied to a single vendor — Omada mainly covers TP-Link products, while we were planning to use Ruckus access points. Still, it was a good introduction for participants to see how predictive design often fails in real environments. That’s exactly why active and passive scanning through surveys is so important — which I’ll explain later in the camp section.</p><p>As teaching assistants, we also had to adjust the worksheets — making sure they weren’t too hard, but not too easy either. Seeing participants perform well was a relief. Maybe that means our work wasn’t so bad, haha.</p><h3>Stage 2: The Camp</h3><p>The camp started on January 12, 2026, and ran for three days. Since not all participants were available for the offline camp, only those eligible and present in Malang could attend. Unlike the e‑workshop, the offline camp used real devices. We brought 18 Ruckus controller-based access points, along with 6 standalone access points from Direktorat Teknologi Informasi (DTI). We wanted participants to experience configuring real devices — something they might not get to do in basic faculty courses.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*so-Lu6G9LvJyO-gvoQu3dQ.jpeg" /><figcaption>Checking everything thoroughly.</figcaption></figure><p>As teaching assistants, we had plenty to prepare. We had to ensure the prerequisites were clear, the presentation material was solid, and all controller-based access points were assigned to the correct zones. At one point, we discovered that one access point didn’t match its zone — turns out we had brought the wrong device, but DTI staff quickly fixed it. Again, we had to adjust worksheets so participants could focus on material and hands‑on work without drowning in paperwork.</p><p>On the night before the first day, we held a big meeting to make sure everything was ready. We checked that worksheets matched the goals, presentations were good, and prerequisites were understandable. Groups were assigned by spin wheel, and I became TA for Group 2 — Farrel, Radit, Dinda, and Melody. Yep, the same Melody from the Manado camp — still not a fan of crimping, haha. Luckily, we had pre‑prepared cables, so she was spared — at least for a while. Prof. Husni also reminded us that advanced pilot projects shouldn’t involve crimping, since that’s already covered in the regular camp. So, lucky you, Mel.</p><h4>Day One — Survey and Design</h4><p>As usual, the camp began with an introduction from the committee, followed by a brief lecture on wireless networks and radio frequency. Honestly, kind of boring for most participants — and for me too — but necessary to set the foundation for the hands‑on sessions.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*teQAZi_jDlTLjb11YHEOlg.jpeg" /><figcaption>Where it all began.</figcaption></figure><p>The first hands‑on was network scanning. Groups were assigned floors in the faculty building by spin wheel. My group got the 2nd floor — not the easiest, but definitely simpler than the 3rd or 4th floors, which had many rooms. The 2nd floor had a few classrooms along the edges and a central void facing the lobby below.</p><p>Participants performed both passive and active scanning across the area using Netspot. Due to limited license keys, participants used version 2 while teaching assistants had version 3, which caused some confusion with the UI. Our group even used the wrong coverage mode at first, but we still managed to gather useful data such as coverage results, RSSI, SIR, and more. After that came break time — my favorite part, since it’s when we could interact with participants outside of tasks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*N5a5d0FkTQ7WHBjpgqXIeg.jpeg" /><figcaption>Predictive design that wasn’t always accurate.</figcaption></figure><p>The second session was predictive design using Omada Design Hub. As mentioned earlier, it only supports TP-Link products, so we used TP-Link EAP650 as the equivalent of Ruckus R550. Before designing, participants analyzed issues with the faculty’s current setup. In our case, not all areas had 5GHz coverage, which slowed speeds despite having five access points. Each group was only given three access points to work with, so even modest improvements saved resources.</p><p>One concern was that we only had a single shared account for all groups — under my email, no less. This was intentional, so participants could focus on AP placement rather than wall materials. Teaching assistants handled the wall setup beforehand to save time. What surprised me was my group’s dedication. While most groups left at 3 p.m., my group stayed behind to finish their worksheet — and I really appreciated their effort.</p><h4>Day Two — Network Deployment</h4><p>Day two was supposed to be simple — a lecture from DTI staff before lunch, then network deployment afterward. I thought it would be quick — cables were ready, brackets mounted, so it should’ve been plug‑and‑play. But of course, things didn’t go as planned.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xRi0Tj7TvfHTyYRuAtiOWg.jpeg" /><figcaption>Network devices fired up.</figcaption></figure><p>Before the network deployment, during the lecture, participants were also allowed to fire up their network devices. This was done to ensure they understood the controller UI and how to operate it. That familiarity would certainly be helpful later when deploying their own devices — like adjusting transmit power, channels, and other settings.</p><p>After all the lectures, it was finally time for the network deployment. There were only two staircases available, so I was glad our group was on the 2nd floor and could go first. The first AP was mounted successfully, cable worked fine, and it appeared on the controller immediately. But the mounting looked loose, so we re‑screwed it, which took longer than expected.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZbnDesA1P11cWQJIVfH65A.jpeg" /><figcaption>Struggle of mounting the APs.</figcaption></figure><p>The second AP was a nightmare. The PoE worked, the AP powered on, but it never appeared online. Worse, no green lights on the switch — clear sign of a faulty cable. And we had no backup for the 40m cable. We ended up doing the one thing Melody hates most — crimping. She did it anyway — maybe she’s made peace with it — but it still didn’t work. The RJ45 connectors had wide gaps, making wires overlap easily. By 3 p.m., most groups were leaving while we were still stuck.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*75GzmzsR7OELQ7OoHHcyRQ.jpeg" /><figcaption>You could see the wide smile on Melody’s face — I think she’s finally starting to love crimping /jk.</figcaption></figure><p>While I searched for better connectors, the group mounted the third AP, which thankfully worked fine. Some worked on the worksheet, others configured the controller. For reference, you can configure APs even if they’re offline — once they come online, they fetch the configuration automatically. One funny thing did happen that day. Because of improper planning in the wiring design — and the fact that cable lengths were limited — we ended up passing cables through the void, literally flying across the air. At that point, I was just afraid someone might fall off because of it, haha.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*m4DruJtjwlpa3C5SfBhPSA.jpeg" /><figcaption>Cable flying through the void.</figcaption></figure><p>I eventually found good connectors. I crimped one end — still no luck. Tried the other end, with help from DTI staff, but one or two pins still failed. At that point, I was ready to give up, thinking the cable was cut somewhere in the middle. But we insisted on trying again — and voila, it finally worked. Even with a few dead pins, the AP showed up online. Afterward, participants did active scanning to confirm the deployment was better than before. It was a long, exhausting day. By 5 p.m., the TAs had a brief meeting. Melody and Dinda stayed behind to finish tasks, while others packed up. Most groups had left long before. That day was the most exhausting of the week — for me and probably for everyone else too.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4LPJ5-n89APnGiOWYbyz3w.jpeg" /><figcaption>Prof. Husni doing his thing.</figcaption></figure><p>One thing always stood out at the end of that day, something that made Melody and me laugh in silence. It was always Prof. Husni, either playing around with the 8K display in our faculty or tinkering with the sound system. It wasn’t much, but it kept us entertained while we finished the worksheet.</p><h4>Day 3 — The Enterprise Setup</h4><p>Everyone was late, but Melody ended up being the tie breaker — once she arrived, our group became the first to be completely present. The focus was on RADIUS server setup, which was relatively easy. We just had to configure a new WLAN with WPA Enterprise and connect it to the RADIUS server we had installed earlier. Our group was the first to finish.</p><p>However, there was this one funny thing that happened. Instead of creating a new WLAN configuration, we accidentally modified the existing one to switch over to WPA2 Enterprise. That change immediately caused the RADIUS server — previously connected through WPA2 Personal — to disconnect. The result? No one could connect to our network anymore, because authentication was now tied to a RADIUS server that was unreachable.</p><p>It felt exactly like the classic mistake of turning on a firewall on your remote VM but forgetting to whitelist port 22. Suddenly, you’re locked out of your own system. My lecturer called it a “lifetime experience” — and honestly, it really was.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dOGqFfqPeVMlMwdEFDPZiw.jpeg" /><figcaption>Configuring what needed to be configured.</figcaption></figure><p>Back at the venue, the group worked on their presentation and worksheet, both due by 1 p.m. Thanks to their dedication, everything was completed on time. We presented second, summarizing our three days of work. One mistake slipped through — we used a different device for validation scanning on day two. That was my fault — I forgot to remind them. Still, not many questions were asked afterward. Nothing outstanding, but each group had unique cases based on their assigned floors, which was fun to see.</p><p>Finally, the awarding ceremony. I wasn’t expecting anything, but somehow we got 2nd place. Everyone was shocked — Dinda and Melody thought their presentation was the worst. Maybe it wasn’t the flashiest presentation, but what mattered was the effort they poured in. Staying late, fixing cables when things went wrong, and refusing to give up even when everyone else had already left — that dedication spoke louder than slides or polished delivery.</p><p>Getting second place felt less like a prize and more like recognition of their persistence. For me, it was a reminder that being a TA isn’t just about guiding the technical work — it’s about standing alongside a group, sharing the struggles, and celebrating the small victories. In the end, that’s what made this camp special — not the perfect setups or flawless worksheets, but the resilience, teamwork, and the stories we’ll carry long after the Wi-Fi signals fade.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uLiRSABH4SOKWidBgEJ6mQ.jpeg" /><figcaption>Unexpected ending (love you guys).</figcaption></figure><p>Thank you for coloring the very first week of 2026, guys — it wouldn’t have been the same without your energy and dedication.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c3a8b2de343d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[MQTT: Protokol Ringan yang Menghubungkan Dunia IoT]]></title>
            <link>https://radityanala.medium.com/mqtt-protokol-ringan-yang-menghubungkan-dunia-iot-a41ed96730a9?source=rss-26bf16164f14------2</link>
            <guid isPermaLink="false">https://medium.com/p/a41ed96730a9</guid>
            <category><![CDATA[mqtt]]></category>
            <category><![CDATA[iot]]></category>
            <dc:creator><![CDATA[Yoga Raditya Nala]]></dc:creator>
            <pubDate>Sat, 03 Jan 2026 15:05:40 GMT</pubDate>
            <atom:updated>2026-01-03T15:09:26.623Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EDOYZc1CO0tygZDuYO5gug.png" /><figcaption>Image Generated with Microsoft Copilot</figcaption></figure><p>Kita semua tahu HTTP(S) adalah protokol yang menjadi tulang punggung web yang kita pakai tiap hari, termasuk website Medium yang teman-teman pakai untuk baca tulisan ini. Tapi tahukah kamu, kalau dunia IoT justru jarang menggunakan protokol ini? Di balik sensor dan perangkat kecil yang saling bicara lewat internet, ada protokol lain yang jauh lebih ringan dan sering dipakai, yaitu <em>Message Queuing Telemetry Transport</em> (MQTT).</p><h3>Introduction to IoT and Protocols</h3><p>Ketika mendengar istilah IoT, kebanyakan dari kita mungkin akan membayangkan tentang sensor-sensor yang terhubung ke sebuah microcontroller seperti Arduino atau ESP32. Tapi sebenarnya, sekumpulan sensor saja belum membuatnya menjadi “IoT”. Justru kemampuan perangkat itu untuk saling bertukar data, entah dengan server atau perangkat lain melalui internet dengan sebuah protokol tertentu yang menjadikannya bagian dari <em>Internet of Things</em>.</p><blockquote>Tapi apa sih protokol itu?</blockquote><p>Internet itu luas—mungkin seluas dunia, kita bisa melakukan berbagai hal, termasuk ngobrolin apapun yang kita mau. Tapi ada satu hal penting supaya apa yang kita obrolkan atau lakukan bisa dipahami satu sama lain. <strong>Kita perlu kesepakatan atau standar untuk berbicara</strong>, misalnya bahasa.</p><p>Dalam sebuah bahasa, tentunya ada <em>set of rules</em> atau aturan khusus yang harus dipatuhi seperti <em>grammar</em>, yang disetujui oleh banyak pihak, supaya saling mengerti apa yang dibicarakan. Nah, protokol itu seperti bahasa. Protokol <strong>mengatur bagaimana cara banyak pihak saling berkomunikasi</strong>, sehingga satu sama lain saling mengerti apa yang ingin dikomunikasikan. Sama seperti bahasa, protokol itu ada banyak. Misalnya HTTP(S) yang biasanya dipakai di website, MQTT yang akan kita bahas, CoAP yang juga sering dipakai di IoT, dan lain sebagainya.</p><h3>What is MQTT?</h3><p>Kalau sebelumnya kita udah bahas mengenai apa itu protokol dan macam-macamnya, sekarang waktunya kita fokus ke MQTT. Tapi sebelum lanjut, kita perlu tahu dulu, apa sih sebenernya MQTT itu dan apa bedanya dengan berbagai protokol lain?</p><blockquote>“MQTT (Message Queuing Telemetry Transport) is a lightweight, publish-subscribe based messaging protocol designed for resource-constrained devices and low-bandwidth, high-latency, or unreliable networks.”<br>— (EMQX, 2025)</blockquote><p>Kalau kita lihat dari definisi di atas, maka kurang lebih MQTT itu adalah protokol yang ringan dengan <strong>sistem <em>publish-subscribe</em></strong> yang memang didesain <strong>untuk <em>device-device</em> dengan <em>resource </em>terbatas</strong>, misalnya bagian <em>network-</em>nya. Tak heran, protokol ini banyak dipakai oleh <em>device-device</em> kecil untuk keperluan <em>internet of things</em> seperti ESP32 yang memiliki <em>resource </em>terbatas, tetapi perlu mengirimkan data melalui internet.</p><h3>Publish-Subscribe Mechanism</h3><p>Jika di dalam protokol HTTP ada <em>request </em>dan <em>response</em>, maka di MQTT konsepnya sedikit berbeda. Dalam MQTT, terdapat konsep <em>publish </em>dan <em>subscribe</em>. <strong><em>Publish </em>digunakan ketika kita ingin mengirimkan data</strong> ke <em>topic </em>tertentu, sementara <strong><em>subscribe </em>digunakan ketika kita ingin mendapatkan data</strong> dari <em>topic </em>tertentu.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TPUwWf6WEKqiQxSw6BhOXQ.png" /><figcaption>The Publish-Subscribe Mechanism</figcaption></figure><p>Dalam implementasi sistem IoT yang membaca data dari sensor dan ingin mengirimkan data tersebut ke server, maka biasanya node IoT tersebut akan <em>publish </em>data sensor-sensor ke <em>topic </em>tertentu. Sementara itu, <em>end user</em> yang ingin melakukan visualisasi atau sekadar melihat data dari node tersebut, perlu melakukan <em>subscribe</em> ke <em>topic </em>yang sama.</p><p>Mekanisme ini bisa dianalogikan dengan grup Whatsapp yang punya dua aktor utama, <em>admin </em>yang punya hak akses untuk kirim pesan, dan <em>member </em>yang hanya punya hak akses untuk baca pesan. Dalam konteks tersebut, <strong><em>admin </em>merupakan seorang <em>publisher</em></strong><em> </em>yang mampu mengirim pesan (data), sementara <strong><em>member-member</em> lain merupakan <em>subscriber</em></strong><em> </em>yang hanya dapat menerima atau membaca pesan (data) tersebut. Kedua aktor tersebut harus berada di grup (<em>topic</em>) yang sama untuk dapat mengirim atau menerima pesan (data).</p><h3>Client &amp; Broker</h3><p>Jika dalam protokol HTTP ada <em>client </em>dan <em>server</em>, maka di <strong>MQTT ada <em>client </em>dan <em>broker</em>.</strong> Keduanya tentu memiliki peran yang berbeda. <strong><em>Client</em> merupakan aktor yang berada di bagian ujung (<em>edge</em>) </strong>dan terkoneksi dengan <em>broker</em>, bisa berupa <em>publisher </em>ataupun <em>subscriber</em>. Sementara itu, broker bertugas untuk mengatur lalu lintas data yang lewat, meneruskan data dari <em>publisher </em>ke para <em>subscriber</em>.</p><p>Dalam implementasinya, client ini bisa jadi node IoT ataupun <em>end user</em>, sementara <strong><em>broker</em> adalah <em>provider</em></strong><em> </em>yang menangani koneksi, <em>topic</em>, dan data-data yang saling dipertukarkan, misalnya EMQX, Mosquitto, HiveMQ, dan lain sebagainya.</p><h3>Topic</h3><p>Nah, topic ini sebenarnya sudah berkali-kali aku singgung sebelumnya. Singkatnya <strong><em>topic </em>ini semacam <em>channel </em>atau wadah</strong> untuk menampung data sementara sebelum dikirimkan oleh <em>broker </em>ke <em>subscriber</em>. Mirip dengan URL, <em>topic </em>biasanya berbentuk <em>path </em>dan memiliki hierarki dengan pemisah berupa tanda slash (/), misalnya sebagai berikut.</p><pre>esp32/1/temperature<br>esp32/1/humidity<br>esp32/2/ultrasonic<br>raspi/1/ph</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aZaEmd5ZXxyt-H1EIhiiJw.png" /><figcaption>How Topics Interconnect Publishers and Subscribers</figcaption></figure><p>Nama <em>topic </em>yang digunakan sangat fleksibel dan tergantung kebutuhan. Pada contoh di atas, aku membagi <em>topic </em>berdasarkan jenis node (ESP32 dan Raspberry Pi), lalu terdapat <em>sub-topic</em> nomor <em>node</em>, dan terakhir adalah <em>sub-topic</em> jenis data sensor. Namun, kembali lagi ke kebutuhan kalian, sekali lagi, <strong>nama <em>topic </em>yang bisa digunakan sangat fleksibel</strong>, asal menggunakan slash sebagai pemisah antar <em>level topic</em>.</p><p>Idealnya, masing-masing data yang dikirim dari <em>node </em>memiliki <em>topic</em> masing-masing, sehingga akan lebih mudah apabila ingin melakukan visualisasi tanpa <em>decode payload</em> terlebih dahulu.</p><blockquote>“Hah? Gimana-gimana?”</blockquote><p>Misalkan <em>node </em>IoT kita menggunakan sensor DHT dan kita perlu mengirim data <em>temperature </em>dan <em>humidity </em>ke MQTT <em>broker</em>, ada beberapa opsi yang bisa kita lakukan. Pertama, kita bisa saja “membungkus” kedua data tersebut menjadi satu <em>payload </em>saja, yang kemudian kita kirim ke satu <em>topic</em>, misalnya sensor/dht dengan payload sebagai berikut.</p><pre>{<br>  &quot;temperature&quot;: &quot;24.6&quot;,<br>  &quot;humidity&quot;: &quot;50.2&quot;<br>}</pre><p>Cara tersebut sebenarnya <strong>bisa saja dilakukan</strong>, nantinya data yang dikirim oleh <em>broker </em>ke <em>subscriber</em> juga dalam bentuk <em>payload </em>tersebut. Namun, sebagai konsekuensi, data tersebut <strong>tidak bisa langsung dibaca</strong> oleh tools visualisasi (e.g., Node-RED Dashboard), tetapi perlu dilakukan pemrosesan terlebih dahulu untuk mengambil value yang sesuai.</p><p>Tentunya berbeda dengan <em>approach </em>yang mengirimkan masing-masing data secara <em>independent </em>ke <em>topic </em>berbeda, misalnya sensor/temperature dan sensor/humidity dengan value yang berupa <em>plaintext</em>, sehingga bisa langsung dibaca oleh tools visualisasi tanpa adanya <em>overhead</em> berupa <em>parsing payload</em> sebelum divisualisasikan. Sehingga, menurut aku, <strong>jauh lebih efisien menggunakan satu <em>topic</em> untuk satu jenis data</strong> apabila data yang dibutuhkan memang tidak memerlukan atribut tambahan.</p><h4>Hierarchy &amp; Wildcards</h4><p>Kalau kalian ingat, sebelumnya aku pernah <em>mention </em>kalau sebenarnya MQTT <strong><em>topic</em> ini punya hierarki yang dipisahkan oleh tanda slash</strong>. Setiap kita menggunakan tanda slash, kita seolah membuat <em>sub-topic</em> baru. Kalau dalam contoh di atas, level paling tinggi (<em>root</em>) adalah nama <em>node-</em>nya, e.g., esp32, raspi, etc. Setelah itu, masing-masing tipe <em>node </em>punya <em>sub-topic</em> sendiri, yaitu identifier dari <em>node</em>. Masing-masing <em>node </em>juga punya <em>sub-topic </em>sendiri lagi, yaitu berbagai jenis data yang ada pada node tersebut.</p><pre>[topic]/[sub-topic]/[another sub-topic]</pre><blockquote>“Terus, fungsinya hierarki ini apa?”</blockquote><p>Sekarang coba bayangkan, semisal kita ingin mendapatkan data temperature dari semua sensor, maka mungkin langkah yang kalian pikirkan adalah dengan iterasi/<em>subscribe </em>secara manual topic di bawah ini sebanyak jumlah node. Tentunya, cara tersebut <strong>cukup rumit</strong> dan <strong>tidak efisien</strong>.</p><pre>esp32/1/temperature<br>esp32/2/temperature<br>esp32/3/temperature<br>...<br>esp32/x/temperature</pre><p>Untungnya, <strong>MQTT menyediakan <em>topic wildcard</em></strong> yang memungkinkan kita untuk <strong><em>subscribe</em> ke banyak <em>topic </em>sekaligus</strong> dalam waktu yang sama <strong>dengan <em>pattern </em>tertentu</strong>. Ada dua tipe <em>wildcard </em>yang bisa digunakan, yaitu sebagai berikut.</p><ul><li>+ merupakan <em>single level wildcards</em>, bisa dipakai di tengah <em>topic query</em></li><li># merupakan <em>multi levels wildcards</em>, dipakai di akhir <em>topic query</em></li></ul><p>Apabila diimplementasikan ke kasus di atas dengan menggunakan <em>wildcards</em>, maka bisa menggunakan topic esp32/+/temperature. <em>Pattern </em>tersebut akan <strong>mengambil semua <em>topic </em>dengan hierarki <em>topic </em>awal</strong> berupa esp32 <strong>dan diakhiri dengan</strong> temperature, tanpa memedulikan <em>identifier </em>dari <em>node-</em>nya.</p><pre>esp32/+/temperature<br><br>esp32/1/temperature<br>esp32/2/temperature<br>esp32/3/temperature<br>esp32/4/temperature<br>...<br>esp32/anything/temperature</pre><p>Kasus lain yang mungkin adalah ketika kita ingin <strong>mendapatkan data dari semua <em>node </em>dan semua <em>sensor</em></strong>. Dalam hal ini, kita bisa melakukannya dengan mudah menggunakan bantuan <em>multi levels wildcards</em>, seperti esp32/# atau raspi/# yang akan mengambil data dari topik dengan awalan esp32 atau raspi tanpa memedulikan <em>sub-topic</em> di bawahnya. Dikarenakan <em>multi levels wildcards</em> ini akan mengambil data dari semua sub-<em>topic </em>di bawahnya, maka <em>wildcards </em>jenis ini hanya bisa digunakan di akhir <em>topic</em> saja, tidak di tengah-tengah <em>topic </em>seperti <em>single level wildcards</em>.</p><pre>esp32/#<br><br>esp32/1/temperature<br>esp32/1/humidity<br>esp32/2/ultrasonic<br>esp32/3/ph<br>esp32/4/temperature<br>...<br>esp32/anything/anything</pre><p>Ketika menggunakan <em>wildcards</em>, selain <em>payload</em>, umumnya <strong>MQTT <em>broker </em>juga akan mengirimkan <em>topic </em>asal</strong> dari <em>payload </em>tersebut, sehingga mempermudah kita dalam memisahkan data dari satu <em>topic </em>dengan <em>topic </em>yang lain.</p><h4>System Reserved Topics</h4><p>Selain itu, <strong>MQTT juga punya reserved <em>topic</em></strong><em> </em>dengan <em>prefix </em>$SYS yang tidak bisa digunakan sebagai topik lain. Ada beberapa <em>reserved topics</em>, diantaranya adalah sebagai berikut.</p><ul><li>$SYS/broker/clients/connected menunjukkan berapa banyak <em>client </em>yang terhubung ke <em>broker</em></li><li>$SYS/broker/clients/disconnected menunjukkan berapa banyak <em>clients</em> yang telah <em>disconnected</em></li><li>$SYS/broker/clients/total menunjukkan total <em>client </em>yang pernah terhubung</li><li>$SYS/broker/messages/sent menunjukkan total <em>message </em>yang terhubung</li><li>$SYS/broker/uptime menunjukkan berapa lama MQTT <em>broker </em>telah berjalan</li></ul><p>Topic-topic tersebut sangat berguna karena dapat <strong>menunjukkan bagaimana kondisi atau performa MQTT <em>broker</em></strong><em> </em>yang sedang berjalan saat ini.</p><h3>Quality of Service (QoS)</h3><blockquote>“The term “Quality of Service” in MQTT defines the reliability guarantee level for message delivery between devices.”<br>— (HiveMQ, 2024)</blockquote><p>Quality of Service atau QoS ini <strong>merujuk pada “kepastian” pengiriman data</strong> oleh broker. Kalau kalian ingat, mekanisme <em>publish-subscribe</em> ini tidak memiliki mekanisme <em>response </em>seperti pada HTTP, yang mana membuat <em>sender/publisher</em> tidak tahu dengan pasti apakah pesannya benar-benar terkirim atau tidak.</p><p>Tingkat Quality of Service dapat <strong>diatur berdasarkan tiga level</strong> berikut. Sebelumnya, QoS ditentukan per-<em>link</em>, bisa jadi <em>sender </em>merupakan <em>publisher </em>ataupun <em>broker</em>, dan <em>receiver </em>bisa jadi merupakan <em>broker </em>ataupun <em>subscriber</em>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/742/1*9srpq_sUyX_KgM12SqPnsw.jpeg" /><figcaption>QoS 0: The Fire and Forget Mechanism</figcaption></figure><p>Pertama dan paling sederhana, <strong><em>at most once</em> (QoS 0), berbasis <em>fire and forget</em></strong>, tidak ada mekanisme <em>acknowledgement </em>dari <em>receiver</em>, <em>publisher </em>tidak peduli pesannya berhasil diterima atau tidak. Dengan cara ini, data bisa sampai jauh lebih cepat. <em>In fact, </em><strong><em>this is the fastest method</em></strong>. Namun, seperti kata orang, tidak ada sistem yang sempurna.</p><p>Mekanisme ini memang yang paling cepat, tapi <strong>tidak ada yang bisa menjamin bahwa data yang dikirim benar-benar sampai</strong> atau tidak, bisa jadi gagal kirim dari <em>publisher </em>ke <em>broker</em>, atau bahkan dari <em>broker </em>ke <em>subscriber</em>. Mekanisme ini <strong>cocok dipakai untuk data yang sifatnya <em>real-time</em></strong> dan <strong>toleran terhadap data <em>loss</em> atau <em>non-critical </em>data</strong>, ketika satu atau dua data hilang tidak memengaruhi performa sistem secara keseluruhan, misalnya data <em>temperature </em>dan <em>humidity</em>, yang data baru selanjutnya dapat segera menggantikan data yang hilang.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/742/1*LVnMr1f5cqH8fn0AGI3a-Q.jpeg" /><figcaption>QoS 1: The At Least Once Mechanism</figcaption></figure><p>Selanjutnya, pada <strong>QoS level 1</strong>, mekanisme PUBACK (<em>Publish Acknowledgement</em>) <strong>menjamin pengiriman pesan <em>at least once</em></strong>, dalam kata lain pesan diterima setidaknya sekali oleh <em>receiver</em>. Ketika pesan berhasil diterima, <em>receiver </em>akan mengirimkan <em>acknowledgement </em>berupa pesan PUBACK. Terlihat bagus untuk mencegah <em>data loss</em> bukan? Benar, tapi lagi-lagi tidak ada sistem yang sempurna.</p><p>Mekanisme ini memang menjamin tidak adanya <em>data loss</em>, tapi akibat mekanisme yang sama, <strong>terdapat kemungkinan duplikasi pesan</strong>. <em>Lah, kok bisa?</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6DwYwW5yRJZrOtuWc7VofA.jpeg" /><figcaption>Message Duplication in QoS 1</figcaption></figure><p>Pada <em>case</em> 1 di atas, dapat dilihat proses PUBLISH pertama oleh <em>sender</em> gagal. Ketika <em>sender </em>tidak menerima PUBACK dalam <em>threshold </em>waktu tertentu, <em>sender </em>akan mengirimkan ulang pesan tersebut melalui mekanisme PUBLISH. Ternyata, setelah pesan PUBLISH ke dua tersebut dikirimkan, <em>receiver </em>menerima pesannya, dan membalas dengan PUBACK. Dari sini, dapat dilihat kalau PUBACK benar-benar membantu dalam menangani data <em>loss</em>.</p><p>Tapi, kalau kita lihat ke <em>case </em>2, ada potensi data <em>duplication</em>. Lebih tepatnya, ketika pesan PUBLISH pertama berhasil sampai, dan <strong>receiver mengirimkan pesan PUBACK, tetapi <em>response</em> tersebut tidak pernah sampai</strong> ke <em>sender</em>. Konsekuensinya, ketika sender tidak menerima PUBACK, ia akan menganggap terdapat data <em>loss</em> dan <strong>mengirimkan kembali pesan</strong> yang sama, <strong>padahal pesan tersebut sebelumnya telah diterima</strong> oleh <em>receiver</em>. Di sisi lain, receiver tidak dapat membedakan, apakah pesan tersebut memang pesan baru, atau memang pesan duplikat. <em>At this point, </em><strong><em>data duplication under QoS 1 is truly inevitable</em></strong>.</p><p>Mekanisme ini menjamin pengiriman data setidaknya sekali, prosesnya tidak sesimpel QoS 0, tetapi juga tidak sekompleks QoS 2. Namun, kelemahannya ada pada kemungkinan duplikasi data yang dikirimkan. Mekanisme ini <strong>cocok untuk digunakan di sistem yang toleran terhadap duplikasi data</strong>. Kelemahan tersebut juga bisa diatasi pada layer lebih tinggi (misal layer aplikasi) dengan menggunakan <em>unique identifier</em> untuk deteksi duplikasi data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/742/1*heMQIoLdP_-AmUH6KlZuAA.jpeg" /><figcaption>QoS 2: The Exactly Once Mechanism</figcaption></figure><p>Untuk mengatasi kelemahan dari QoS 1 berupa data <em>duplication</em>, <strong>QoS 2 menawarkan mekanisme <em>exactly once</em></strong> yang menjamin pengiriman data pasti sekali melalui PUBREC (<em>Publish Received</em>), PUBREL (<em>Publish Release</em>), dan PUBCOMP (<em>Publish Complete</em>).</p><p>Masalah yang muncul pada QoS 1 adalah <em>receiver </em>tidak tahu dengan pasti apakah pesan yang diterima itu merupakan pesan duplikat atau pesan baru. Pada QoS 2, setelah sender PUBLISH, ia akan menunggu konfirmasi penerimaan pesan dari <em>receiver </em>melalui PUBREC. Pada tahap ini, <strong>Apabila PUBLISH terkirim lebih dari sekali karena retransmisi, <em>receiver</em> akan menandai pesan tersebut sebagai duplikat</strong>, dan akan mengambil salah satu saja.</p><p>Kemudian, sender akan mengirimkan PUBREL, bahwa pesan dengan Packet ID tertentu siap untuk diproses/<em>deliver</em>, yang secara implisit juga menyatakan bahwa packet ID tersebut mungkin digunakan kembali nanti, sehingga pesan dengan packet ID yang sama bukanlah duplikat, tetapi pesan baru. Setelahnya, <em>receiver </em>akan membalas dengan PUBCOMP, tanda pesan telah selesai diproses.</p><p>Singkatnya, setiap <strong>PUBLISH tambahan dengan Packet ID yang sama sebelum PUBCOMP dianggap duplikat</strong>. Melalui mekanisme seperti ini, <em>receiver </em>jadi tahu, mana pesan yang duplikat dan yang bukan. Mekanisme ini <strong>memiliki kompleksitas dan <em>overhead </em>yang paling tinggi</strong> jika dibandingkan dengan dua level di bawahnya, tetapi juga menjamin tidak adanya data <em>loss </em>atau <em>duplication</em>, yang <strong>cocok dipakai di dalam sistem yang sangat sensitif</strong> terhadap perubahan data.</p><h3>Penutup</h3><p>Pada akhirnya, MQTT bukan sekadar protokol ringan untuk IoT <em>devices</em>, tetapi fondasi komunikasi yang membuat ekosistem IoT benar-benar bisa berjalan. Dengan mekanisme <em>publish-subscribe</em> yang sederhana, fleksibilitas <em>topic </em>dan <em>wildcard</em>, serta <em>level</em> QoS yang bisa disesuaikan sesuai kebutuhan, MQTT memberi ruang bagi sensor, perangkat, dan aplikasi untuk saling terhubung dengan cara yang efisien.</p><p>Tantangannya memang ada, mulai dari potensi duplikasi data hingga kompleksitas (QoS 2). Namun, justru di situlah letak kekuatan MQTT, fleksibel. Ia menyediakan pilihan sesuai kebutuhan sistem. Jadi, entah kita ingin membangun dashboard sederhana, jaringan sensor berskala besar, atau sistem yang ultra sensitif terhadap perubahan data, MQTT selalu punya cara untuk menjembatani komunikasi.</p><h3>Reference</h3><p>EMQX. (2025, August 6). <em>Mastering MQTT: The Ultimate Beginner’s Guide for 2025</em>. <a href="https://www.emqx.com/en/blog/the-easiest-guide-to-getting-started-with-mqtt">https://www.emqx.com/en/blog/the-easiest-guide-to-getting-started-with-mqtt</a></p><p>HiveMQ Team. (2024, February 20). MQTT Topics, Wildcards, &amp; Best Practices — MQTT Essentials: Part 5. <em>HiveMQ</em>. <a href="https://www.hivemq.com/blog/mqtt-essentials-part-5-mqtt-topics-best-practices/">https://www.hivemq.com/blog/mqtt-essentials-part-5-mqtt-topics-best-practices/</a></p><p>HiveMQ Team. (2024, May 20). What is MQTT Quality of Service (QoS) 0,1, &amp; 2? — MQTT Essentials: Part 6. HiveMQ. <a href="https://www.hivemq.com/blog/mqtt-essentials-part-6-mqtt-quality-of-service-levels/">https://www.hivemq.com/blog/mqtt-essentials-part-6-mqtt-quality-of-service-levels/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a41ed96730a9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[APIE Camp 2025: A Journey in Manado]]></title>
            <link>https://radityanala.medium.com/apie-camp-2025-a-journey-in-manado-c33f27bc3ff3?source=rss-26bf16164f14------2</link>
            <guid isPermaLink="false">https://medium.com/p/c33f27bc3ff3</guid>
            <dc:creator><![CDATA[Yoga Raditya Nala]]></dc:creator>
            <pubDate>Sat, 20 Dec 2025 13:47:40 GMT</pubDate>
            <atom:updated>2025-12-20T13:56:59.137Z</atom:updated>
            <content:encoded><![CDATA[<p>It was the middle of the night, and I was trying to wrap up a task for a freelance project I’d been working on for months. Out of nowhere, an email popped up from one of the APIE staff members, titled “<em>Congratulations</em>” I was stunned. Our lecturers had already told me I wouldn’t be joining the camp this time. Honestly, I had made peace with it—maybe even felt relieved, considering how busy I already was. I tried to brush it off as a half-asleep hallucination, but I couldn’t stop staring at it. Eventually, I opened it, and it turned out to be real. I was accepted to the 8th APIE Camp, to be held at Universitas Sam Ratulangi (UNSRAT) the following month.</p><p>In that moment, fear washed over me,<br><em>fear that I wouldn’t know what to do,<br>fear that I wasn’t good enough,<br>fear that I’d only disappoint my peers,<br>and many other unspoken fears.</em></p><p>The same fears that stopped me from registering for the previous camp at Universiti Sains Malaysia (USM). But this time, I had already made up my mind. <strong>At least once in my life, I should try</strong>. There was one quote that pushed me to finally register, the same one I had pinned on my Twitter profile back then.</p><blockquote>“I’d rather regret doing something than regret not doing it later.”</blockquote><p>Still, the weeks before the camp were hectic—exams, final projects, everything piling up at once. The camp preparation itself wasn’t too demanding. We were only assigned to explore the AWS Academy Lab so the camp would run smoothly. I managed to finish it just in time. We also had a final briefing covering what we’d be doing, how much money to prepare, and other practical details. I had hoped to revisit DNS, since it was the only part I struggled with during the APNIC lab, much harder than routing, honestly. But academic life kept me so busy that I barely noticed how quickly the departure day arrived.</p><h3>The Departure Day — 16/11</h3><p>It was a cold Sunday morning, the kind that made staying in bed dangerously tempting. I had pushed myself to the limit the night before trying to finish a project due that evening. Still, lingering in bed never really brings me much joy, so I got up, packed, got ready, had breakfast, and double-checked everything. Our flight to Manado was scheduled for 2:50 p.m. from Juanda International Airport near Surabaya. We had to leave Malang around 10 a.m., arranged by a travel service. Since my house was the farthest, I was picked up first, followed by Syifani, Liyya, and Melody. There were actually five of us, but Akbar chose to travel separately since he lived much closer to the airport.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*Xn-qFdMTXYUFOIHtmkPEfQ.jpeg" /><figcaption>The Place Where Everything Began</figcaption></figure><p>We arrived at the airport around 1 p.m. and prayed while waiting for Akbar—who, amusingly, was already inside eating at a restaurant. After that, we headed in, chatted a bit, and queued to check in our luggage. That’s when the first funny moment of the trip happened. The airline had a 10 kg limit for checked baggage. Mine and Akbar’s were fine, but the girls’ bags were over the limit. No wonder—Syifani even brought an iron, <em>haha</em>. Eventually, we headed to the terminal, passed through security, and waited, only to learn that our flight had been delayed to 4 p.m.</p><p>Thankfully, there were no further delays. The aircraft arrived, boarding was announced, and I felt relieved knowing I could finally relax. It was my first time flying, so I hoped everything would go smoothly. I was originally seated next to Akbar in the middle row, but it turned out I’d been reassigned to an emergency exit window seat. I wasn’t thrilled, but at least it had more legroom. Still, I couldn’t rest. The seat felt too high, my neck hurt, I had a migraine, and I was hungry. It was a direct two-and-a-half-hour flight from Surabaya to Manado, and the combination of boredom and discomfort wasn’t pleasant.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3G2FveZIbAILsblRi4czTg.jpeg" /><figcaption>A Very Long and Tiring Day</figcaption></figure><p>Eventually, a voice came through the intercom announcing that we’d be landing soon. I was beyond relieved. We touched down around 9:30 p.m., and exhaustion hit everyone immediately. We were greeted by Ryo and Joshua—both serving as teaching assistants for the camp. I already knew Ryo since he was one of my seniors from the faculty’s lab, while it was my first time meeting Joshua from UNSRAT. I didn’t talk much—I was too tired and, <strong>more importantly, starving</strong>. The hotel was about 16 km from the airport. When we arrived, we checked in and regrouped at the lobby. My roommate, Rayyan, wasn’t in the room yet, but his belongings were there. I assumed he was out looking for food like the rest of us. The room had a beautiful sea view, which I instantly loved.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ED9FeFEMVxU-7cF9NU3HUw.jpeg" /><figcaption>A Beautiful Night Under Manado Skies</figcaption></figure><p>Our plan was to find food at the mall beside the hotel since we had direct access to it. But Melody was unreachable—we guessed she had fallen asleep, so we went ahead without her. To our disappointment, the mall was closing. Once again, I had to postpone the one thing I wanted most—food. We decided to walk outside instead. Along the way, we ran into others. Viet, a NoC member from Duy Tan University (DTU) Vietnam; Sabina, another TA from UNSRAT; Minh from DTU; and Shota from Keio University. Not everyone joined since many had already eaten, and it was nearly midnight. Only Viet and, later, Egar—another NoC member and one of my seniors, came along. Not long after, we finally bumped into Melody, who had actually been on a call with her mother (sorry Mel, <em>haha</em>).</p><p>We eventually found a nearby coffee shop. I actually enjoyed the walk. <strong>The night felt peaceful</strong>—busy streets actually, but far less chaotic than in Malang. Perfect, really. I didn’t feel like ordering anything specific, just something to fill the hunger. While waiting for the food, we talked about random things. I was a bit shy and not very confident with my English, so I mostly listened. One thing stood out, though—it was loud outside. Cars kept passing by with booming speakers, pretty similar to Malang, except in Malang, they uses big trucks with even bigger speakers, <em>haha</em>.</p><p>After eating, we headed back to the hotel. That’s when I finally met Rayyan in person. He’s from ITB. It seemed like he had been waiting to make sure I could get into the room, assuming I didn’t have the keycard. Once I arrived and opened my laptop, he went straight to sleep—probably just as tired as I am. We had to wake up early anyway, the bus to UNSRAT would leave around 7:45 a.m. I wanted to sleep too, but my project refused to let me. I ended up reading research papers, trying to form a solid argument for my final report. But eventually, fatigue won, and I finally gave in to sleep.</p><h3>Network Deployment — 17/11</h3><p>I woke up early on the first day, and I think that was the only morning I actually caught a sunrise while I was there. As usual, I checked WhatsApp, opened my laptop, and worked on a final project from another course. Honestly, this was probably the busiest week of the semester. I had to stay focused on the APIE Camp while juggling multiple final projects at the same time. After finishing some of my work, I got ready and took a bath. The funny part was that the bathroom didn’t even have a proper lock, and the glass was kind of transparent—so I was a bit paranoid. Thankfully, my roommate was still asleep, <em>haha</em>.</p><p>Once I was done, I headed to the lobby for breakfast with Akbar and the others. This might have been my favorite moment of the camp. We could take anything we wanted, and everything was free. I grabbed fried rice with chicken and beef, plus some fruit and drinks. I also met up with Rayyan, Syifani, Melody, Liyya, the TAs, and the NoC team. After we all finished eating, we waited in the lobby with a bunch of faces I didn’t recognize yet. Two participants from Universitas Syiah Kuala (USK) had missed their flight and were scheduled to arrive that morning. I felt bad for them—they didn’t get any rest at all. The van eventually arrived—the same kind that had picked us up from the airport the night before. We were split into three groups, and I ended up in the third van with the UB team, except Melody. For some reason, <em>she always got separated from us, haha.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1pnsPW3GzdAa0wSqTIho5Q.jpeg" /><figcaption>First Shot, The Beginning of the Journey</figcaption></figure><p>Our first stop was the faculty hall, where the dean gave a short welcoming speech, followed by introductions from the APIE staff, i.e., Kudo-san, Marcos-san, Arima-san, and Mr. Imtiaz from APNIC. We also took our first group photo with all participants, TAs, and staff. After that, we headed to the Teknologi Informasi dan Komunikasi (TIK) building. A few more participants joined us—people I hadn’t seen at the hotel, which made sense because they were from UNSRAT.</p><p>At the venue, we were greeted with snacks and drinks, and we received APIE merchandise—a small thermos, which I loved. I’d been wanting a small tumbler for a while but never got around to buying one, so the timing felt perfect. I also finally met the three UNSRAT participants who weren’t staying at the hotel, i.e., Regina, Crystal, and Fadhlan. Inside the lab, we were divided into four groups. I ended up with Crystal, Tian, and Zikri—Group 4. At that point, I barely knew any of them. I had only met Crystal that morning. Tian had arrived the night before, but we never talked. Zikri had just landed that morning after missing his previous flight, along with Muhazar.</p><p>Before jumping into networking tasks, we introduced ourselves one by one. That’s how I learned more about people like Jai from USM, who surprised all of us. He’s currently pursuing a PhD, not a bachelor’s degree like the rest of us.</p><blockquote>“Well, I guess he’s proof that it’s never too late to learn something new,” I murmured to myself.</blockquote><p>I also met Alief, a TA from ITB who had already graduated. We even had a small game where each group submitted a unique item and others had to guess who it belonged to. It wasn’t really about the item—it was to help us remember each other’s names. There was also a scoring system in the camp. We could earn points by answering questions, completing tasks and missions, showing up on time, or simply winning the games, just like this one.</p><p>The first lecture, an introduction to networking, was delivered by Mr. Alwin from UNSRAT. It was basic material, but important as a refresher. After that, we moved on to the hands-on part, making UTP cables. It sounds simple, but not everyone found it easy. Zikri and I managed to get ours right on the first try. I’d practiced this many times in our own lab. Tian and Crystal struggled a bit, especially with positioning all eight wires into the RJ45 connector. Honestly, that part is always tricky. In total, we needed to create six cables, four for group devices, and two for the uplink and server.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*byyXuw4RcidNEQZr7Qqc-w.jpeg" /><figcaption>Melody truly has such a strong spirit, haha</figcaption></figure><p>Some people still hadn’t finished even during the break, including Melody. She kind of reminded me of how I used to code back then—<em>chasing deadlines and pushing my body past its limits, haha</em>.</p><p>The Muslim participants went to pray at the mosque while the others waited outside, since we would go straight to lunch afterward. Again, we were free to choose anything we wanted. I picked rice with soup and grabbed a bottle of water. We chatted about random things while eating—just getting more comfortable with each other. Afterward, we returned to the venue. Even though it was just behind the building, some of us, myself included, were lazy and took the van anyway.</p><p>Back in the room, Mr. Daniel continued with material on IP addressing, subnetting, routing, NAT, and several other technical topics. At first, I felt a bit lost — I had forgotten some basics, like calculating network and broadcast addresses. I even struggled to convert decimal IPs to binary, only to find out later that calculators were actually allowed. Akbar answered many of the questions and collected a lot of points. Meanwhile, I felt a little disappointed in myself for not being able to answer even one, haha.</p><p>After that, we were given a blank topology and asked to assign IPs ourselves based on what we had just learned. The topology was simple. The first router connected to the uplink, servers, and staff PCs. The second router connected to a switch, which connected to the access points. Both routers were interconnected. The first router acted as the Integrated Services Router (ISR), and the second was the ISR-NAT, because it handled NAT for the network behind it. Each topology received a public /28 subnet, just enough to connect the routers and servers. NAT on the ISR-NAT supported all end users connected through the access points, including us. I was already used to this kind of setup. A few months earlier, our lab held an event similar to APIE Camp, with almost the same equipment and topology. So for me, it felt familiar and refreshing.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*h6ptUZxGVGJ47QihWRryYQ.jpeg" /><figcaption>Our Lovely Network Devices</figcaption></figure><p>The only issue was that our network couldn’t reach the internet. Strangely, we could ping the uplink just fine. We spent hours troubleshooting before finally discovering that the documentation used a different uplink subnet than the one we received. More challenges appeared when configuring IPv6. The subnet wasn’t /64, so Stateless Address Autoconfiguration (SLAAC) will never work. The NoC team later changed it to /64 to make things easier. Overall, the first day was the toughest—there were many devices to configure, each with different requirements, all under tight time pressure. But in the end, we managed to get everything working. We were finally able to reach the internet, like 8.8.8.8 and 1.1.1.1, from the access point network. Two groups weren’t as lucky, they had to stay late and come back early the next morning to fix their setup with help from the TAs and the NoC team.</p><p>We got back to the hotel around 7 p.m. and got ready for dinner. Zikri and Muhazar didn’t join us, probably because they were exhausted and hadn’t had any real chance to rest beforehand. Regina, Crystal, and Fadhlan didn’t come either, mainly because they weren’t staying at the hotel, and it would’ve taken extra effort for them to go there first before heading to the restaurant. The restaurant, Hotways, was quite far from the hotel, but even though I was already tired, I actually enjoyed the walk. As I mentioned earlier, the nights there were peaceful, and I loved strolling together with everyone. From the walkway, we could clearly see the night sea, something I rarely get to experience in Malang. Sabina and Joshua usually led the way since they knew the area well, while the rest of us followed behind. Ryo chatted with Shota, Syifani with Liyya, Akbar with Egar, the ITB participants with Al, and I usually talked with Melody—or simply enjoyed the night.</p><p>Eventually, we arrived at the restaurant. Some of us ordered right away, while the others waited patiently for the food to be prepared. Once everything was ready, we moved to the second floor, which had enough space for all of us. I didn’t talk much—partly because I was starving, but also because my mind kept drifting to the final projects I had missed. After we finished eating, we split up. Sabina wanted to buy something nearby, while the rest of us headed straight back to the hotel, including me, because I was completely exhausted. Back at the hotel, I planned to work on some of my final projects before going to sleep. But once again, my body gave up, and I ended up crashing early. I was just too tired.</p><h3>Virtualization &amp; Cybersecurity — 18/11</h3><p>On the second day, I woke up a bit later than I had the day before. Rayyan showered first since he needed to be there around 7 a.m., while the rest of us didn’t have to arrive until about 9. The morning went pretty much the same as usual—we had breakfast, then waited for everyone in the lobby, but it felt a bit calmer.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*R99f5Mwjp67DSwSdqkvohg.jpeg" /><figcaption>Our Beloved Network Topology</figcaption></figure><p>I spent some time working on a digital version of the topology we built the day before, just to make things easier since my handwriting is pretty hard to read. I was sitting with Zikri then, talking about the updated topology I had just recreated. While we were waiting for the transportation to arrive, Sabina went around scanning our ID cards so we could earn points for being in the lobby on time. Not long after that, the transportation finally arrived. Sabina and Joshua quickly told us to head down to the ground floor. Unlike the previous day, this time it was a large bus, and all of us could fit into a single vehicle.</p><p>There was one phrase I kept seeing throughout the week, and it became my favorite.</p><blockquote>Sitou Timou Tumou Tou</blockquote><p>It was written at the entrance of UNSRAT, visible almost every day during the camp. It carries the meaning that <strong>humans should humanize others</strong>—simple, yet something many still struggle to live by.</p><p>The material that day began with a brief introduction to virtualization and why it matters. It was delivered by the NoC team, i.e., Egar and Viet. There was also a short overview of Linux, though I think most of us already use it on a daily basis anyway. I was a bit surprised to learn that Zikri actually uses Hyprland as his daily driver, and I couldn’t help teasing him about it.</p><blockquote>“A daily driver with a driver that breaks daily,” I told him the moment I saw his desktop.</blockquote><p>No offense to him, of course. It’s just something I’m pretty sure most of us have experienced at some point, haha. Even with my KDE Plasma setup, I often end up frustrated by how easily it can break. Thanks to the early-oom package, which prevents system-wide crashes by meticulously monitoring tasks that exceed memory limits.</p><p>After that, we were instructed to open the server dashboard—Proxmox, to create the VMs that would be used as our DNS servers. This was the moment our previous configurations were put to the test. The same reason why the groups that failed to configure their network had to stay late and come in early. Each VM had to be assigned the IP address we reserved the day before. If we had missed anything during the previous setup, that was it—we wouldn’t be able to continue with the rest of the day’s exercises.</p><p>Just when everything seemed to be going according to plan and the VM creation process returned successful responses, a new problem appeared—our VMs couldn’t access the internet. At that moment, I prayed wholeheartedly that the issue wasn’t on our side. Fortunately, a short while later, it was announced that there was a problem with the server in Japan. We were told to take a break while the NoC team worked on fixing it.</p><p>After we finished lunch, we returned to the venue to listen to Mr. Husni’s presentation on cybersecurity, specifically, the security measures we needed to implement on the virtual machines we had just created. The IP addresses we assigned earlier were actually real public IPs, reachable from the internet, so leaving them unsecured was definitely not a good idea.</p><p>We started by setting up the basic protections—disabling password authentication and switching to key-based authentication only. We also configured the firewall to allow only essential traffic, e.g., HTTP on port 80 and SSH on port 22. There’s actually a funny story behind that firewall setup. Once, I accidentally locked myself out of my own machine because I enabled UFW without allowing TCP/22 first. As a result, the server became completely unreachable remotely—and honestly, it turned into a surprisingly memorable learning experience, haha.</p><p>To be honest, the second day felt less hectic than the first. Maybe it was simply because I was already familiar with Linux, Proxmox, and virtualization. Back at the faculty, I used to handle part of the organization’s digital infrastructure, i.e., creating and provisioning VMs, managing domains through Nginx Proxy Manager, and so on. So this task felt relatively easy and even enjoyable for me.</p><p>After everything was done, we finally headed back to the hotel. That night felt different. The rain was pouring heavily outside, and none of us were in the mood to walk around. I decided to work on my final project for a bit, then asked Akbar if he wanted to go out to find some food. He agreed, and Melody joined us as well.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NaarRfw4d3CJpgJ2XcFLhw.jpeg" /><figcaption>A Beautiful Night Beneath the Weight of Final Projects</figcaption></figure><p>The plan was to stop by Hypermart first since they wanted to buy a few things, then head to Mantos to look for food. We ended up choosing Solaria, even though it was a bit pricey.</p><blockquote>“Well, it better be worth it,” I told myself.</blockquote><p>On the way there, I spotted Kesha and some of the other APIE participants eating at Hokben, right next to Solaria. We didn’t eat in, though—we decided to take everything to go. Since I was racing against a tight deadline for my final project, I inevitably agreed. In the end, I ate my dinner while reading research papers back at the hotel. I couldn’t help but feel a little sorry for myself, haha.</p><h3>Domain Name System (DNS) Servers — 19/11</h3><p>Then came the next day—the one I feared the most. Mostly because the material wasn’t something I was fully comfortable with. My last real experience with it was during the APNIC Lab in the APIE online course, and even then, it took me hours just to debug a single DNS server configuration. We were guided by Mr. Imtiaz, who introduced us to everything related to DNS.</p><p>It was actually quite interesting to learn about. If I had to say, DNS might be the system that makes the internet truly usable. I mean, we just type a domain name and get taken to the site we want. Imagine if DNS never existed and having to memorize a 32-bit IP just to visit a website. Not to mention 128-bit IPv6 addresses, haha.</p><p>The tasks, however, were quite a lot. We needed to configure a recursive resolver, a primary DNS server, and a secondary DNS server. The last VM was for the web server, which would later run Apache or Nginx. Our group divided the work, one VM per person—and I got the webserver, which required the least effort. I mostly ended up monitoring their progress using tmux, a simple terminal multiplexer that lets multiple remote sessions share the same TTY. That way, I could see what everyone was doing and try to follow each task.</p><p>A recursive resolver is a server that acts on behalf of the client—whether it’s a browser or another application to look up the IP address of a given domain. It’s the component that performs the full lookup process, starting from the root servers and moving down through the DNS hierarchy until it finds the correct address. The primary DNS server, on the other hand, stores the actual records that map names to IP addresses within its domain. You can think of it as a simple database that the resolver consults when searching for the right information. The secondary DNS server serves a similar purpose, but instead of holding the original records, it replicates them from the primary server. It functions as a failover or backup in case the primary server goes down.</p><p>Redundancy was one of the key themes emphasized during this camp, and the DNS session demonstrated it clearly. There was a real-world case where an attack on one of the world’s major DNS providers succeeded, taking down a significant portion of the internet. That incident alone is more than enough proof that redundancy isn’t optional—it’s essential in the networking world.</p><p>And as I mentioned earlier, the IP addresses we received were all public, including the ones used for DNS. We were assigned subdomains under the <em>.soi.asia</em> domain, and Mr. Imtiaz, who represented APNIC, was responsible for delegating each subdomain to our DNS servers. Our group successfully set up the main infrastructure—the recursive resolver, primary server, and secondary server.</p><p>However, after Mr. Imtiaz delegated our domain to our nameserver, nothing worked. The <em>dig</em> result for our domain still didn’t point to our DNS at all. We tried to figure out what went wrong, but nothing made sense, and none of our tweaks fixed the problem. Eventually, I started thinking in a different direction.</p><blockquote>“What if the issue isn’t on our side but on Mr. Imtiaz’s side?” I told my group.</blockquote><p>I had a solid reason for it. When we ran <em>dig</em> on our domain, it still pointed somewhere else—some IP that looked like it belonged to a previous camp. It clearly wasn’t a DNS propagation issue since hours had passed, and other groups were already done and could access their domains. So, together with Ryo, I approached Mr. Imtiaz and explained the problem. And sure enough, the domain was still pointing to an old IP, one that wasn’t even in our topology. Not long after he fixed the delegation, <em>voila</em>—everything worked flawlessly.</p><p>That night felt different. Crystal, Fadhlan, and Regina came over to the hotel, and most of them were planning to go out and eat together. I really wanted to join them. However, earlier that morning, I had received some bad news. My final project was already three days past the deadline, and the consequences were serious. For every day late, ten points would be deducted, and worse, the project carried the same weight as the final exam. If I didn’t finish it that night, the damage would be hard to recover from. So I spent the night buried in research papers, pushing myself to complete the report. Even though I managed to finish it before the day ended, something still didn’t feel right. I felt upset with myself, for not managing my time better, for being late, and for not pushing hard enough earlier when I still had the chance.</p><h3>IoT &amp; Cloud — 20/11</h3><p>That day somehow began more calmly, and everything seemed to move a little slower—at least for me. It was the last day we were scheduled to be at UNSRAT, and I’ve always had this habit of looking at things differently when I know it’s the last time. I found myself staring at the entrance, the room, trying to take in every remaining moment. It was also the final set of material we would receive in this program—IoT, something I genuinely enjoy, along with a bit of cloud-related content.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/901/1*hs31fZaH1T0QEvz3rJcG5A.jpeg" /><figcaption>Unemployed IoT Engineer at its Finest</figcaption></figure><p>The first course was about IoT. It was delivered by Mr. Pinrolinvvic, a lecturer from UNSRAT as well. The material was actually quite simple, interconnecting sensors like the DHT11 and an ultrasonic sensor. Everything went smoothly, and the rest of my group explored the topic carefully. This time, we had to deal with physical cables and hardware, where even a small mistake could ruin the entire hands-on session.</p><p>Watching them work somehow reminded me of my very first mistakes with IoT and embedded systems. Once, I connected an ESP8266 microcontroller to a USB power supply that should have worked — except the adapter was supplying almost 30 volts, far beyond what the ESP could handle (around 5 volts). It was a funny, but painfully memorable, moment. So I mostly stood there, watching them have fun while making sure everything stayed safe, especially that no one swapped the GND and 3.3V pins, haha.</p><p>Our group finished first, and I even took a moment to step outside and enjoy some of the food provided, savoring the last bit of joy I could get from this place, I guess, haha. After that, the material started to shift toward cloud topics, specifically Amazon Web Services (AWS). It was delivered by Mr. Aris, a professional working in cloud from Singapore (if I’m not mistaken).</p><p>The plan was to connect the IoT node to an MQTT server provided by AWS IoT. For me, the cloud setup itself was a bit challenging. There were a lot of steps to follow, and worse, my account didn’t behave like the others’, as if some features had been cut off. No wonder — just a few days earlier, I had created a high-resource VM for my wireless network final project to host a 5G network core Kubernetes cluster, which indeed required a lot of resources to run.</p><p>Luckily, Tian got my back, he created all the necessary services to make the MQTT service run smoothly, and it did. When we connected the IoT node to the access point and started publishing data to the AWS domain, the data showed up perfectly on the dashboard, indicating that everything was running just fine. Once again, we became the first group able to do that (<em>kind of proud of it</em>). Even Mr. Pinrolinvvic and Mr. Aris praised our group for it, haha.</p><p>One last obstacle for us was trying to use IPv6 to connect to the AWS infrastructure instead of the usual IPv4. We tried everything. Our IoT node received an IPv6 address from the access point with some tweaks, but we had no way to be sure that this address was actually being used to connect to AWS. I tried another approach, querying the AWS endpoint using dig. And yes, I was able to get the IPv6 address of that endpoint. However, when I tried to use it directly, it led to no results. It seemed that somewhere along the way, direct IP access simply wasn’t allowed on that infrastructure. In the end, we (along with the staff) finally gave up on that part and just pretended IPv6 was used, haha.</p><p>Earlier that evening, Mr. Aris also introduced us to the concept of load balancing, where multiple servers can serve the same resource and split the traffic so that no single server gets overwhelmed. We implemented this by creating multiple web servers connected to a single load balancer gateway. We also modified each web server’s page so we could tell which server was actually serving the request.</p><p>I ended that day in peace, knowing very well that everything I had been afraid of was finally over. All the main materials and tasks had been completed successfully. That night felt different. We all knew the next day would be the final presentation day. We had to present our ideas and implement everything we had learned so far. The plan was simple: find food as quickly as possible, then work on the presentation. As people often say,</p><blockquote>“Humans may plan, but God decides the final verdict.”</blockquote><p>All those plans quickly turned into fairy tales. What started with just Akbar, Melody, and me slowly grew — we ran into Regina and Crystal, then Rayyan and Zikri showed up, followed by Jai and the rest of the others.</p><p>Honestly, we didn’t really know where to go. But Jai confidently led the way. He didn’t know the name of the restaurant, but he was sure he knew the route, and we trusted him. Funny enough, halfway there, he started getting confused himself, and we practically got lost. Eventually, we ended up at a nearby restaurant and ordered some food. Regina and her team discussed their presentation there. Unfortunately, other group didn’t bring any devices or even something as simple as a paper and a pen, so we couldn’t really discuss anything. Ryo also joins us later, and we just enjoyed the rest of the night, I guess.</p><p>Long story short, after we finished eating and having fun, we headed back to the hotel. Since my group didn’t plan to work everything out that night, Crystal went back to her house, while Regina came with us. Her group was working somewhere on the upper floor, while Rayyan’s group settled in our room. I hadn’t expected so many people to show up, but the crowd grew quickly — first Melody, Minh, and Muhazar, then Zikri, Egar, and even Ryo. Still, it was all good fun. We exchanged ideas, talked about how the presentations might go, and even asked for a few spoilers from the previous camp.</p><p>With almost every group busy preparing their slides, I felt motivated to do the same. Together with Zikri, we tried to fill out as many slides as possible. Tian and Crystal weren’t very responsive in the group chat — no surprise, since it was already close to midnight and we had to wake up early the next morning. By the time we wrapped everything up, it was around 2 a.m. With only four hours left to sleep, none of us wasted a second. After Melody, Minh, and Muhazar said goodnight and left, I finally let my eyes rest, relieved to end the night on a productive note.</p><h3>Industrial Visit and Trip to Tomohon — 21/11</h3><p>And here we were, at the end of the journey. That day felt calmer than ever. Even though we didn’t have the pleasure of a full night’s sleep, it was still the day we had all been waiting for. The plan was quite simple, an industrial visit to PT Telin, an Indonesian national company that hosts the submarine cable landing site in Manado, followed by a trip to Tomohon for the final presentation — plus some food and fun, of course.</p><p>The journey to PT Telin itself didn’t take very long, but it was peaceful. I managed to get enough sleep while we were on the way. We didn’t take the big roads, just smaller ones winding through the mountains and forests. That was one of the things I really like about Manado — the shore and the mountains aren’t that far apart. Even Mr. Husni mentioned the same thing to me, haha. It’s quite different in Malang, where you need at least three hours from the city just to reach the shore.</p><p>Upon our arrival, we were greeted by quite a few armed forces. No wonder — it was one of the nation’s vital facilities that needed to be guarded, haha. After that, we were warmly welcomed by some of the PT Telin staff. They even gave us some merchandise — a pocket-sized waterproof bag, which I really liked. Then they explained what they actually do there, the impact they make, and the history behind the place. The talk went on for quite a while, and I started to feel a bit impatient. Well, I guess I wasn’t alone — Kudo-san himself said the same thing (in a friendly way, of course).</p><blockquote>“Enough with the talk, show me the facilities”<br>Kudo-san said after giving his speech.</blockquote><p>It kind of gave me the same vibe as that famous quote from Linus Torvalds, haha.</p><p>Finally, they divided us into two groups. I was placed in the second group, so we had to wait for the first group to finish their exploration. No phones or cameras were allowed; everything had to be kept inside a special box provided by the staff. Not long after, we were allowed to go inside. I was expecting to see the physical form of the submarine cable, but it turned out I was wrong. Instead, we were shown some of the equipment they use — like the systems to control power, the backup power supply, and devices to read data from the cable. The visit didn’t last very long. I thought it would take hours to explore the entire place, but in reality it was only about 15 to 20 minutes at most.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*O3_T3vdg7RzKCV05mgy7Hg.jpeg" /><figcaption>The Happiest Day of the Week</figcaption></figure><p>Still, it was fun. The staff were friendly and even gave us snacks, food, ice cream, and a cup of coffee. You could tell how happy we all were back then — it was basically the most peaceful day of the entire week, haha. We then headed back to the bus to continue our journey to Tomohon.</p><p>Just like before, the journey to Tomohon didn’t take long. Sabina played some music on the bus, which coincidentally matched my taste — even The Black Parade came on, haha. I sang along throughout the ride, without realizing we were already close to our destination. However, because of Friday prayer, some of us needed to get off first, and the bus would pick us up later. From the mosque, it wasn’t a long journey — less than half an hour, so it wasn’t a big problem.</p><p>Upon our arrival, I was amazed by how beautiful it was. Trees were everywhere, the mountains felt close, and the cold air carried a peaceful vibe. Still, I grew nervous. It was 1 p.m., and the final presentation was scheduled just an hour later. My mind was racing with potential questions that might be asked, flaws that could be found in our ideas, and countless other worries. I couldn’t even eat at that time — I didn’t feel hungry at all. I think I was already fed enough with all the ideas, haha.</p><p>Long story short, the time has come. The first group came first. It was Melody, Rayyan, Muhazar, and Minh. They present some sort of smart system for schools. Cool, but I don’t see problem or the urgency of implementing those smart system. I mean, yeah it could be helpful, but not really urgent at this point, but still good ideas. A lot of question is asked, I grew more nerveous back then, haha.</p><p>The second group presented their smart parking idea. It was Syifani, Jai, Regina, and Kesha. Their plan was to use a single controller for each parking spot — yes, you heard that right, one controller for every spot. They based their case on UB, my university, where parking spaces are usually difficult to manage. I really wanted to ask why they didn’t consider using computer vision instead, since controllers are quite expensive, especially if implemented for each slot. As it turned out, Egar asked that very question. I understood their reasoning, they needed to track every single spot — but still, it felt like an expensive idea.</p><p>The third group came with another issue, wildfires in Indonesia. They presented an early warning system designed to detect wildfires and report them for early countermeasures. Their plan was to use LoRa, which was a good idea, except for one challenge — the signal would need to pass through dense forest. It was a solid concept, but I would have loved it if they had also explained how they planned to conserve power. After all, the system would be deployed in the forest and would need to last for a long time, right?</p><p>Then came my group’s turn. Our project was about an anti-bullying system using computer vision, implemented through the school’s CCTV network. I was expecting a lot of questions, but it turned out only a few were asked. Egar questioned why we chose an on-premise server instead of a cloud server for the AI models. I explained that it was for performance and safety reasons, since CCTV footage is sensitive data that should not pass through a public network. Later, Mr. Husni asked about privacy concerns, and I replied that we could minimize exposure by using the on-premise server.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*lQ_aIjXKW9q7SVRHeClI4g.jpeg" /><figcaption>Treasure Hunt</figcaption></figure><p>And that was it — that’s how everything was done. Mr. Alwin then took the stage and introduced us to a game called treasure hunt. The rules were simple, find an egg, answer the questions, and get the clue for the next egg. All of us were very enthusiastic back then — even Tian, who usually didn’t have much spirit for activities, looked so excited. I swear to God, he was the one who loved running around searching for the eggs. I guess he just needed some fun and challenge during the camp, haha.</p><p>My group became the first to finish the game, and we answered all the questions correctly (I was kinda proud of that). Thanks to Tian, who was really eager to find all the eggs, haha. We just enjoyed the rest of the evening while waiting for the others to finish. I also took a short walk with Melody and Syifani under the lovely amber skies. We could also see Mr. Imtiaz taking some videos — or maybe vlogging, I wasn’t really sure. To our surprise, it was already time to depart. We saw others waving at us to take pictures, so we rushed back, grabbed our belongings, and took our very last photos together.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mae30efSZtEq-D9JVrzCBA.jpeg" /><figcaption>Final Shots, Closing a Beautiful Journey</figcaption></figure><p>Without taking another delay, we finally departed back to the city for dinner and the awarding night. The journey was quite long, but I couldn’t sleep — I just enjoyed the road and the scenery. Upon arrival, the scene was beautiful: clear skies and the sparkling city lights. It was at Casa Bakudapa, which Mr. Alwin told us basically means “a place to meet.”</p><p>We took seats wherever we wanted. Somehow, I ended up with Melody and Muhazar beside me. We enjoyed food and drinks, taking whatever we wanted before leaving Manado the next day. I sat there, savoring the moment, and helped Muhazar record some footage for his vlog. After everyone finished eating, Kudo-san took the stage and announced the end of the APIE camp program. He began calling our names group by group to give us the camp certificates. I watched them one by one, knowing deep down that I might never have the pleasure of seeing them all again in person.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*r0sevn9sS0ZQ7qn8PI1Mww.jpeg" /><figcaption>A Night of Pure Happiness This Year</figcaption></figure><blockquote>“So this is it, that’s how it ends,” I said to myself.</blockquote><p>I never expected that APIE camp would end this beautifully — and this fun, of course. All I had imagined before was endless configurations, heavy workloads, and stressful days. But it wasn’t like that. I mean it, really. And if I had to mention my happiest days throughout the year, <strong>they were right here, in Manado, with all those people.</strong></p><p>As the night drew to a close, I felt nothing but gratitude, for the laughter, the lessons, and the friendships that had colored this journey. Manado gave me more than just memories, it gave me a reminder that joy often comes when you least expect it. <strong>Thank you all — for everything.</strong> This camp became unforgettable because of you.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c33f27bc3ff3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Inside Forge Day: The Chaos, the Code, and the Collaboration]]></title>
            <link>https://radityanala.medium.com/inside-forge-day-the-chaos-the-code-and-the-collaboration-876d419f3926?source=rss-26bf16164f14------2</link>
            <guid isPermaLink="false">https://medium.com/p/876d419f3926</guid>
            <category><![CDATA[iot]]></category>
            <category><![CDATA[networking]]></category>
            <dc:creator><![CDATA[Yoga Raditya Nala]]></dc:creator>
            <pubDate>Sun, 26 Oct 2025 13:34:56 GMT</pubDate>
            <atom:updated>2025-10-27T03:40:16.801Z</atom:updated>
            <content:encoded><![CDATA[<p>It was Saturday, and we were all still feeling sleepy with our tired souls. The sun is shining in just the right way with a cold breeze. We spent the night before working on a final report that needed to be submitted immediately after the event. Yep, that’s us, doing anything near the deadline, even though it was announced a long time ago—<em>deadliner</em> they said.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*SzmgE-mB9M21M9ssypB00g.png" /><figcaption>Participants and staff of Forge Day 2025</figcaption></figure><p>As one of the education institutions that adhere to “<em>tridharma</em>”—or rather three duties (i.e., education, research, and community service), our lab often held a community service, though usually on a smaller scale than today’s event. Even though we’re still feeling sleepy, that morning was different. That morning, our community service titled “<strong>Forge Day</strong>” will be held. There are a lot of research topics that will be presented in two days. The first day will be about <strong>network infrastructure</strong> and <strong>cybersecurity</strong>, while the second day will be about <strong>virtualization &amp; containerization</strong> and the <strong>internet of things</strong>. The audience is vocational high school students, we are indeed looking for an audience that is a bit familiar with the topic we brought in.</p><h3>First Day</h3><p>After registration, greeting, and welcoming sessions from the committee, we all got busy. Some of the lab’s members will be the presenter, while the rest of us will be helping on the back to ensure that this event goes smoothly as planned. The first topic, network deployment, officially began. The room felt crowded and lively. The presentation started with how the internet was founded. Started from how the USSR managed to launch a satellite into orbit, which in turns caused the US government to panic and founded The Defense Advanced Research Project Agency, which later be known as ARPA. Its research is not mainly focused on the internet or network at all, however its research institutions were scattered all across the states, making communication and coordination difficult. They had to fly across the continent just to have a conversation, and surely this isn’t effective or efficient. In the late ’69, ARPA funded ARPANET (ARPA Network), connecting research institutions between the US west coast and east coast.</p><p>After participants understood how it all started, the topic started to shift to how each device could understand each other through the internet. TCP/IP is now officially introduced to the participants as the agreed communication protocol worldwide, and all those interconnected networks are now officially called internet. Then, the material started to shift to more technical terms, such as IP address, subnetting, and numerous routing protocols. <strong>Not only using a theoretical approach, there are also hands-on sessions</strong> to implement all that theory using simulators like GNS3. Don’t ask me why we’re not using a real device, it’s not easy to get one (and surely not cheap). A single Cisco router could precede a five month salary in my city, haha.</p><p>Long short story, the first topic is done, we continue to the next topic, cybersecurity. This transition might seem a bit off, considering both topics were not connected directly, especially this topic is more focused on the application layer, from the side of web development. However, this material perhaps will be <strong>more related to the participants</strong>, since some of them are coming from software engineering classes, which often do a lot of work on web development. The materials brought in are quite interesting, such as SQLi and Cross-Site Scripting (XSS). Although modern frameworks often block these attacks by default (e.g., Laravel, ExpressJS, Go Fiber, etc) by default, it is still necessary for all those participants to learn how things actually worked behind the scenes (validation, sanitation, encoding, etc). Just like before, this topic also has a hands-on session, where participants are able to try to attack the provided lab environment using materials provided before.</p><h3>Second Day</h3><p>The next day, not much changed. It’s more calm, not as hectic and busy as the first day, we already had a picture of how things will go. The first topic for this day is virtualization and containerization. This topic talks about what virtualization is, what’s the difference with containerization, what’s the benefit, etc. The hands-on session is quite straightforward, participants were asked to <strong>create a virtual machine</strong> from type 1 hypervisors (i.e., Proxmox). After successfully creating a VM, participants were instructed to <strong>install Docker, pull some images, and run containers</strong> that later will be used on the next topic. And here’s where the thing started to get interesting.</p><p>A few days before, Direktorat Teknologi Informasi (DTI) Universtias Brawijaya—you can think of them as network administrators at our university, announced that they will do a <strong>network migration that might affect digital activity at our university</strong>. Of course we get the message and understand what it means, but there is nothing we can do at that moment. Participants are already told to come at the specified time, rooms already reserved, food already prepared, etc. <strong>We can only hope and pray</strong> that the migration won’t affect our event. Unfortunately, what we feared the most finally happens. When participants are updating the Ubuntu repositories before installing Docker, the internet access within our university is lost. We couldn’t do anything at that time, it was far beyond our reach. Worse, <strong>if the VM still didn’t have internet access, the next topic is potentially cancelled</strong>, because it is dependent on the VM created from this topic.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Jj898EMoA65As9zz_FNgqw.png" /><figcaption>Network engineers that keep the event running</figcaption></figure><p>Luckily, the toughest moment finally passed. <strong>One of the lab members, </strong><a href="https://medium.com/u/e03462772573"><strong>Akbar Fikri Abdillah</strong></a><strong>, came up with a brilliant workaround</strong>—he suggested mimicking our university network using a small router and connecting it to a phone’s cellular data as a substitute for the campus network. I have to admit, Akbar is one of the brightest minds I’ve ever known in the lab—or perhaps even in the whole faculty, haha. It went smoothly, all participants were able to install Docker, and in turn successfully pull a few images and run containers needed in the next topic, such as Mosquitto and Node-RED.</p><h3>Internet of Things</h3><p>Then here we are, at the last topic, the internet of things. It is quite special for me, because I am also one of the presenters that will introduce to participants how these things actually work. <strong>I started it with a simple question</strong>, yet many of my fellows failed to answer,</p><blockquote>“What is IoT?”</blockquote><p>Few hesitated to answer, but finally there is an answer that satisfied me, things that are interconnected through the internet. A lot of people are often misunderstood that IoT is as simple as a node reading sensors and doing something from the data using actuators. Well, it’s actually far beyond that. It’s about how we collect those data, sending them through a server, orchestrating them, and creating a decision based on those collected data. And all of them were done using the help of the internet.</p><p>I don’t want to yap too much about the theory, but unfortunately it had to be done, to ensure that they left all the misunderstanding they had before. I also <strong>explain about MQTT</strong>, a lightweight protocol that is often used in IoT using the publish/subscribe mechanism. Of course, a lot of them are still confused with what that means. I then used a simple analogy using a messaging app where there are two main actors, administrator (publisher) who have the right to deliver a message (data) and members (subscriber) that are only able to read the message. And yeah, it’s actually as simple as that, and they all did understand how MQTT protocol works, at least the core concept of it.</p><p>Not only that, <strong>I also introduce them to Node-RED</strong>, a simple yet powerful visualization tool that can be used to visualize the data collected from MQTT through the internet that was initially published by each node. I also ensure that they understand the basic syntax of Arduino that later will be used by them, such as setup and loop function, serial, digital and analog, and so many others. It might feel boring to some, for sure, but this one is necessary to ensure they have basic understanding on how to do the hands-on session after this theoretical presentation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vWMuHDOKmLb4C6HaPur-Mg.png" /><figcaption>ESP32 reading data from DHT sensor indefinitely</figcaption></figure><p>After that, the hands-on session finally began, each <strong>participants were given a Bluino Development Boards</strong>, a learning boards with ESP32 as its microcontroller with a lots of sensors (e.g., DHT, Ultrasonic, Gyroscope, PIR, etc) and actuators (e.g., relay, LED, buzzer, display, etc). This development board is just right for teaching them, they just need to connect the signal cable to the microcontrollers, no need to worry that they will swap the VCC and GND cable that might burn the sensors, haha. This session was organized by two of my friends, the first one is about DHT sensors.</p><p>Just like before, there is a material about how DHT actually works and how we can take advantage of it. This one was certainly more interesting, because they’re not only studying something theoretical, but something they can see and utilize directly. After that, participants started to connect ESP32 to their laptop and wrote some code under the instruction of the presenter. The first task is to <strong>read the data from the DHT sensor</strong> (using help from the Adafruit library) and print it to the serial monitor. Most of them pass this task easily and the temperature and humidity data successfully printed to the serial monitor.</p><p>Next, they were asked to <strong>integrate actuators into the existing system,</strong> in this case, a buzzer that would ring if the temperature or humidity exceeded a certain value. This task introduced an <strong>event-driven architecture</strong>, where the node reacts to incoming data by performing an action. Participants handled this part easily. Soon, the room was filled with random buzzer rings, sounding like a <em>chaotic digital orchestra</em>.</p><p>Then came the most challenging part, <strong>sending data from the DHT sensor to an MQTT server</strong>. The ESP32 now had to act as a publisher, sending messages to specific topics. To do this, they first needed to connect to the access point provided by the staff to gain network access. On top of that, <strong>they had to set up a Node-RED dashboard</strong> to subscribe to the same topics and visualize the data in real time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5TiN2OSSxL_0SXew3vmcyw.png" /><figcaption>A glimpse of me debugging code that was expected to work</figcaption></figure><p>And, of course, that’s when the real debugging began. <strong>Some participants couldn’t connect to the MQTT server</strong>, often because of something as small as a wrong IP address or a mismatch in the network configuration. One particularly funny case had someone unable to compile their Arduino code at all. I joined in to help troubleshoot, but after several failed attempts, I gave up, completely stumped—only to find out later that another instance of the Arduino IDE had been left running in the background the whole time. Thankfully, with the help of the teaching assistants, everyone managed to get their systems running. It took longer than expected, but seeing the MQTT messages flow into the dashboard made every bit of the struggle worth it.</p><p>After successfully building a node using a single sensor, it is time to explore other sensors. Unfortunately, the time is almost over. Because of that time constraint, the last session about ultrasonic sensors is just a theory presentation by another lab member. As the last slide was presented, the screen went dark, and the chatter began to fade, the room slowly filled with that strange mix of relief and pride. Some participants were still tinkering with their boards, others were already asking questions far beyond the material we had planned. We looked at each other—tired, but genuinely happy. The sleepless nights, the network outage, the chaotic setup, somehow, everything felt worth it.</p><h3>The End</h3><p>Forge Day wasn’t just another community service event. It was a reminder of how fragile yet powerful collaboration can be, how something as simple as a lost connection could push us to think creatively and adapt. We came as sleepy students rushing deadlines, but left as a team who learned to face unexpected challenges head-on. At the end of the day, technology isn’t just about devices or code, <strong>it’s about people working together to make things possible</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zY6i_msdbeSrsIhXh_0JSA.png" /><figcaption>From left to right: Nizar, me, and <a href="https://medium.com/u/a3db964821ea">Ahmadnafim</a></figcaption></figure><p>I would also like to extend my heartfelt appreciation to the IoT teams for their invaluable support throughout this event. Without them, this event would not have been possible. Thank you all.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=876d419f3926" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Transpubin: An Initiative Efforts to Revitalize Public Transportation]]></title>
            <link>https://radityanala.medium.com/transpubin-an-initiative-efforts-to-revitalize-public-transportation-f2a391a5f961?source=rss-26bf16164f14------2</link>
            <guid isPermaLink="false">https://medium.com/p/f2a391a5f961</guid>
            <category><![CDATA[iot]]></category>
            <category><![CDATA[internet-of-things]]></category>
            <dc:creator><![CDATA[Yoga Raditya Nala]]></dc:creator>
            <pubDate>Wed, 22 Oct 2025 15:08:42 GMT</pubDate>
            <atom:updated>2025-10-23T01:53:37.215Z</atom:updated>
            <content:encoded><![CDATA[<p>Angkot is one of the <strong>oldest public transportation systems in Malang</strong>, one of the major cities in East Java, Indonesia. It can be traced back to 1931 for its appearance (Novrian, 2025), long before Indonesian independence. Even though the system might be different from what we’re seeing today, one thing remains, it’s cheap and affordable for people.</p><h3>The Decline of Angkot</h3><p>However, its popularity has been decreasing ever since. Several factors contribute to this decline—the rise of online transportation, an inadequate fleet, reckless driving, lack of real-time monitoring from the ministry, and long, uncertain waiting times, etc. Perhaps, at this point you’ll ask a few questions.</p><blockquote>“Why bother? Why not just move on to the new technology?”</blockquote><p>As of this moment when I’m writing this story, <strong>three thousand people are on the edge of unemployment</strong> (Midaada, 2024). When this condition doesn’t get better, three thousand families will have no chance to make a better future. Three thousand families whose only hope is that the system somehow improves. It’s also getting worse if we’re taking into account the fare of this transportation. It <strong>only costs about 5.000 IDR—about 30 cents USD</strong> for almost all routes with the longest route spanning for about 18km. You also only charged once, no matter how long the distance you took. Because of this, <strong>drivers sometimes overload the passenger, sacrificing other passenger comfort just to get more money</strong>, and it is totally understandable considering the very low cost fare. The ministry didn’t seem to pay much attention to them, whether they were overloading, driving recklessly, smoking while driving, or simply speeding.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KglkLGuCXgC-fJSPFUp2ew.jpeg" /><figcaption>A gloomy evening, waiting for someone who won’t even arrive</figcaption></figure><p>Another reason people—including me, avoid it, is the <strong>lack of a fixed schedule</strong>. It is really up to the driver whether they want to arrive at where and at what time. Passengers really have no control over them. We often left waiting for them to arrive with uncertainty whether they will eventually arrive or never at all. At some point, I waited for almost an hour, yet none of them arrived. <strong>Drivers often drive recklessly—smoking, speeding, and breaking rules that endanger passengers</strong>. Not only that, many of their fleets are obsolete or inadequate. The Department of Transportation itself said that <strong>from the 1.666 fleet, only 200 remain in good condition</strong> to operate. This of course raises more questions whether this old transportation is still worth using or not, considering its safety that is often left aside.</p><h3>Introducing Transpubin</h3><p>And here I am-with two friends and our lecturer—determined to find a way technology can breathe new life into this old system. So what do we really want to propose here? Let us introduce <strong>Transpubin, a smart system that integrates real-time location tracking, passenger count, and smart payment</strong>, all in one small device powered by ESP32. This device, which we refer to as a node, will be installed in every vehicle. Each node will send its location to a centralized server using a lightweight protocol such as MQTT. Payment data, on the other hand, will use a request–response protocol such as HTTP or HTTPS to ensure reliable transaction handling.</p><blockquote>“But why would we need all those sensors to solve these issues? How far could it actually help?”</blockquote><p>To answer that, we need to go to the very beginning of this article. In an earlier section, I’m stating that Angkot’s popularity started to decrease dramatically since the appearance of online transportation, <em>but why tho?</em> Turns out, the reason behind it is simply that <strong>most of the problems we address in Angkot before are already answered by modern architecture of online transportation.</strong></p><p>When using online transportation we don’t have to wait at the side of the road for hours with uncertainty. We just click order and wait for them to arrive with an estimated time of arrival and specified price. The drivers were also strictly monitored and regulated. They can’t speed because the automation detection system will send speed information to their system, thus adding a penalty to the driver, which can result in suspension. Customer complaints were also highly regarded, customer service often takes this matter seriously and didn’t hesitate to take necessary action against the driver. The rating system also makes the driver as kind as possible to increase customer satisfaction with the hope that customers will give them a high rate and avoid complaints. In essence, what online transportation offers isn’t a new kind of vehicle, it’s a new kind of information system. Transpubin aims to give Angkot that same backbone of digital intelligence—visibility, regulation, and trust.</p><h3>How The System Works</h3><p>The core idea is straightforward yet powerful. Each node integrates essential sensors, <em>NEO6M</em> for GPS, <em>MFRC-522</em> for RFID, and an RGB LED indicator, under the control of an ESP32 microcontroller. Because most Angkot routes operate across wide and often infrastructure-limited areas, we opted for GSM modules such as the <em>SIM800</em> series to send data over cellular networks to a centralized server. Using community-maintained Arduino libraries like <em>TinyGSM</em> and <em>PubSubClient,</em> we implemented both HTTP and MQTT protocols for data transmission with minimal overhead.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1013/1*7vYqydp5KN-YurqJYIxmug.png" /><figcaption>High Level System Design of Transpubin</figcaption></figure><p>Each node publishes its location data to the MQTT broker (we use HiveMQ for this implementation). The backend subscribes to the relevant topics, processes incoming payloads, and rebroadcasts them to the frontend through Server-Sent Events (SSE), enabling <strong>real-time visualization</strong> as long as the nodes continue transmitting. SSE was chosen over WebSocket due to its lower complexity and native support for unidirectional, event-based updates from server to client. To a passenger, this just means one thing, <strong>they can finally see where their Angkot is, live, without waiting endlessly</strong> at the roadside.</p><p>Among the topics published, one specifically carries location data, providing the current coordinates of each Angkot. These coordinates can be leveraged for various features such as speed estimation, direction detection, and real-time tracking. Each node continuously emits these payloads to the HiveMQ broker, which are then consumed and processed by the backend. Those coordinate data can be utilized into many features, such as speed measure, direction detection, and real-time tracking. Each node will emit these payloads to the HiveMQ which is later consumed by the backend.</p><pre>{<br>  &quot;vehicleId&quot;: &quot;1234xxx&quot;,<br>  &quot;lat&quot;: &quot;40.6892&quot;,<br>  &quot;long&quot;: &quot;-74.0445&quot;,<br>  &quot;timestamp&quot;: 2025-10-15T23:43:14+00:00<br>}</pre><p>We store latitude and longitude as strings rather than floating-point numbers to avoid rounding errors that can accumulate during serialization and parsing, especially when coordinates are transmitted frequently.</p><p>Based on those data, the backend can theoretically calculate vehicle speed by dividing the distance between two consecutive location points by the time interval between them. However, because the Earth is a sphere, computing distance directly from latitude and longitude requires a spherical geometry formula rather than simple Cartesian math. To address this, we use the Haversine formula, which estimates great-circle distance between two points on a sphere given their latitudes and longitudes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*k-c-AbkkCqqVeL8xjEqXXA.jpeg" /><figcaption>The Harvesine Formula</figcaption></figure><p>r is the earth radius (approx. 6371km), phi is the latitude (in radians), and lambda is the longitude (in radians).</p><p>After successfully determining the distance between two points—or rather payloads, we can finally use a simple formula by dividing the distance we just calculate using the Haversine formula with the delta from timestamp embedded in each payload. By doing this, we can theoretically estimate each vehicle’s speed and present it to the user and stakeholders.</p><p>At first glance, the computation seems straightforward, it looked simple in theory, but real-world conditions reminded us that theory rarely survives the first field test. Just like that when we finally realized that inexpensive <strong>GPS modules often struggle under signal obstructions</strong>. We had another idea of using signal trilateration to obtain location. The concept itself is benefiting from the fact that the GSM module will get signals from multiple base stations. Each of those base stations will be received as different signal strength from the node. Based on those environmental noise, making distance estimation imprecise without extensive calibration.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Vx82DHElUKdVKd8_yC_a6Q.jpeg" /><figcaption>A fading light, where everything begins to fall apart</figcaption></figure><p>Furthermore, our <strong>GSM module suffered a similar problem</strong>, it often failed to connect under certain conditions. It simply doesn’t get a signal from any provider, rendering the whole system unresponsive. If it is able to send data, it will be at a very slow rate. Well, no wonder why it happens, GSM which only uses 2G signals are often discarded by cellular providers and switched to the latest technology such as 3G, 4G, or even 5G. <strong>It is indeed a great idea to get newer models such as SIM7000 series</strong>, but our budget simply says no, it is extremely expensive compared to the GSM models.</p><p>There was also this one funny moment when I bought the wrong GSM modules. At first, I didn’t know much about the different types, and SIM800 and SIM900 were the ones that popped up in my Google search. Without doing any further research, my brain simply concluded that SIM900 had to be superior—after all, it had a bigger number. But I was completely wrong. The SIM800 was actually smaller, more compact, and supported more frequencies, making it far more reliable. In the end, I bought the newer models anyway, even though both still underperformed under certain conditions.</p><p>After numerous trials, we finally managed to get both the GPS and GSM modules working together. We began testing inside the building, on the 9th floor, but the GPS didn’t receive any data at all—it looked completely dead. At first, I thought I’d been scammed with defective units. We moved to the 4th-floor balcony, at least with some view of the sky, yet it still showed no signs of life. Even worse, the GSM module stopped working entirely.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PCg7ISvQ45sKVy4XKzY43Q.jpeg" /><figcaption>The happiest night when everything started working according to plan</figcaption></figure><p>Finally, as a last resort, we tested the system in an open field. To our surprise, it worked flawlessly. The node began receiving coordinates with an error range of about 5 to 10 meters—<strong>accurate enough for our needs</strong> at that time. We were thrilled; even the GSM module performed perfectly, successfully sending data to the MQTT broker. By then, we understood what we needed to do, <strong>mount both the GSM and GPS antennas above the vehicle</strong> to ensure a clear line of sight to the sky, even if it was just an interim solution. However, location tracking solved only half of our problem—the other half was determining who was actually inside the vehicle.</p><p>The RFID sensor is used to support the <strong>smart payment and passenger counting systems</strong>. Our first idea was to use a load cell sensor to detect the overall weight of the passenger and it seems promising at first glance. However, judging how crowded a vehicle could be only from its weight sometimes can’t be representative. Because of that, we decided to look the other way around, utilizing the same RFID sensor that was initially used for payment, to detect the number of passengers.</p><p>The concept is straightforward. Whenever a passenger boards the vehicle, they are required to tap their ID card on the RFID sensor. The system then registers the card ID to the server and increments the passenger counter for that specific vehicle. Similarly, when a passenger disembarks, they must tap their ID card again. Upon receiving an RFID tap, the system checks whether the corresponding ID already exists in the temporary database. If the ID is found, it can be inferred that the passenger is exiting the vehicle—the counter is decremented, and the ID is removed from the temporary list. Otherwise, if the ID is not found, the passenger is boarding, so the counter is incremented and the ID is added. The algorithm can be summarized in the following pseudocode.</p><pre>if (card.isPresent()) {<br>  if card.isInDatabase() {<br>    card.remove();<br>    counter--;<br>  } else {<br>    card.add();<br>    counter++;<br>  }<br>}</pre><p>However, despite the simplicity of the initial concept, several unexpected challenges emerged. Our first approach was to use the national ID embedded in each citizen’s identification card as the user identifier. This method would have allowed passengers to use their existing ID cards without prior registration, while the server automatically created a new account based on the received ID.</p><p>In practice, however, this approach proved infeasible. When attempting to read the card data, <strong>the national ID field was found to be encrypted</strong>, and no accessible identifier could be retrieved. Due to these technical limitations, we were forced to revise our approach. The system was eventually adapted to rely on the card’s internal identifier, which required manual registration before use—introducing an additional layer of inconvenience compared to the original design. This challenge reminded us of the broader reality of working across multiple domains, where every improvement often reveals new constraints. It also shaped the next phase of our development.</p><p>Talking about Transpubin wouldn’t be complete without mentioning its application layer. Three of us come from a web development background—two, including myself, are backend developers with months (if not years) of experience, while the third specializes in machine learning and data engineering. <strong>None of us, however, are experts in IoT or embedded systems</strong>, which explains why our early prototypes were quite messy. Still, that made our direction clear, we would finish the IoT prototype together, then move swiftly into web development—if time allowed.</p><p>As shown earlier, the architecture is simple—a backend server acts as an intermediary between IoT nodes and the client, while also managing user data and payments in the database. Everything went smoothly until we realized one crucial gap—we didn’t have a UI/UX designer to handle the website’s front end. Because of that, we decided to postpone frontend development and focus on the core communication between the IoT node and the backend server. We experimented with various protocols, including REST API, MQTT, and Server-Sent Events (SSE). The biggest challenge came when configuring MQTT to run over WebSocket, since plain TCP connections aren’t well supported through proxies in the classic PubSubClient library. Fortunately, our broker (HiveMQ) natively supports MQTT over WebSocket, which saved us a lot of trouble.</p><p>Another major challenge came from the encrypted data inside national ID cards. Initially, we wanted to use this data to automatically fetch user accounts without requiring them to register through our website, which would’ve added an extra layer of complexity. As an interim solution, we simply “pretended” the detected ID from the RFID reader was identical to the user’s national ID. This let us focus on the system’s core functionality first and find a proper solution later as the project evolved.</p><h3>Future Works</h3><p>The successful implementation of our proof of concept has proven that the system can work in practice, and it opens a wide range of future research opportunities. The data collected from each node, such as vehicle position, speed, and even passenger count can serve as the foundation for developing more advanced applications and optimizations, such as</p><ul><li><strong>dynamic traffic light control</strong> based on real-time traffic density and vehicle speed data collected across multiple nodes,</li><li><strong>public transportation route optimization</strong> using aggregated passenger tracking data to identify underused or overcrowded routes,</li><li><strong>passenger behavior analytics</strong>, to understand peak usage patterns and improve service planning through data-driven decisions,</li><li><strong>comparative study on quality of service (QoS)</strong> using different GSM modules for real-time location transmission, focusing on latency, connection stability, and data throughput under varying signal conditions, and others.</li></ul><h3>Conclusion</h3><p>Technology isn’t something we can deny. A war against it can’t be won by any man—only by another technology. Perhaps the effort to revitalize Angkot through modern tools is the right kind of war—one fought not with destruction, but with hope. Transpubin might not solve everything, but it shows that innovation isn’t just about moving forward, sometimes, it’s about looking back and lifting what’s been left behind.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JoPZA12D3G7cuifCNFePxA.jpeg" /><figcaption>The Remnant of the Transpubin Prototype</figcaption></figure><p>We know Transpubin is still far from perfect. There are many moving parts that require further attention, numerous stakeholders to consider, and countless details that need refinement. The modules fail under bad signal, the sensors sometimes misread, and many ideas remain only on paper. The current prototype, though functional, can also be greatly improved—most notably by replacing the messy wiring with a proper printed circuit board (PCB) design.</p><p>But even so, this project is our small attempt to ask a bigger question, <strong>can technology revive something that’s been left behind?</strong> Maybe the answer isn’t clear yet—but every working prototype, every line of code, every failed test moves us a step closer to understanding how the old and new can coexist.</p><p>This project was first conceived for the Gemastik 2025 Smart City Competition. We never made it to the finals—<strong>but ideas don’t need trophies to live</strong>. I decided to let this one live on, for anyone who still believes that even the most traditional systems can find new life through technology.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ARqXo-SM4UQHsfqUFfDzbw.jpeg" /><figcaption>The people who made this proof of concept happen (somehow)</figcaption></figure><p>Last but not least, I would like to express my deepest gratitude to my friends who have kept this project alive for more than half a year, maintaining their commitment and passion even when things didn’t always go as planned. This project still needs a lot of refinement, but it has come this far because of their dedication. I also want to extend my sincere thanks to our lecturer, who has guided us throughout this journey. Without their support and encouragement, this project would never have become a reality.</p><p>If you’re interested in this project, feel free to visit our <a href="https://github.com/Angkutin-Gelatik">GitHub</a> and build upon it.</p><h3>Reference</h3><p>Calculate distance between two latitude-longitude points? (Haversine formula). (2017, July 21). Stack Overflow. <a href="https://stackoverflow.com/a/27943">https://stackoverflow.com/a/27943</a></p><p>Knolleary. (n.d.). <em>GitHub — knolleary/pubsubclient: A client library for the Arduino Ethernet Shield that provides support for MQTT.</em> GitHub. <a href="https://github.com/knolleary/pubsubclient">https://github.com/knolleary/pubsubclient</a></p><p>Midaada, A. (2024, July 10). Angkot di Malang Banyak Tak Layak Jalan, Ribuan Sopir Terancam Nganggur. Okezone News. <a href="https://news.okezone.com/read/2024/07/10/519/3032457/angkot-di-malang-banyak-tak-layak-jalan-ribuan-sopir-terancam-nganggur">https://news.okezone.com/read/2024/07/10/519/3032457/angkot-di-malang-banyak-tak-layak-jalan-ribuan-sopir-terancam-nganggur</a></p><p>Novrian, A. (2025, August 21). Transformasi Angkot di Kota Malang dari Era Kolonial sampai Sekarang. Jawa Pos Radar Malang. <a href="https://radarmalang.jawapos.com/kota-malang/816462540/transformasi-angkot-di-kota-malang-dari-era-kolonial-sampai-sekarang">https://radarmalang.jawapos.com/kota-malang/816462540/transformasi-angkot-di-kota-malang-dari-era-kolonial-sampai-sekarang</a></p><p>Vshymanskyy. (n.d.). <em>GitHub — vshymanskyy/TinyGSM: A small Arduino library for GSM modules, that just works</em>. GitHub. <a href="https://github.com/vshymanskyy/TinyGSM">https://github.com/vshymanskyy/TinyGSM</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f2a391a5f961" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introduction to Git]]></title>
            <link>https://radityanala.medium.com/introduction-to-git-b7784a415b11?source=rss-26bf16164f14------2</link>
            <guid isPermaLink="false">https://medium.com/p/b7784a415b11</guid>
            <category><![CDATA[coding]]></category>
            <dc:creator><![CDATA[Yoga Raditya Nala]]></dc:creator>
            <pubDate>Tue, 19 Aug 2025 12:01:35 GMT</pubDate>
            <atom:updated>2025-08-21T07:37:38.019Z</atom:updated>
            <content:encoded><![CDATA[<p>Pernah ga sih, kalian bikin sebuah dokumen, proposal, atau apapun itu, tapi <strong>punya banyak banget versi</strong>, sampai kalian sendiri bingung dengan mana versi <em>file</em> yang masih <em>draft</em> dan sudah siap untuk dikumpulkan?</p><p>Hal yang sama juga dialami sama developer dan programmer di era awal teknologi. Tepatnya di sekitar tahun 2005, ketika sebuah <em>project</em> besar yang dikenal dengan nama Linux Kernel sedang berkembang. Kala itu, update dan patch <strong>didistribusikan menggunakan <em>archived files</em> biasa secara manual</strong>. Hal itu tentunya menjadi kekurangan sendiri, terlebih lagi ketika ada banyak sekali developer dan kontributor yang mengerjakan project yang sama. Bayangkan, mereka harus mendistribusikan file kode ke banyak orang sambil berharap nanti hasil akhir file nya <strong>tidak saling konflik</strong> dan overlap. Belum lagi misal ternyata ada kode dari salah satu kontributor yang menyebabkan aplikasi <em>crash</em>, <em>error</em>, dan berbagai kesalahan lainnya, maka perlu <strong>mencari lagi versi aplikasi sebelumnya dari tumpukan draft</strong> di file explorer untuk dimasukkan lagi ke kode utama.</p><p>Hingga pada akhirnya, Linus Torvalds yang merupakan salah satu pengembang Linux akhirnya memutuskan untuk membuat Git, sebuah <strong>Version Control System</strong> yang memungkinkan developer untuk tracking kode mereka, melihat apa saja yang mereka ganti dari versi sebelumnya, hingga <strong>mengembalikan kode ke versi sebelumnya</strong> apabila ditemukan bug atau error. Kerennya lagi, Git <strong>bisa dihubungkan ke remote repository</strong>, membuat distribusi kode lebih mudah sembari memastikan integritas kode, sehingga memudahkan para developer untuk berkontribusi dalam <em>project</em>-<em>project</em> besar lain.</p><h3>Version Control System</h3><p>Seperti yang aku bahas sebelumnya, salah satu fitur yang bikin Git ini banyak dipakai adalah kemampuannya untuk <strong><em>revert</em> file</strong> atau lebih tepatnya <em>state</em> <strong>ke versi tertentu</strong>. Git memiliki kemampuan ini karena ia mencatat setiap perubahan antar versi, sehingga bisa merekonstruksi <em>state</em> atau keadaan dari versi yang diinginkan.</p><blockquote>Git basically takes a picture of what all your files look like at that moment and stores a reference to that snapshot. To be efficient, if files have not changed, Git doesn’t store the file again, just a link to the previous identical file it has already stored. (What is Git? — <a href="https://git-scm.com/">https://git-scm.com</a>)</blockquote><p>Sebagai tambahan, Git tidak hanya mencatat perubahan yang dibuat saja, tapi lebih ke <strong>mencatat bagaimana keadaan satu folder (<em>repository</em>) ketika versi (<em>commit</em>) tersebut dibuat</strong>. Ketika ada file yang tidak berubah, maka cukup catat referensi ke file itu saja, sehingga dapat menghemat tempat dengan tidak mencatat isi semua <em>file</em>. Baru ketika ada <em>file</em> yang berubah, maka <em>content</em> perubahan tersebut dicatat sebagai bagian dari versi tersebut.</p><h3>Git v. GitHub</h3><p>Kebanyakan dari kita mungkin sudah mengenal GitHub, tapi tidak sedikit juga yang mungkin kurang mengerti perbedaan antara GitHub dengan Git. Singkatnya, Git ini sistem yang bakal track perubahan kode di local system kita, sementara <strong>GitHub</strong>, ini <strong>tempat untuk menyimpan repository secara <em>remote</em> alias <em>online</em></strong>. GitHub ini sangat memudahkan kita untuk sharing kode dengan teman atau dengan kontributor lainnya. Berbeda dengan era pengembangan Linux Kernel yang patch didistribusikan manual melalui email, di era kita sekarang, kita bisa mengerjakan sebagian kode di local system, lalu kalau sudah selesai bisa upload ke GitHub, dan diakses orang lain. Kontributor lain juga bisa melakukan hal yang sama, sehingga setiap perubahan akan terkumpul jadi satu di remote repository. Walaupun mungkin terlihat sama, perbedaan keduanya sangat signifikan, yang satu sebagai sebuah sistem untuk mengontrol versi kode, yang satunya lagi sebagai tempat untuk menyimpan tracking tersebut secara <em>online</em>.</p><h3>Staging</h3><p>Staging di sini merupakan sebuah terminologi yang merujuk pada sebuah <em>waiting list</em> yang berisikan <em>file</em>-<em>file</em> yang ingin kita simpan perubahannya (<em>commit</em>). Mungkin sedikit berbeda dengan menyimpan dokumen secara tradisional yang cukup dengan klik tombol <em>save</em>, di Git, ketika kita ingin menyimpan sebuah versi, kita perlu <strong>memilih <em>file</em> apa saja, atau lebih tepatnya, bagian mana saja yang ingin disimpan</strong>. Dengan demikian, kita bisa menyimpan kode yang sudah siap dikirim, sembari mengerjakan fitur lain tanpa khawatir untuk merusak kode utama. File-file yang kita ubah dalam sebuah repository berada dalam <em>working directory</em>, yaitu <em>state default</em> ketika terjadi perubahan pada sebuah file. Baru ketika kita ingin menyimpan perubahan tersebut, kita pindahkan <em>state</em> dari file-file tersebut ke staging area untuk di-<em>commit</em>.</p><p>Cara melakukan staging sebenarnya juga cukup mudah, cukup dengan memilih file yang ingin dimasukkan ke staging area jika menggunakan IDE, atau mungkin dengan menggunakan commands berikut jika melalui CLI.</p><pre>$ git add &lt;nama_file&gt; # menambah satu file<br>$ git add &lt;direktori&gt; # menambah satu direktori<br>$ git add --all # menambah semua file</pre><p>Contohnya, ketika aku membuat file baru bernama README.md dan ingin menambahkannya ke staging area, maka aku bisa menggunakan command berikut.</p><pre>$ git add README.md # atau dengan git add --all<br>$ git status<br>On branch master<br>Your branch is up to date with &#39;origin/master&#39;.<br><br>Changes to be committed:<br>  (use &quot;git restore --staged &lt;file&gt;...&quot; to unstage)<br> new file:   README.md</pre><p>Dapat dilihat pada output eksekusi di atas, git status berhasil menunjukkan bahwa README.md telah masuk ke staging area, dan siap untuk di-commit. Untuk mengeluarkan sebuah file dari staging area (unstage) sebenarnya cukup mudah, bisa dengan menggunakan command yang diberikan di atas, yaitu melalui git restore --staged &lt;file&gt; .</p><h3>Commit</h3><p>Nah, seperti yang aku bahas di awal tadi, pasti kita pernah mengalami kondisi di mana sebuah file itu punya banyak banget versi yang bikin kita jadi bingung. Contohnya mungkin di bawah ini, ketika ada sebuah tugas yang mungkin punya banyak revisi. Kebanyakan dari kita mungkin umum mengirimkan file tersebut ke rekan-rekan satu kelompok untuk bergantian menambahkan pekerjaan masing-masing, yang akhirnya menyebabkan menumpuknya banyak file. Di bawah ini, ada banyak sekali file dengan title yang kurang lebih sama dan cenderung kurang informatif. Kita tidak tahu versi mana yang paling baru atau yang paling siap untuk dikirim sebelum memastikan isinya satu persatu.</p><pre>akademik<br>└── tugas_akhir_paw<br>    ├── tugas_akhir_paw.docx<br>    ├── tugas_akhir_paw(1).docx<br>    ├── tugas_akhir_paw(2).docx<br>    ├── tugas_akhir_paw_fix.docx<br>    ├── tugas_akhir_paw_fix(1).docx<br>    └── tugas_akhir_paw_final_fix.docx</pre><p><strong><em>Commit</em></strong> di sini bisa dianalogikan dengan <strong>membuat versi baru</strong>, tetapi lebih jelas mengenai apa saja yang diubah, siapa yang mengubahnya, dan kapan. Kalau sebelumnya kita sudah memahami mengenai staging area, maka commit ini adalah proses menyimpan <em>file</em>-<em>file</em> yang ada pada <em>staging area</em> tersebut menjadi satu kesatuan versi. Ketika melakukan <em>commit</em>, kita juga bisa menyertakan informasi mengenai apa saja yang kita ubah dalam file-file tersebut melalui <em>commit message</em>.</p><p>Untuk melakukan commit sebenarnya juga cukup mudah, pastikan terlebih dahulu file-file yang ingin di-<em>commit</em> (atau dibuat versi baru) telah masuk ke staging area. Setelah itu, kita bisa menggunakan command git commit -m &lt;pesan commit&gt; seperti di bawah. Setelah itu, kita akan mendapatkan informasi mengenai berapa banyak file yang diubah dan berapa line yang ditambah atau dihapus.</p><pre>$ git commit -m &quot;docs: add README file&quot;<br>[master 142eb31] docs: add README file<br> 1 file changed, 20 insertions(+), 0 deletions(-)<br> create mode 100644 README.md</pre><p>Setelah melakukan <em>commit</em>, kita bisa melihat <em>log</em> dari semua <em>commit</em> yang telah kita lakukan. Sebagai contoh, pada <em>project personal</em> yang aku kerjakan di bawah, ada beberapa <em>commit</em> dari user lain beserta informasi mengenai apa yang mereka ubah dari kode utama pada <em>commit</em> tersebut.</p><pre>$ git log<br>commit 142eb3132a70097ced7e60ead68b5b5d8859900e (HEAD -&gt; master)<br>Author: Yoga Raditya &lt;******@gmail.com&gt;<br>Date:   Mon Aug 18 19:32:29 2025 +0700<br><br>    docs: add README file<br><br>commit 72eb24d1f5047d41a5bd0ee7f7d27e7d50192ac5 (origin/bugfix/error-check)<br>Author: agungaryansyah &lt;******@gmail.com&gt;<br>Date:   Sat Apr 19 15:58:48 2025 +0700<br><br>    fix: use errors.Is() instead of directly comparing error<br><br>commit b30f54024adb1855bf7467c2ef138248c6741c1d<br>Author: agungaryansyah &lt;******@gmail.com&gt;<br>Date:   Sat Apr 19 15:53:10 2025 +0700<br><br>    fix: change every errors.new to a static error</pre><p>Jika dibandingkan dari kasus awal tadi, tentunya Git ini sangat bermanfaat untuk menyimpan track dari berbagai versi file yang kita buat, terlebih lagi jika pesan commit yang kita sampaikan sangat informatif. Sebenarnya tidak ada aturan khusus yang membatasi mengenai pesan commit ini, tetapi beberapa project atau perusahaan memiliki aturan pemberian pesan <em>commit</em> sendiri untuk mempermudah dalam melakukan version tracking. Misalnya saja ketika ditemukan bug, maka akan lebih mudah bagi kita untuk mencari tahu di mana (atau di versi mana) masalah mulai muncul, sehingga dapat segera dilakukan penanganan. Berbeda lagi ketika pesan <em>commit</em> diisi dengan pesan-pesan sembarangan (atau bahkan sumpah serapah karena frustasi) yang menyebabkan pencarian versi menjadi lebih sulit. Terdapat konvensi yang secara umum memberikan arahan mengenai bagaimana pesan <em>commit</em> seharusnya diberikan, yang umumnya disebut sebagai <em>Semantic Commit Message</em>.</p><h3>Push</h3><p>Setelah berhasil melakukan berbagai commit, tiba saatnya bagi kita untuk membagikan repository kita ke orang lain, atau mungkin sekadar untuk disimpan secara online ke remote repository. Pertama, kita perlu set terlebih dahulu <em>remote origin</em> URL dari remote repository tujuan kita. Singkatnya, langkah ini bertujuan untuk memberi tahu Git ke mana kita ingin “meng-<em>upload</em>” repository kita.</p><pre>$ git remote add origin &lt;origin_url&gt;</pre><p>Kalau kita berbicara soal GitHub, maka cara mendapatkan origin URL tersebut sangat mudah, cukup dengan mengambil URL dari repository yang baru kita buat di GitHub. Sementara itu, git remote -v bisa dipakai untuk mengonfirmasi <em>remote origin</em> URL yang baru saja ditambah. Sebagai contoh, di bawah ini saya mencoba menambahkan <em>remote origin</em> URL dari GitHub ke salah satu repository <em>local</em> yang saya buat.</p><pre>$ git remote add origin https://github.com/yogarn/repository<br>$ git remote -v<br>origin https://github.com/yogarn/repository (fetch)<br>origin https://github.com/yogarn/repository (push)</pre><p>Setelah berhasil menambahkan remote origin URL, kita dapat melakukan proses push dengan menggunakan command git push origin &lt;branch&gt;, dengan branch yang secara default adalah <em>main</em>. Branch sendiri bisa dibahas lebih lanjut karena merupakan hal yang jauh lebih kompleks dari <em>flow</em> sederhana <em>staging-commit-push</em>.</p><pre>$ git push origin main<br>Enumerating objects: 4, done.<br>Counting objects: 100% (4/4), done.<br>Delta compression using up to 16 threads<br>Compressing objects: 100% (2/2), done.<br>Writing objects: 100% (3/3), 809 bytes | 809.00 KiB/s, done.<br>Total 3 (delta 1), reused 0 (delta 0), pack-reused 0 (from 0)<br>remote: Resolving deltas: 100% (1/1), completed with 1 local object.<br>remote: <br>remote: GitHub found 5 vulnerabilities on yogarn/repository&#39;s default branch (3 high, 2 moderate). To find out more, visit:<br>remote:      https://github.com/yogarn/repository/security/dependabot<br>remote: <br>To https://github.com/yogarn/repository<br>   82c1137..142eb31  master -&gt; master</pre><p>Dapat dilihat pada output command di atas, repository berhasil di “<em>upload</em>” ke <em>remote</em> repository. Sekarang kita dapat melihat repository tersebut secara <em>online</em> melalui GitHub.</p><h3>Pull</h3><p>Lalu bagaimana jika ada orang lain yang membuat perubahan, lalu melakukan push ke remote repository kita? Bagaimana cara kita “download” perubahan tersebut dengan mudah? Tenang, fitur “pull” dibuat untuk melakukan tugas tersebut. Command nya cukup mirip, yaitu git pull origin &lt;branch&gt;, dengan branch merujuk pada branch mana yang ingin kita “download” perubahannya. Dalam contoh di bawah, saya mencoba untuk mengambil perubahan dari branch default main dan hasilnya menunjukkan bahwa ada beberapa file yang masuk ke local repository saya karena kontributor lain membuat perubahan pada branch main di remote repository.</p><pre>$ git pull origin main<br>Updating 1139c5e..82c1137<br>Fast-forward<br> internal/repository/cart.go | 10 +++++++ — -<br> internal/repository/user.go | 3 — -<br> internal/service/book.go | 3 ++-<br> internal/service/checkout.go | 7 +++ — — <br> internal/service/payment.go | 40 ++++++++++++++++++++ — — — — — — — — — — <br> pkg/jwt/jwt.go | 2 +-<br> 6 files changed, 33 insertions(+), 32 deletions(-)</pre><p>Dengan demikian, proses pengerjaan kode program jadi jauh lebih mudah. Cukup fokus <em>coding</em> tanpa perlu takut ada <em>error</em>, karena sewaktu-waktu masih bisa di-<em>revert</em>. Pengembangan juga jadi lebih lancar, karena masing-masing developer bisa saling melihat perubahan satu sama lain, dan proses jadi lebih cepat dengan fitur <em>push-pull</em>, tanpa perlu distribusi patch secara manual seperti di masa lampau.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b7784a415b11" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>