<?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 Ruwan Yatawara on Medium]]></title>
        <description><![CDATA[Stories by Ruwan Yatawara on Medium]]></description>
        <link>https://medium.com/@ruwanyatawara?source=rss-3c2d4df53380------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*KvjwUjUZT3k3_Dzu.</url>
            <title>Stories by Ruwan Yatawara on Medium</title>
            <link>https://medium.com/@ruwanyatawara?source=rss-3c2d4df53380------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 08:21:20 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@ruwanyatawara/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[The People’s AI Cloud]]></title>
            <link>https://ruwanyatawara.medium.com/the-peoples-ai-cloud-45b434407bc8?source=rss-3c2d4df53380------2</link>
            <guid isPermaLink="false">https://medium.com/p/45b434407bc8</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[distributed-systems]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[future]]></category>
            <dc:creator><![CDATA[Ruwan Yatawara]]></dc:creator>
            <pubDate>Sun, 01 Nov 2026 18:36:38 GMT</pubDate>
            <atom:updated>2026-11-01T18:40:44.035Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vkuDLOSRjH4tdYCQTEcmsQ.jpeg" /></figure><p>I was born in the late ’80s. Which means I grew up in the 2000s downloading torrents, watching files being put together piece by piece from the computers of complete strangers scattered all over the world. No company in the middle, no central server. Everyone who downloaded also uploaded, and the more people joined, the faster it got. <a href="https://en.wikipedia.org/wiki/BitTorrent">BitTorrent</a> had only been around since 2001, and to a teenager, it felt like pure magic!</p><p>Then came <a href="https://bitcoin.org/bitcoin.pdf">Bitcoin</a> in 2009. Same magic, but this time with money attached. I got hooked, read up on crypto and, like a lot of people my age, did a bit of mining of my own. Strangers who would never meet, running a global computer together and getting paid for it. How cool was that?</p><p>So yes, I have been a fan of distributed computing and peer-to-peer networks for most of my life. Which is why two stories from this year really got me thinking.</p><p>In early March 2026, drones <a href="https://news.bloomberglaw.com/us-law-week/drone-strikes-damage-amazon-data-centers-in-the-uae-and-bahrain">struck two Amazon data centers in the UAE</a> and damaged a third in Bahrain. AWS warned that recovery would be prolonged “given the nature of the physical damage.” Then in September, Russian strikes <a href="https://rsf.org/en/ukraine-russia-pounds-data-centres-kyiv-disrupting-media-outlets">knocked several Kyiv data centers offline</a>. All this while the AI industry keeps pouring concrete into campuses that each draw around a gigawatt of power, and the <a href="https://www.iea.org/reports/energy-and-ai/executive-summary">IEA</a> expects data center electricity use to more than double by 2030.</p><p>So the kid who ran torrents in me asked the obvious question. If strangers can share files and run a currency together, why can’t they run AI? Chop the work up into small, encrypted pieces, send them out to idle machines around the world, pay each one for honest work, and put the answers back together at home.</p><p>Sounds simple, right? Well.. it’s a better idea than you’d think, and a lot harder to execute than its fans would like to admit. Let me walk you through both sides.</p><h3>We’ve done this before</h3><p>Don’t get me wrong, the idea itself is old. Back in 1999, <a href="https://setiathome.berkeley.edu/classic.php">SETI@home</a> turned idle home computers into a giant search for alien signals. In March 2020, volunteers simulating COVID-19 proteins pushed <a href="https://foldingathome.org/about">Folding@home</a> past one exaFLOPS, the first computing system of any kind to get there. And today, marketplaces like <a href="https://akash.network/">Akash</a> and <a href="https://io.net/">io.net</a> already rent out spare GPUs peer to peer, while Prime Intellect has trained a <a href="https://primeintellect.ai/blog/intellect-1-release">10-billion-parameter model</a> on GPUs spread across five countries.</p><p>But there’s a catch with the Bitcoin comparison. Bitcoin miners solve a puzzle that is deliberately pointless, picked because anyone can check the answer with a <a href="https://bitcoin.org/bitcoin.pdf">single hash</a>. AI work actually produces something useful, and checking whether a stranger did it honestly is a whole lot harder. Keep that in mind, we’ll come back to it. But first, why bother at all?</p><h3>Power and war</h3><p>Here’s the thing about AI’s appetite for power. It’s huge, yes, and it’s also piled up in a handful of places. The <a href="https://www.iea.org/reports/energy-and-ai/executive-summary">IEA</a> says nearly half of US data center capacity sits in just five regional clusters, and around 20% of planned projects could be delayed by grid constraints. Data centers already eat up <a href="https://www.cso.ie/en/releasesandpublications/ep/p-dcmec/datacentresmeteredelectricityconsumption2024/">22% of Ireland’s metered electricity</a>, up from 5% in 2015. In the eastern US PJM market, capacity prices jumped roughly tenfold, and the market monitor pinned <a href="https://ieefa.org/resources/projected-data-center-growth-spurs-pjm-capacity-prices-factor-10">63% of that rise on data centers</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n8w6ESWSz2vCYEMW9gT8dQ.png" /></figure><p>Every new campus needs new power plants, new transmission lines and years of waiting in a connection queue. Spread the same work across homes, offices and labs, and it runs on connections we already have, moving to wherever there’s power to spare. Even Google has signed <a href="https://www.datacenterdynamics.com/en/news/google-partners-with-im-and-tva-to-expand-use-of-demand-response-at-its-ai-data-centers/">deals to pause machine-learning jobs</a> when the grid is under stress.</p><p>Then there’s the problem of data centers being potential targets for attacks. Back in 2022, Ukraine moved its government data out of the country because, as its digital minister put it, <a href="https://www.theregister.com/2022/11/30/ukraine_cloud_migration/">“Russian missiles can’t destroy the cloud.”</a> A network spread across thousands of locations has no single building worth striking. As an SRE, that’s the one property I’d want above everything else.</p><p>Of course, it comes at a cost. Home GPUs add load to local networks that <a href="https://www.tdworld.com/electrification/article/21278952/right-sizing-residential-transformers-for-evs">electric vehicles are already straining</a>. And spreading out means trading one big target for lots of small weak spots: the Aisuru botnet launched a <a href="https://www.cloudflare.com/threat-intelligence/research/report/aisuru-botnet">record 29.7 Tbps attack</a> from about 500,000 hijacked home devices. Still, given the choice, I’d rather defend a million small things than one big one.</p><h3>The work already fits!</h3><p>Of course, all of this only matters if the work can actually be split up and handed out. So which AI work eats the most compute? Ask most people and they’ll say training. Fair enough, a single frontier training run now draws <a href="https://epoch.ai/blog/power-demands-of-frontier-ai-training">over 100 MW</a>. But at the companies serving AI to billions of people, running the models is what costs the most. <a href="https://arxiv.org/pdf/2111.00364">Meta</a> splits its AI power capacity roughly 10:20:70 across experiments, training and inference. <a href="https://arxiv.org/pdf/2204.05149">Google</a> found about three fifths of its machine-learning energy went to inference. The frontier labs are the exception; OpenAI <a href="https://epoch.ai/data-insights/openai-compute-spend">spends more on research</a> than on serving its users.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n_0gQYD9J7bcCdLxu5t9GQ.png" /></figure><p>Why does this matter? Because inference is exactly the kind of work that breaks down nicely into small, independent jobs. And there’s plenty to run. Open-weight models, the ones anyone can download, are everywhere now; Chinese models alone took <a href="https://api-inference.huggingface.co/blog/huggingface/state-of-os-hf-spring-2026">41% of Hugging Face downloads</a> in the year to March 2026. A 24 GB gaming GPU can serve a compressed 30B-class model all by itself, and the giants, like DeepSeek V4-Pro at <a href="https://simonwillison.net/2026/apr/24/deepseek-v4">about 865 GB</a>, can be split across several machines with tools like <a href="https://github.com/exo-explore/exo">Exo</a>.</p><h3>What makes it hard ?</h3><p>So far so good. Now for the part the fans usually skip.</p><p>Privacy is the first big problem. Splitting data into pieces doesn’t make it private. A node running part of a neural network sees what that part produces, and <a href="https://par.nsf.gov/servlets/purl/10208164">researchers have shown</a> a malicious node can use that to rebuild the original input. There is encryption that lets a machine compute on data it can’t read, but it’s <a href="https://proceedings.mlr.press/v202/ran23b.html">orders of magnitude slower</a>. Trusted hardware enclaves are fast, but NVIDIA only offers <a href="https://docs.nvidia.com/datacenter/cloud-native/confidential-containers/latest/supported-platforms.html">confidential computing</a> on its data center cards, and the GeForce card in your living room isn’t one of them. So for now, sensitive data and anonymous home machines simply don’t mix. Training on public data is a lot easier though; if the data is public and the model is open, there’s very little left to protect.</p><p>Then there’s proof, and this is where that Bitcoin catch comes back to bite. What stops a dishonest node from sending back garbage and asking to be paid anyway? Running every job twice will catch it, but that doubles the cost. Cheaper checks are on the way, like Prime Intellect’s <a href="https://arxiv.org/abs/2501.16007">TOPLOC</a> fingerprints, but this is still the hardest nut to crack.</p><p>And then energy. I would love to tell you this is automatically greener. Unfortunately, it isn’t. Data center hardware does <a href="https://arxiv.org/pdf/1903.01699">more work per watt</a> than many consumer devices, and one <a href="https://arxiv.org/pdf/2609.11940">2026 study</a> found running AI on a phone used about three times the energy per token that a server did. The green argument only holds up when the hardware is already switched on and sitting idle. Bitcoin is the cautionary tale here; an incentive to compute very quickly turns into an incentive to buy hardware just to earn.</p><h3>Where to start ?</h3><p>With all of that in mind, where do you start? It comes down to two questions. How sensitive is the data, and how much do the machines need to talk to each other while they work?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*G-Ppju2JywytMi85jF52sg.png" /></figure><p>That bottom-left corner is already a big market: public data, batch inference on open models, synthetic data and evaluations. These are the jobs that get pushed to the back of the queue on expensive GPU clusters today. The other corners will open up as trusted hardware and faster training methods come along.</p><h3>Who gets paid ?</h3><p>Getting into those other corners needs one more ingredient, and that’s trust. No one in their right mind would send their bank data to some random gaming PC. The way around that is to make trust something providers can earn, and get paid more for.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*p68w2eQ9LY1H01LRPs-_BA.png" /></figure><p>Bits of this already exist. <a href="https://docs.vast.ai/host/datacenter-status">Vast.ai’s</a> top tier asks for an ISO 27001 certificate, a registered business and identity checks. On <a href="https://akash.network/docs/other-resources/security">Akash</a>, independent auditors vouch for providers, and customers pick which auditors they trust. <a href="https://salad.com/faq">Salad</a> pays gamers for their idle PCs and says it has paid out over $7 million so far.</p><p>Take it one step further and you get a box people buy purely to make money from. A certified mini data center sitting in a home or a shop, rented out by the hour to whoever bids the highest, including AI agents, which can now <a href="https://decrypt.co/339752/google-reveals-ai-agent-payments-protocol-coinbase-ethereum-foundation">pay for things on their own</a>. And believe it or not, it has already started. <a href="https://www.datacenterdynamics.com/en/news/electrical-panel-company-span-launches-xfra-distributed-data-center-offering/">SPAN and NVIDIA</a> are piloting GPU servers mounted outside people’s homes, and <a href="https://www.theregister.com/2023/01/12/heata_offers_free_hot_water/">Heata</a> uses servers to heat household water!</p><p>So what could one of these boxes earn? Vast.ai says an RTX 5090 <a href="https://vast.ai/article/how-much-money-can-you-earn-renting-out-your-gpu-on-vast-ai">earns its host</a> about $0.30 to $0.60 an hour. Let’s do some rough math. At $0.45 an hour and 80% utilization, that’s about $260 a month, minus roughly $62 of electricity at $0.15 per kWh. That pays back a $1,999 card in about ten months, if demand holds.</p><p>And that “if” is a big one. H100 rental rates dropped from over $8 an hour to <a href="https://latent.space/p/gpu-bubble">$1 to $2 in 2024</a>, then shot back up almost 40% between <a href="https://newsletter.semianalysis.com/p/the-great-gpu-shortage-rental-capacity">late 2025 and early 2026</a>. Bitcoin’s history is a sobering reminder too. Home miners got squeezed out by industrial miners with cheap power, and those same companies are now signing AI deals worth <a href="https://iren.gcs-web.com/news-releases/news-release-details/iren-secures-97bn-ai-cloud-contract-microsoft">billions</a>. Trying to beat them on price is a losing game. The garage box has other cards to play: sitting close to the people who use it, reusing its heat, and soaking up spare grid capacity that would otherwise go to waste.</p><h3>So, will it work ?</h3><p>Will the world’s idle computers replace the giant AI data center? Probably not. <a href="https://epoch.ai/gradient-updates/how-far-can-decentralized-training-over-the-internet-scale">Epoch AI</a> finds decentralized training is growing fast, but it’s still about 1,000 times smaller than frontier training.</p><p>And that’s okay. Think of it like rooftop solar sitting alongside the power plants. Public, parallel, checkable work, which happens to include the fastest-growing slice of AI demand, can run on a people-powered grid. Private and tightly coupled work stays in trusted facilities until the technology catches up.</p><p>Most of the parts are already on the shelf: scheduling from volunteer computing, payments from blockchains, open-weight models, and AI agents that can pay their own way. What’s missing is someone to put it all together, with a certified tier people can actually trust, honest energy accounting, and payments that don’t feel like a casino.</p><p>The kid who ran torrents in me thinks it’s only a matter of time. If you’re building something like this, or you think I’ve got it all wrong, I’d love to hear from you!</p><h3>Further reading</h3><ul><li><a href="https://arxiv.org/abs/2209.01188">Petals: running large models across volunteer GPUs</a> (ACL 2023)</li><li><a href="https://arxiv.org/abs/2311.08105">DiLoCo: training with 500x less communication</a> (Google DeepMind, 2023)</li><li><a href="https://arxiv.org/abs/2412.01152">INTELLECT-1: a 10B model trained across continents</a> (Prime Intellect, 2024)</li><li><a href="https://arxiv.org/abs/2505.21684">Templar: paying contributors for useful training work</a> (2025)</li><li><a href="https://arxiv.org/abs/2607.13332">Agora: 330 consumer GPUs train an 8.6B model</a> (Pluralis Research, 2026)</li><li><a href="https://www.iea.org/reports/energy-and-ai/executive-summary">IEA: Energy and AI</a> (2025)</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=45b434407bc8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Prepared, Not Surprised: The Working Values of an SRE]]></title>
            <link>https://medium.com/operational-sympathy/prepared-not-surprised-the-working-values-of-an-sre-692304236b99?source=rss-3c2d4df53380------2</link>
            <guid isPermaLink="false">https://medium.com/p/692304236b99</guid>
            <category><![CDATA[sre]]></category>
            <category><![CDATA[site-reliability-engineer]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[engineering]]></category>
            <dc:creator><![CDATA[Ruwan Yatawara]]></dc:creator>
            <pubDate>Mon, 07 Sep 2026 05:31:43 GMT</pubDate>
            <atom:updated>2026-09-07T12:30:26.105Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CTbYjEKDeoaiJQAZLHCa-g.jpeg" /></figure><p>A seasoned Overlander never just turns the key and drives. Whether the journey is a weekend sprint or a deep expedition into the wilderness, the road demands a ritual: inspecting tires, fluids, bearings, and fuel margins. These checks aren’t thrilling, nor do they promise a flawless trip. But they are the exact difference between a breakdown you can wrench yourself out of, and a crisis that strands you completely out of reach.</p><p>Site Reliability Engineering asks the same of us, except the vehicle is a production system carrying customer traffic and the trip never ends. What follows is a two part article that outlines the value set of an SRE actually doing the job (the way I see it): how we show up to a shift, how we hold a problem task, and how we engineer the systems we’re responsible for.</p><h3>Two rivers, two outcomes</h3><p>In August 2026, a section of glacier collapsed on the Nepal–China border, sending a wall of ice, mud, and rock down the Trishuli river corridor. The region did have an early warning system, built up over years and credited with preventing casualties in past monsoon floods. But those sensors had been tuned to detect seasonal, monsoon-driven flooding rather than a sudden glacial outburst, and the upstream monitoring stations were themselves swept away before they could relay anything meaningful. One official later put it plainly: they had sensors, but nobody was watching them closely enough to know the moment they went dark. The flood killed close to a thousand people, with thousands more left missing.</p><p>A year earlier, a strikingly similar hazard played out in the Swiss Alps. The Birch glacier, above the village of Blatten, had been showing precursory signs of instability, with rockfall accelerating from centimetres a year to metres a day. Authorities didn’t wait for a confirmed prediction of the exact moment of collapse. They evacuated the village’s roughly three hundred residents and all livestock while the mountain was still moving. When the glacier finally let go, it destroyed about ninety percent of Blatten. Only one person was unaccounted for, out of a village that could easily have lost hundreds.</p><p>The physical hazard in both cases was comparable. The outcome was not. The difference wasn’t luck, and it wasn’t a better forecast, because nobody predicted either collapse to the minute. The difference was that one system was built, staffed, and rehearsed to act on ambiguous, early, incomplete signals, and the other had a single point of failure sitting exactly where the disaster would strike first.</p><p>That is the whole argument for SRE as a discipline. We will never predict every incident down to the minute. What we control is whether our monitoring, our escalation paths, and our people are positioned to act on the first ambiguous signal, or whether our sensors get swept away right when we need them most.</p><h3>Part one: how we show up</h3><p>The way I see it, these five values are the ones that decide whether a bad day becomes a bad quarter. They cost nothing but discipline, and they are worth putting on a wall as a mantra.</p><h4>1. Prepared, Not Surprised</h4><p>You cannot forecast every failure mode, and chasing that certainty is a trap. What you can do is keep the basics in a known-good state: dashboards that reflect reality, alerts tuned to the failure you’re actually likely to see rather than only the one you saw last time, runbooks that are current, and an on-call engineer who knows where the spare tyre is kept. Blatten wasn’t saved by a precise prediction. It was saved because the people responsible had already decided, in advance, what they would do when the mountain started moving. Preparedness is what turns an unpredictable event into a survivable one.</p><h4>2. Check It Before You Trek It</h4><p>Routine health checks are the least glamorous work we do and the most consequential. A check that runs but that nobody reads is not a check, and a single sensor sitting in the path of the very failure it’s meant to detect is not coverage. Nepal’s stations were swept away because they were the only thing watching, positioned where the disaster would hit first, with no fallback behind them. Layer the visibility: synthetic checks, real user metrics, infrastructure health, and a human being who actually looks. Then verify the checkers, because silence from a monitor is ambiguous until you prove it means all clear.</p><h4>3. See It. Say It. Own It.</h4><p>An SLA breach or a P1 incident is never somebody else’s until they log in. Ownership means you carry it until it is genuinely handed off, acknowledged, and being worked on, not until you’ve sent a message into the void and moved on. A problem task for an incident that you responded to is not someone else’s job to fix or follow up. This is the value that most directly separates a team that drops the ball from one that doesn’t. Nothing sits unattended because everyone assumes someone else had it. If you saw it, it’s yours until it demonstrably belongs to someone else. This is pride we can take in a job properly done.</p><h4>4. Silence Is Not an SLA</h4><p>Leads and stakeholders should never be the last to know. If a problem task is trending toward a repeat occurrence, if an SLA is at risk, if something simply looks wrong before you’re sure why, raise it early and raise it plainly. Nepal’s official alerts reportedly reached people late, in one language only, and without clear guidance on what to actually do. An escalation that arrives too late, or too vague to act on, is barely better than none. Nobody has ever been criticised for flagging a risk that turned out to be nothing.</p><h4>5. Reliability Is a Habit, Not a coincidence</h4><p>Every incident, near-miss, and missed SLA is a rehearsal for the next one. Swiss authorities could evacuate Blatten with confidence because they had spent years building the habit of acting on early, ambiguous signs of instability, shaped by many smaller events before it. Postmortems, RCAs, and pattern-spotting across incidents only earn their time if they change what the team actually does next time. Reliability isn’t something you achieve during an outage. It’s what you were doing in the quiet months beforehand.</p><h3>Part two: how we engineer</h3><p>While the earlier five values govern the shift. These govern the system, and without them the first five slowly become impossible to sustain.</p><h4>6. Blame the System, Not the Engineer</h4><p>Values three and four only survive in a team where speaking up early costs you nothing. The moment an engineer calculates that raising a risk is more dangerous than the risk itself, escalation stops and the sensors go dark, exactly as they did upstream on the Trishuli. Postmortems examine what the system allowed, not who was unlucky enough to be holding the pager. Cultivating this environment is a leader’s responsibility, not an engineer’s, and it is the precondition for every other value on this list. Leadership must create a space where the team feels completely safe coming clean about the exact realities of an incident, including malpractices that went unnoticed, without fearing their jobs are on the line for doing the right thing.</p><h4><strong>7. The User Decides If You’re Up</strong></h4><p>A green dashboard is an opinion; a customer who cannot log in is a fact. Reliability is measured from the outside, at the boundary where the user interacts with the system. This is why Service Level Indicators (SLIs) must reflect actual customer experience rather than internal system comfort. When our metrics and the customer’s reality disagree, the customer is right, and our instrumentation is defective. This truth demands total alignment across the board — from the initial product sale to monitoring the exact use cases the customer cares about. Product development, system solutions, and SRE practices must orbit this single objective. If that alignment doesn’t exist, our primary job is to build it (or push those responsible to do so).</p><h4>8. One Hundred Percent Is the Wrong Target</h4><p>Perfect reliability is neither achievable nor worth what it costs, and pretending otherwise leaves the team with no way to reason about risk. To manage this effectively, we rely on risk registers to make the invisible visible. We must be clear about what risks we are comfortable holding, what non-negotiable risks we must address as a business, and how exactly we plan to overcome them. SLOs and error budgets exist so we can dynamically manage this balance: deciding how much unreliability we can afford, spending it on change and velocity, and knowing when to stop. An SLA without a defined objective behind it is a promise nobody knows how to keep. But to enforce these targets, manual tabs are not enough. Telemetry and monitoring must be incorporated natively into the tools we manage. First we document the risk, then we automate the tracking, and finally, we enforce the budget.</p><h4>9. If You Did It Twice, Automate It</h4><p>Manual, repetitive, no-lasting-value work is toil, and toil is the tax that quietly consumes the capacity you need for the first five values. An engineer spending their week on the same recurring fix has no attention left for the ambiguous signal at the edge of the dashboard. Automating toil isn’t a productivity nicety. It’s how a team buys back the room to be prepared.</p><h4>10. Simple Beats Clever</h4><p>Every additional layer, exception, and clever optimization is another thing that fails at three in the morning in front of someone who didn’t build it. Simplicity is a reliability property, and once lost it is extraordinarily hard to recover. The system you can explain on a whiteboard is the system you can fix under pressure.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-v2cFO86wQC9urpW3zqfkA.jpeg" /></figure><h3>Conclusion</h3><p>Nobody does a pre-trip vehicle check because they expect that specific trip to go wrong. They do it because the cost of the check is minutes, and the cost of skipping it, on the one trip in a hundred where it matters, can be everything. The same logic runs through all ten values here — the discipline of showing up prepared, and the engineering discipline of building systems that don’t depend on someone remembering to be prepared.</p><p>Most days that rigour looks unnecessary: the runbook nobody opens, the error budget nobody spends, the automation that quietly removes a task no one misses. Then one day the ground actually moves. Nepal had sensors. Blatten had a warning just as ambiguous. What Blatten had that Nepal didn’t was a system still standing when the signal came, and people who were still watching for it.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=692304236b99" width="1" height="1" alt=""><hr><p><a href="https://medium.com/operational-sympathy/prepared-not-surprised-the-working-values-of-an-sre-692304236b99">Prepared, Not Surprised: The Working Values of an SRE</a> was originally published in <a href="https://medium.com/operational-sympathy">operational-sympathy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Buying and Modifying a Land Cruiser 79 Series V8 (VDJ79)]]></title>
            <link>https://ruwanyatawara.medium.com/buying-and-modifying-a-land-cruiser-79-series-v8-vdj79-3fdfc06ed19a?source=rss-3c2d4df53380------2</link>
            <guid isPermaLink="false">https://medium.com/p/3fdfc06ed19a</guid>
            <category><![CDATA[adventure]]></category>
            <category><![CDATA[land-cruiser]]></category>
            <category><![CDATA[modification]]></category>
            <category><![CDATA[offroad]]></category>
            <category><![CDATA[suspension]]></category>
            <dc:creator><![CDATA[Ruwan Yatawara]]></dc:creator>
            <pubDate>Sun, 22 Jun 2025 20:12:41 GMT</pubDate>
            <atom:updated>2025-10-10T05:48:05.362Z</atom:updated>
            <content:encoded><![CDATA[<h3>My 79 Series V8 (VDJ79) Build Journal</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*o_B72hXYsoSuXsWfTQjUow.jpeg" /><figcaption>79 In a trail at Ingiriya</figcaption></figure><p>There are cars… and there are CARS! The 70 Series Land Cruiser, in my humble opinion, deserves to be addressed as a CAR — in bold, capital letters! It is truly one of those vehicles whose designers unapologetically chose not to compromise by turning it into a practical, family-friendly, everyday-use automobile. Instead, what you get is an ‘80s-style, sparsely equipped, leaf-sprung truck with the aerodynamics of a brick that rides like a sack of potatoes rolling down a hillside! While this might sound like a vehicle unworthy of a second glance, it has somehow become one of the most coveted vehicle in the Australian outback and the preferred mode of transport for remote area travel by both the United Nations and terrorist organisations alike.</p><p>But why? The answer lies in its unparalleled toughness, simplicity, and reliability. This vehicle is engineered to endure 25 years of relentless abuse in the harshest conditions, all while demanding nothing more than basic maintenance. You don’t buy the 70 Series for luxury; you buy it for its unyielding durability. For enthusiasts like us, it’s a blank canvas — a raw slab of Carrara marble, waiting to be sculpted into something extraordinary.</p><p>Its design is boxy, simple, rugged, and utterly timeless. Twin locking solid axles out of the factory — just as any proper truck should. Under the hood sits a deliberately detuned V8 turbo diesel, so underpowered it feels almost criminal — again to serves one purpose: longevity. Every aspect of this machine is built to last and easy to repair in the bush, a quality that makes it not just a vehicle, but an enduring legend.</p><p>It is this combination of factors that made me lust after a 70 Series since I was a teenager!</p><h3>Inspiration</h3><p>I grew up surrounded by Land Cruisers. Some of my earliest memories include drives to the hill country in a 40 Series long-wheelbase cruiser owned by my dad’s friend. I learned to drive in my dad’s old 60 Series, but his pride and joy — a KZJ77 Land Cruiser he’s owned for nearly 25 years — holds the most sentimental value for me.</p><p>Before the 79, I drove a 2006 Hilux for over seven years. Reliable and versatile as it was, its off-road performance often left a lot to be desired. I always dreamed of owning a 70 Series but it always felt like a distant aspiration given how much I thought it would cost!</p><figure><img alt="A news paper clipping from 2020 advertising the 79 V8 in LK" src="https://cdn-images-1.medium.com/max/1024/1*lqgxhsnDy9y3RJ_OaCK_hg.jpeg" /><figcaption>A news paper clipping from 2019, advertising the 79 series.</figcaption></figure><p>And then, this happened! Sri Lanka being the quirky country it is, has some interesting vehicle tax regulations. Despite the heavy taxes enforced on other vehicle categories, import of dual purpose vehicles (double and single cabs) are permitted at a flat tax regardless of engine capacity. Making the V8 powered 79 series priced remarkably close to a Hilux (difference was just Rs. 2 Million LKR at the time this add was published).</p><p>After seeing it, I couldn’t resist the itch to have one for myself. One fine day — on my birthday, no less — on my way back from work, I decided to swing by the local Toyota dealership and try my luck at driving one. That fateful day, I got behind the wheel of a gray 79 with 33-inch tires and a few basic off-road mods. I was so thrilled, I nearly stalled the car pulling out of the dealership during the test drive!</p><p>The V8, though modest in power, felt just right for the vehicle. The exhaust note under load and the train-like chugging at idle were nothing short of mesmerizing. I was completely smitten — love at first sight! From that moment on, I was determined to get one, no matter what it took. And so, the hunt began!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CH9TPDklODSpI8gUpt_P2g.jpeg" /><figcaption>The 79 I got to test drive at Toyota Lanka back in early 2019</figcaption></figure><h3>The Hunt</h3><p>I initially wanted to get a Merlot Red 76, but with the Lankan vehicle taxes getting a v8, 76 imported would have certainly cost me both my kidneys and all my limbs!</p><p>It had to be a double cab and Toyota at that time had a few unregistered examples including a Merlot red in stock. Despite the discounts offered, I was still unable to afford it without risking financial peril. Therefore I had put on hold my idea of getting a 79, indefinitely. Then, came Covid! All imports were restricted. And the Toyota Lanka vehicle stocks evaporated overnight! I thought my dream of owning a brand new 79 was all but lost.</p><p>Fast forward 1 year, I could feel the itch coming back. I was monitoring the vehicles advertised for sale and a silver 79 caught my eye. It had only 11,000 km on the clock and the gentleman trying to sell, was one of the nicest I have ever met. It had a 2 inch OME Nitrocharger Kit installed and a custom fabric canopy from a reputed off-road modification stable bolted on as well. It was the first time I got to properly push a 79, on road. The owner wanted me to hammer it to see its true grit, and man-oh-man was I not a happy camper! It felt wonderful and the vehicle absorbed all the bumps on the road superbly! I was in love once again. However, coming home. I was heart broken — hear I was with hopes of building my own 79 from ground up. But this one had everything already done to it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*06UcfILpWFboSoQ2eXcJAA.jpeg" /><figcaption>The silver 79 that I never got!</figcaption></figure><p>I wanted a blank canvas and this was not it. So I kept on searching. Few days later — I found what I was looking for!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rY3o3pclOfpB4bSyupJqfg.jpeg" /><figcaption>First Look</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C-lCHqCjd1BHeIyMaa3b-Q.jpeg" /><figcaption>First Drive.</figcaption></figure><p>This particular 79 had just 300km on the clock and had been sitting for a while. My guess is that; then owner, after the onset of Covid met with some financial difficulties as this had a hefty lease attached to it. Most people do not anticipate how uncomfortable/rough this thing rides in its stock form, perhaps this is also one of the reasons that fellow gave this up as his next immediate purchase was a Hilux! In any case; I was able to secure a deal for almost the same sum the vehicle would have cost me at Toyota. It was practically brand new, as not even the first free service at Toyota Lanka had been done (first service — more of an inspection at 1000km). Inside of two weeks, I was able to sell off my Hilux, secure a loan and bring home the 79. It was a dream come true!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*otCbTWoHGjtGtfuwTNBH9Q.jpeg" /><figcaption>after getting it home</figcaption></figure><h3>The Mods!</h3><p>The 79 is by no means a complete car, it is at best a blank canvas on top of which you can create your own masterpiece. It is an awesome platform! Solid axels — front and rear lockers, turbocharged v8 engine, manual transfer case and gear box, strong chasis and a ute body. It’s got all the underpinnings for greatness right out of the factory.</p><p>This was my first attempt at modifying a completely stock vehicle. Even though I was aware of all the things I could do, I was a total noob when it came to figuring out which fabricator/supplier/design I should go with, this meant certain things had to done multiple times as my needs and experience changed over time.</p><h3>Canopy</h3><p>This is in fact the first mod I did on the 79, but as mentioned previously, it is also one of the things that I did without much knowledge on the subject. Eventually I made a decision to remove the first version of it and sell it to a friend without suffering that big of a loss.</p><h4>V1 — Steel Frame + Canvas</h4><p>Design-wise, this canopy fit the vehicle profile perfectly. It was sleek, with no jutting parts, and made of steel, making it both affordable and strong. It was installed by a reputed third-party modification centre recommended to me by a few recent acquaintances. While it fulfilled the need for a contraption that allowed the transportation of goods in the back without risk of them being swept off or getting wet, the fit and finish — particularly the quality of workmanship, such as the welds and powder coating — left much to be desired. I kept it on the truck for over a year, but at the first sign of rust, I had to remove it, as I couldn’t stand the sight of it any longer.</p><p>On the plus side, I made so many connections while getting this built. Saman, being one of them; a true master when it comes to stitching canopies!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XTAu9kaZTr5laX5phb9hdQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FGVuSYkDsep6xoei_iwJ7A.jpeg" /></figure><h4>V2 — Stainless Steel Frame + Marine Hooding</h4><p>With all the lessons learned from the failures of the first iteration, I set out to design the second one. I briefly considered installing a stainless steel canopy from RSI in South Africa, as their designs looked fantastic. However, acquiring one at an economically viable price point proved challenging, especially amidst the post-COVID import restrictions we faced in Sri Lanka — given the economic crisis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/999/0*ArUVx_mJUgDjVzUU.jpg" /><figcaption>Source — RSI SmartCap for the 79</figcaption></figure><p>The biggest drawback of the RSI design was that it couldn’t be used to shelter people — something essential if we planned to use the vehicle as a people carrier inside national parks. I had fond memories from my school days, riding in the back of our short-wheelbase Nissan Patrol, and I wanted my son to one day experience that same sense of adventure. So, it had to be a fabric top once again. But this time, on the advice and recommendation of a friend, I decided to build it using stainless steel and aluminium — no rust this time around.</p><p>That is how I got introduced to Harald — a true master of his trade. I have never seen anyone work so expertly with steel ever in my life. He is truly an artist. I spent days at his shed in Kadana, and learned so much from him.</p><p>This design had a few special features.</p><ul><li>The bed rails were encased in a 4mm thick folded and welded stainless steel sheet. The outer edge, where the fasteners are attached, was deliberately angled inward (tucked in) to provide a better seal for the fabric, effectively preventing splattered mud, water, and dust from entering.</li><li>The stainless steel structure could be raised by 8 inches when needed and was secured to the bed rails using six sturdy screws. This design provided improved headroom and easier access for passengers sitting in the back, creating a safari-like experience.</li><li>200 series 3rd row seats on a custom subframe, that fold away when not in use</li><li>Powder coated, folded and welded — aluminium roof with tie down hooks and max-trax mounts.</li><li>Sides that open up, mounts for lights and riverse camera</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*SnCi6_bEikeKzIBaUsjT4g.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8HSQj9Wt_Kwf4YNINv7Xjw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HIhT5_WBVkzWdVE9r_I24w.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bkQ6t8Y_4eUze2enRPnrzw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fdlvDP5DvVzWzijN1fZfFQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1YN4syx5EImMaoWCo6sDfg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rMPso7kAOPIJMd374tAvNg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Sar0UM7arZ2gB4NQV-O9Aw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*w6NE70_EhJh629uXQzgQgA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ICpFH1irkszXqrnULJtDsA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C4V_foDJChX4ULkiEMnEEw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jQL9IpOkIPEuwDof0Jx09Q.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*B4Ih3Rh8wvaPM1DZX7TDVA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*A4Gqe_C7tS0AIctBDuPzzQ.jpeg" /></figure><p>The finished structure looked stellar. It was a labour of love. Quality was exquisite. Everything came out great.</p><p>For the canvas I wanted to go with Marine boat hooding, instead of fragile canvas that would fray with weather. I sourced Aquatica Vinyl Marine Hooding from Paskal a company in Austrailia with the help of a close friend— <a href="https://www.paskal.com.au/aquatica.html">https://www.paskal.com.au/aquatica.html</a></p><p>Ended up paying a premium for this along with taxes, but once Saman did his magic and stitched it all up, I was very pleased with the end result.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yRsthXMaID6uWsE6PoPE5A.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BDjfH2BkrO7VhDyVfrQb4A.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wboU-HwXGxHVr9d_58M7sw.jpeg" /><figcaption>Finished Product</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*72NIJl2pgXqDuYZ5y8rw_A.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*BlE7_kZ8pHMPlQqF3n5wCw.jpeg" /><figcaption>View as seen from above with MaxTrax mounted</figcaption></figure><p>I might some day go for a hard canopy, but I feel that point of time is still many years a away. Until such time, Harold’s masterpiece should stand the test of time.</p><h3>Suspension</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ct_M77Apnv3Vm9KkLBqIdw.jpeg" /><figcaption>us loading up the 79 with all the accessories from Motorcade. I was a happy man that day!</figcaption></figure><p>I knew I wanted to get a 2 inch lift in the truck even before I got it. I was entertaining the idea of going with internal bypass remote reservoir shocks but was hesitant to go with a brand like King Shocks, even though they are the gold standard when it comes to custom adjustable suspension components — purely due to the lack of proper support and presence of a local agent in Lanka. So it had to be ARB BP-51s.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FYH8N62iLPLI%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DYH8N62iLPLI&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FYH8N62iLPLI%2Fhqdefault.jpg&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/a2ba3d37931c44f98b725d25cb140f9a/href">https://medium.com/media/a2ba3d37931c44f98b725d25cb140f9a/href</a></iframe><p>Not only was it individually adjustable in terms of rebound and compression — it was rebuildable meaning when the seals and hydraulics fail some day I can ideally send this back to ARB and get them rebuilt — ready for another round! BP-51s have been built around the concept of high speed valving meaning the damping force is dynamic in response to rapid changes in suspension movement.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1006/1*IflGkHVTgginxDQPK64Suw.png" /><figcaption>picture credits — MotorTrend — <a href="https://www.motortrend.com/news/ome-bp-51-bolt-on-bypass-shocks-for-insane-performance/">https://www.motortrend.com/news/ome-bp-51-bolt-on-bypass-shocks-for-insane-performance/</a></figcaption></figure><p>Long story short — the faster you go the softer the ride gets. Its all thanks to the internal bypass system these shocks have — see picture above.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/597/1*d89j3_Iv_kyanNOGNn0yLw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/597/1*z51J8ksPWMbEx78KT3cStg.jpeg" /><figcaption>Ranjith, a seasoned mechanic at our workshop installing BP-51s amidst fears of Covid, as it was just after the outbreak — hence the mask.</figcaption></figure><p>Installing BP-51s was a brand new experience for us. We had never done anything like it despite the countless suspension upgrades we have done over the years. So we decided to do everything, by the book leaving no space for guesswork. The instructions were spot on and the mounting system for the remote reservoirs was very well thought out! It was a few days of meticulous fun installing these, but the best part was by all means the testing and adjusting that followed right after.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mtZpGJ5FSVIWjp2G7MLHsA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/597/1*qbsocPwj5Gu8u3WIh3LYsQ.jpeg" /><figcaption>Rear Reservoir Mount on Chasis cross member</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/597/1*AP9UCym6tX4auWgzrDvqgg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/597/1*XppxirytTNrEj3duHFB1gw.jpeg" /><figcaption>Front Reservoir mount and Rebound / Compression adjustment</figcaption></figure><p>After much tweaking — and spirited driving over rough roads, I was able to come up with a setting I liked. Note that the suspension upgrade was done after installation of the winch and canopy — therefore, the final weight and weight distribution of the vehicle was somewhat understood. Honestly, after the initial setup — I am yet to revisit make an adjustment and its been more than 3 years!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*O5IbfsUWMqrVmV6swBI-Uw.jpeg" /><figcaption>Stock suspension that was removed.</figcaption></figure><p>Above is a picture of all the standard suspension bits removed from the truck. The suspension was on for just shy of 15,000 Km — so i was able to sell off the lot for a fair price pretty easily.</p><p>While everyone installing a lift kit would surely consider a steering damper, one piece of kit often overlooked is the castor correction bushes on these solid axle based vehicles.</p><p>When a truck is lifted, the caster angle — the angle of the steering axis relative to the vertical — can change, potentially affecting steering stability and alignment. This is the reason why most lifted trucks drive like boats with no steering feel. Installing caster correction bushes helps restore the caster angle to its optimal position, ensuring proper steering geometry. One thing you would notice after installing the lift and doing a alignment is that — the castor angle will be off by more than 3 degrees. The ARB castor correction bushes corrects the castor by 2 degrees. While you would still have castor angle thats a bit off from factory — it will still be positive meaning there would not be any involuntary turning of the wheel when you try to trying to lean in to a corner.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*7PMH7PZkxJ_csJ-P" /><figcaption>Source — <a href="https://www.asli4x4.com/products/offset-bush-for-toyota-landcruiser-78-79-80-series">https://www.asli4x4.com/products/offset-bush-for-toyota-landcruiser-78-79-80-series</a></figcaption></figure><p>That said, the castor correction bushes are a temporary fix — There have been examples of these bushes affecting diff welds over a time when used in extreme rough terrain over long periods of time — the most appropriate way to solve the castor problem is to install new radius arms that twist the diff in to the correct angle. But for a 2 inch lift this is over-kill, and not a must. Superior-Engineering does sell these radius arms. Maybe some day I may consider installing these as well — but so far the need has not arisen.</p><p>Having done the BP-51 installation on my 79, we built up expertise and have since installed this kit on quite a few 70 series Cruisers and Nissan Patrols. On one of these occasions, while installing the kit on a single cab 79, we came across a rear sway bar that had been added as an optional extra. During the initial test drive, I immediately noticed how much stiffer the vehicle felt when cornering. While this setup could potentially limit articulation off-road, I primarily use mine for long-distance travel — so any added stability was a welcome improvement.</p><p>I was lucky enough to score a Superior Engineering rear sway bar for free from a generous friend who’s into hardcore off-roading (he also owns a blue 79 with a rear coil conversion). I sourced the required fitment kit from ARB, and once everything was installed, the difference in cornering performance was clear and worthwhile.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MBfy7UbqF3DXNJIHGYnBTg.png" /><figcaption>ARB Sway Bar Mounts for the Chassis</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*H_MOFqfp1J7-2zQueFkoqQ.png" /><figcaption>installing the Superior Engineering Rear Sway Bar</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JXYuunKSOcu_AISN6mTfaA.png" /><figcaption>As seen from the rear after mounting.</figcaption></figure><h3>Snorkel</h3><p>All 70 Series Land Cruisers come with a raised air intake that many mistake for a proper snorkel. It’s not — it’s made up of multiple pieces and isn’t fully sealed. If you attempt a deep enough water crossing, there’s a real risk of water entering the intake. While most people opt for the Safari Snorkel from ARB, I personally preferred the one offered by TJM — it had a cleaner design and felt better suited for the vehicles looks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VDFvqre5cW18J_U2_TsxeA.jpeg" /></figure><h3>Roof-rack</h3><p>While I dont have much use for one, there were occasions where this came in handy. But mostly I cannot count the times this has saved the roof of my 79 from getting dinged. While this does compromise my ability to enter certain low ceilinged garages. I have zero regrets in getting this rack given how good it looks and the insurance it provides. That said, the ARB Base Rack I am running is definitely on the expensive side, but the modular way you can bolt stuff on to this makes it a worthy accessory for additional things you want to bolt on to it, like my front light bar.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yv70Q-66eHNKRHKwrUDKQQ.jpeg" /><figcaption>79 with the ARB Base Rack + Monkeys</figcaption></figure><h3>Winch and Recovery Point</h3><p>While a fully loaded 79 might not tip the scales beyond 5 tons, I still wanted a winch powerful enough to recover it effortlessly if the need ever arose. I opted for a steel rope over synthetic — not just for durability, but for its ability to withstand the elements with minimal maintenance. The 12,000 pound Zeon series from Warn felt like the obvious choice, even with the more budget-friendly VR series available.</p><p>After fitting the initial version of the canopy, the very next modification I prioritised was the winch — this was partly due to a close call I had while visiting our estate in Elkaduwa. I was navigating a narrow track that skirted the edge of a steep drop — part of the estate tea plot — on a particularly wet and rainy day. The ground gave way beneath one of the tires, causing the entire vehicle to tip dangerously to the side. What saved it from going over was the differential catching on the ledge, stopping the fall just inches from disaster. We eventually had to borrow a boom truck to recover the vehicle. Had it been equipped with a winch, a controlled self-recovery would have been possible.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/591/1*QIgK7XYF-IStUyFugFCgnA.jpeg" /><figcaption>79 Nearly toppled off down a mountain face</figcaption></figure><p>After getting installed — so far, the winch has recovered several vehicles — though never my own. Most recently, it pulled out a Patrol significantly heavier than the 79, buried to its diffs in sand. I also made sure not to compromise on rated recovery points, knowing how critical they are when things get serious. The ARB recovery point features a unique mounting design that ensures its integrity and secure attachment to the chassis, regardless of the direction of the pull — whether straight, upward, or from the side.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iaMwrmuQ0DGd3PkvRAwyUA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2-ZzoD_RXoasUTmgxzaUCA.png" /><figcaption>installing the winch + skid plate + radiator guard + recovery point</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n-5smFdYVloiV_VgmtmrzA.jpeg" /><figcaption>Skid Plate, Winch and Radiator Guard from Armadillo</figcaption></figure><h3>Rear Bar-work</h3><p>I had the rear barwork done at the same place that fabricated the front skid plate and winch mount. While it serves its purpose, I feel the design and mounting could have been executed better. I’m currently considering a redesign, drawing inspiration from Australian-made solutions like the Kaymar bar and the Cruiser Company bar.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*OrDtn08H9-0zkzJ9amQS1g.png" /><figcaption>Rear bar mounted for the first time while Caopy V1 was in place</figcaption></figure><p><strong>Update : 03/10/2025</strong></p><p>I finally got rid of the rear bar from the previous fabricator — and once it was off, my doubts about its integrity and workmanship were confirmed. The build quality was shockingly poor, and frankly, I regret keeping it on the vehicle for as long as I did. Thankfully, I never attempted a snatch recovery from the rear with it, because it would have failed miserably. The crude finish speaks for itself. A hard-earned reminder for me: never take things at face value, and always get hands-on with the finer details.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ajiicqi7LRTF-qZVgCQ9WA.jpeg" /><figcaption>The disappointingly crude old rear bar!</figcaption></figure><p>I collaborated with <strong>Asela Gunasinghe</strong> — a renowned fabricator in Sri Lanka’s 4x4 community — to design a new, simplified bar inspired by the X-Bar for the 79. No compromises were made on materials, craftsmanship, or effort. To enhance functionality, we incorporated <strong>extended side flares</strong> that provide additional protection against side impacts.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/750/0*QSW81VAhxP_RWE-K" /><figcaption>X-BAR — <a href="https://x-bar.com.au/toyota/">https://x-bar.com.au/toyota/</a></figcaption></figure><p>The final product turned out top-notch, delivering everything I envisioned in a rear bar. It features <strong>12 mm steel side recovery points</strong> for uncompromising strength and <strong>10 mm steel mounting plates</strong> to ensure a rock-solid fitment.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8T8TqCcFKUMaOlM3QJwfsg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jGkRoLqNJ2xqBVCjgJnlhA.jpeg" /><figcaption>New rear bar!</figcaption></figure><p>The center recovery point is still being fabricated. It will be going on sometime next week. A very clean, durable and practical design, overall!</p><h3>Lights</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xQ7y6VOOI9tAKrE6NthrUw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*c3rHcj3csBIb9ie7I6w4bQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Lz2Pyq4i5eiR4Ssf6RM6-g.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*idwHXFMhCFWgMFiPyayejg.jpeg" /></figure><p>The 79 originally came from the factory with yellow halogen lights. While these are easy on the eyes and perform decently in rain, they’re not ideal for cutting through pitch-black conditions. To improve nighttime visibility, I opted for Stedi’s Pro Series gear, which offers excellent performance at a reasonable price point. I installed Type X Pro spotlights up front, complemented by a roof rack-mounted 41-inch ST3301 LED light bar with amber lenses for situations where yellow lighting is preferred. Upgrading the factory headlights with an H4 LED conversion also made a noticeable difference.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TcNszcu46WnEqXac230ycQ.jpeg" /><figcaption>Getting the ST3301 PRO 41 INCH light bar fitted, mounted to the front of the ARB Base Rack at Harold’s Worksop.</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Yi4MO0Ve2i-hgHySlI8Tow.jpeg" /><figcaption>STEDI Type X Pros in action</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*55YCpdjVaklDe4M2JbZmyg.png" /><figcaption>STEDI Lights lighting up the lagoon for a midnight dip at Uchchimunai Point, Kalpitiya</figcaption></figure><h3>List of Mods</h3><ul><li>OME BP51 Remote Reservoir fully adjustable Shock Absorbers, Front and Rear</li><li>OME 350Kg constant load rear leaf springs + Greasable Shackles</li><li>OME 50Kg constant load front coil springs</li><li>OME Castor Correction Bushes</li><li>OME Steering Damper</li><li>OME Sway Bar Spacers and Brake Proportioning Kit</li><li>Superior Engineering Rear Sway Bar</li><li>Armadilo Front Skid Plate and Radiator Guard</li><li>Hayman Reese X-BAR inspired custom Rear Bumper</li><li>ARB Recovery Point</li><li>Warn Zeon 12,000 pound winch with Steel Cable</li><li>TJM Snorkel</li><li>MaxLiner tub liner</li><li>Height-adjustable rear canopy using stainless steel framing with Aquatica marine boat hooding. The roof is shaped and welded aluminium checker plate, with mounts for Maxtrax and tie-down points. Fitted with 200 Series flip-up third-row seats in the tub.</li><li>Stedi rear flood lights</li><li>Stedi Type X Pros in the front</li><li>Stedi ST3301 PRO 41 INCH light bar</li><li>Stedi H4 LED head lights</li><li>Kenwood 8 inch head unit with wireless Apple Car Play and Android Auto</li><li>JBL GTO509C x 2 + Club 6520 x 2 Speakers</li><li>Apicool Console Fridge with customised arm rest</li><li>Wetseat Waterproof Seat Covers</li><li>MaxLiner Moulded 3D Carpets</li><li>Cabin Filter from Carbon Filters Austrailia</li><li>3M Window Tint all around</li></ul><h3>Conclusion</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*D4Y4OZ6pAVRisG-iKUcLPg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VTgQv4HzPeu38BQTYWASiw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XqrsyXmsDlkY5n5RguRCIw.jpeg" /></figure><p>After over five years of ownership and millions invested in modifications, the 79 has become far more than just a vehicle to me. It’s a passion — bordering on obsession. It’s taken me to places I once only dreamed of, safely and reliably. No other vehicle has ever had this hold on me. I always find myself coming back to it, and no matter how many times I drive it, I’m always looking forward to the next journey.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3fdfc06ed19a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Don’t postpone your dreams!]]></title>
            <link>https://medium.com/swlh/dont-postpone-your-dreams-9e7715a3c5e3?source=rss-3c2d4df53380------2</link>
            <guid isPermaLink="false">https://medium.com/p/9e7715a3c5e3</guid>
            <dc:creator><![CDATA[Ruwan Yatawara]]></dc:creator>
            <pubDate>Thu, 27 May 2021 04:54:55 GMT</pubDate>
            <atom:updated>2024-12-09T06:48:22.704Z</atom:updated>
            <content:encoded><![CDATA[<h3>Don’t Postpone your dreams!</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1_n7iDdbU_8pbHy2EbMQwA.jpeg" /><figcaption>source : <a href="https://unsplash.com/photos/7nsqPSnYCoY">https://unsplash.com/photos/7nsqPSnYCoY</a></figcaption></figure><p>Dreams, We all have them. If you were to ask 100 people about their aspirations, what they want to chase after, if money and time were no objects, you are bound to get some very diverse answers. Don’t get me wrong I am not talking about careers here. If you really think about it, we do what we do, to earn a living, make ends meet. Sure, our jobs play a huge part in our identity, but there is a lot more to us than our jobs. Our aspiration is the lives we want to build for ourselves and our loved ones and our careers help finance that.</p><p>Not everything can be bought, yes. Some things need to be experienced, felt, discovered or even built. We all get our kicks in very different ways, thats how the world works. But the bottom line is, in this day and age all of these require a pretty penny. Sometimes more than we are ready to shell out, given our list of current commitments.</p><p>I watched this short video about a man in Japan sometime ago; A middle aged man who lives alone in a small apartment — not atypical in Japan, yes. But what is most atypical about his predicament is that more than half of his living space is occupied by a state of the art sound system that is so advanced, it is sensitive to the most minute of fluctuations in the city’s power grid. It is so vast that he has had to dispose of his bed. He sleeps on a couch, that requires moving in order to open his refrigerator. The setup has been so expensive and beyond his means, he had not been able to buy a single article of clothing for himself over the past 15 years!</p><p>What a miserable life, you think… or is it? Despite all the sacrifices he has had to make, He is happy. He sits on his couch, puts on a track and immediately transcends! In his eyes, he is living the good life. He has grasped what makes him happy, and has come to terms with it.</p><p>While I certainly don’t advocate putting your life on hold and living beyond your means, like this Gentlemen, here. There is a certain kind of satisfaction in doing that. We all envy him, for he has had the guts to chase after his dream no matter the consequence. It’s a truly rare thing, really. In a time and age where everyone is talking about investments, risk management and career growth.. here is a man shelving out his life’s saving to live his dream. We all dream of doing that. In the deepest corners of our hearts, we all know what we want. It can be an exotic vacation to a place you always wanted to go see. It can be your dream Car, Motorcycle or that flashy new Guitar! We all wish we had the courage like this guy here, to pull the trigger and just go for it.</p><p>We are all taught from a young age to save our money, keep it for later, invest, put it in a bank, let it grow… etc. etc. It’s about saving what you can today, so you have enough money to get around when you are older, a form of delayed satisfaction, if you will. But is it really worth it? Will you be able to enjoy that fast car when you are 75 and sickly? Can you climb that mountain when you are past 60? What can you do with 100 acre ranch and four story house when you cant even limp from one end of the room to the other; without gasping for breath!</p><p>In reality, time is a fleeting mistress, it waits for no man, or woman! You get old, weak and powerless. With time, your psyche matures, but your body gets weaker. You can slow it down, but this is the grim reality.</p><p>So enjoy what you want to enjoy, when the time is right. There is no moment like the present, your time is limited. People get old, grow apart, times change but life goes on. So strive to achieve the things that matter to you, I mean the things that truly matter, not the whimsical fantasies that come over you watching other people on social media. Don’t put it off till next year, or the one after, cos the risk of you never getting to enjoy your life and live your dreams run high. Maybe take baby steps if you have to, but make sure you get there. Someday!</p><p>You owe it to yourself!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9e7715a3c5e3" width="1" height="1" alt=""><hr><p><a href="https://medium.com/swlh/dont-postpone-your-dreams-9e7715a3c5e3">Don’t postpone your dreams!</a> was originally published in <a href="https://medium.com/swlh">The Startup</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[MySQL Tips and Tricks for Better performance]]></title>
            <link>https://medium.com/swlh/mysql-tips-and-tricks-for-better-performance-15fab69ea818?source=rss-3c2d4df53380------2</link>
            <guid isPermaLink="false">https://medium.com/p/15fab69ea818</guid>
            <category><![CDATA[performance]]></category>
            <category><![CDATA[tuning]]></category>
            <category><![CDATA[mysql]]></category>
            <category><![CDATA[database]]></category>
            <category><![CDATA[sql]]></category>
            <dc:creator><![CDATA[Ruwan Yatawara]]></dc:creator>
            <pubDate>Tue, 16 Jun 2020 02:49:55 GMT</pubDate>
            <atom:updated>2020-06-16T19:41:22.048Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3K01U_X9_oywa4pjJ5ouMg.jpeg" /></figure><p>NoSQL databases maybe the cool kids on the block, but the days of Relation Data Bases are far from over. With proper tweaking, the performance of these traditional databases can be very well optimised. Please note that the tips discussed here are applicable for MySQL databases and it’s many variants, although the broader concepts may apply for other systems as well.</p><h3><strong>Optimising for Writes</strong></h3><p>In the recent past, we had to adjust a customer’s IoT Server deployment for high number of writes. Given the large device count and high communication frequency, performance against a non-optimised MySQL DB, proved to be abysmal. To overcome this I ended up adjusting the following few parameters.</p><h4><strong>innodb_buffer_pool_size</strong></h4><p>InnoDB caches tables and indexes, whenever data is accessed in an area of memory called the Buffer Pool. When a change is done to a row, what’s changed first is the page pertaining to this data in the buffer pool. This page will later be flushed and written back to the disk. Note that in MySQL, for an update to be marked as successful, the page need not necessarily be written to the disk, rather just the transaction log (redo-log according to Oracle terminology) will be updated. This is adequate for the transaction to stay safe. In the event of a crash or failure all changes present in the re-do log will be applied upon recovery.</p><p>To be fair, both reads and writes benefit from having an increased <strong>innodb_buffer_pool_size, </strong>also both master and slave nodes (if present). Increase in the buffer pool will increase the likelihood of a given page being present in RAM, when a query is run, leading to better performance. It is usually safe to set this parameter to a value that is about 70% — 80% of the total memory available in the MySQL server instance, equating to about 1/4 the size of your database.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/561/0*GA9S1jgPr22JTpbR.jpg" /><figcaption>source : <a href="https://www.percona.com/blog/2018/06/19/chunk-change-innodb-buffer-pool-resizing/">https://www.percona.com/blog/2018/06/19/chunk-change-innodb-buffer-pool-resizing/</a></figcaption></figure><p>It is to be kept in mind that <strong>innodb_buffer_pool_size </strong>should be a multiple of <strong>innodb_buffer_pool_chunk_size </strong>(If not this will be adjusted automatically by the InnoDB engine, during runtime)<strong>. </strong>This is because the buffer pool consists of individual chunks, of the specified size (by default this is set to 128MB, and I do not recommend you meddle with the default value. If you need to change this refer <a href="https://www.percona.com/blog/2018/06/19/chunk-change-innodb-buffer-pool-resizing/">this article</a>), and chunks will be added or removed depending on the expansion or constriction of the pool.</p><p>Information on how to change buffer pool size available <a href="https://dev.mysql.com/doc/refman/5.7/en/innodb-buffer-pool-resize.html">here</a>.</p><h4>innodb_flush_log_at_trx_commit</h4><p>This property controls the balance between ACID compliance and performance. As we discussed in the earlier topic, for InnoDB engine to know a transaction actually took place, there needs to be a transaction log entry present. MySQL does something called write-ahead-logging to maintain data consistency. This will result in transaction log being updated prior to commit, for every transaction. While this is necessary for transaction completion, it can also be an overhead. Also to be considered is the fact that, the log entry needs to be written to durable storage or the transaction will be lost entirely, in the event of a system crash or volatile storage failure. Therefore, InnoDB engine provides the topic under discussion as a programable parameter to optimise performance.</p><p>The value for <strong>innodb_flush_log_at_trx_commit </strong>can be either 0, 1 or 2. By default this is at 1, meaning transactions logs are written to disk at each transaction commit. This ensure full ACID compliance, and no data loss in the event of a crash. Setting this property as 0 will make sure transaction logs are written and flushed to durable storage every second. In a high write scenario, where a reasonable consideration needs to be given for consistency, I believe setting this value to 2 is ideal. Doing so, will make sure that transaction logs are written at each transaction commit. However, flushing to durable storage will only happen once every second.</p><p>This value can be set in run time as follows.</p><blockquote>SET GLOBAL innodb_flush_log_at_trx_commit=2</blockquote><p>More information on the property available <a href="https://dev.mysql.com/doc/refman/5.7/en/innodb-parameters.html#sysvar_innodb_flush_log_at_trx_commit">here</a>.</p><h3>Optimising for Reads</h3><p>When it comes to optimising for reads, once we cross off the usual culprits like:</p><ul><li>Indexing columns</li><li>Restructuring database schemas</li><li>Optimising querries (Minimising use of ‘like’s, full-text searches, wildcards etc.)</li><li>Implementing Query Caching</li></ul><p>There comes a point where some focus need to be given to adjusting the transaction isolation level, to eke out the maximum performance possible, from the DB engine, without compromising too much on reliability, consistency and repeatability of the results when multiple transactions are making changes at the same time.</p><h4>READ COMMITTED</h4><p>By default the InnoDB engine will be using <a href="https://dev.mysql.com/doc/refman/5.7/en/innodb-transaction-isolation-levels.html#isolevel_repeatable-read">REPEATABLE READ</a> level for transaction isolation. This will ensure a consistent result set across multiple reads within the same transaction. This is made possible by a snapshot which is established upon the first read. If you can afford to relax this consistency a little bit, we can use <a href="https://dev.mysql.com/doc/refman/5.7/en/innodb-transaction-isolation-levels.html#isolevel_read-committed">READ COMMITTED</a> isolation, which will eliminate the dependency on this original snapshot resulting in much higher read throughputs.</p><p>This level of isolation, avoids the phenomenon of a dirty read as only changes that are committed will be taken in to account. However, given that each SELECT uses its own snapshot, the same query executing multiple times within the same transaction will result in a different sets of results (given that other transactions are performing changes to the same set of data, simultaneously). This phenomenon is called <strong>non-repeatable</strong> read.</p><p>For more information on this please refer this <a href="https://dev.mysql.com/doc/refman/5.7/en/innodb-transaction-isolation-levels.html">link</a>.</p><p>On a final note, there are many points of optimisation in InnoDB, what’s discussed here is just a handful. However, all these performance tweaks are highly subjective and scenario dependent. There is always a compromise somewhere, so do your research and ample amounts of testing before settling on a set of properties that works for you.</p><p>Further reading :</p><ul><li><a href="https://dev.mysql.com/doc/refman/5.7/en/innodb-parameters.html">MySQL :: MySQL 5.7 Reference Manual :: 14.15 InnoDB Startup Options and System Variables</a></li><li><a href="https://www.percona.com/blog/2011/02/03/how-innodb-handles-redo-logging/">How InnoDB handles REDO logging - Percona Database Performance Blog</a></li><li><a href="https://www.xaprb.com/blog/2011/01/29/how-innodb-performs-a-checkpoint/">How InnoDB Performs a Checkpoint</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=15fab69ea818" width="1" height="1" alt=""><hr><p><a href="https://medium.com/swlh/mysql-tips-and-tricks-for-better-performance-15fab69ea818">MySQL Tips and Tricks for Better performance</a> was originally published in <a href="https://medium.com/swlh">The Startup</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Rented Life]]></title>
            <link>https://ruwanyatawara.medium.com/the-rented-life-22748d3a188?source=rss-3c2d4df53380------2</link>
            <guid isPermaLink="false">https://medium.com/p/22748d3a188</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[life]]></category>
            <category><![CDATA[inspiration]]></category>
            <category><![CDATA[thoughts]]></category>
            <category><![CDATA[subscription]]></category>
            <dc:creator><![CDATA[Ruwan Yatawara]]></dc:creator>
            <pubDate>Fri, 21 Jun 2019 16:51:57 GMT</pubDate>
            <atom:updated>2020-06-18T10:14:25.721Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XpExyeZ5xv25adnSqFLi1g.jpeg" /></figure><p>Imagine a world where you don’t own anything. Everything you use, your house, car, laptop, mobile phone, and even your pets are rented! This ain’t some sci-fi story from an episode of Black Mirror, this is the new “reality”. The inevitable future.</p><p>Major corporations, worldwide are experiencing slowdowns in business growth. Consumption of their products has plateaued. They are simply, not selling “stuff” at the same rate as they used to. Why? Partly, this is due to us a human race, having reached the brink of what is possible to humanly consume. But mostly, this has to do with the fact that the targeted consumers of these products have already bought their fill of necessities and they won&#39;t consider buying more until what they already have “expires”!</p><p>So how do businesses work around this? They make more innovative, expensive, limited edition, exclusive stuff. They deliberately legacy-fi and expire tech. Look at the new <a href="https://www.amazon.com/Apple-iPhone-256GB-Silver-Unlocked/dp/B07T18NBMX">iPhones</a> for example, they are the most expensive mass-produced personal communicating devices in all of human history to date, but they don’t happen to have a headphone jack, just so Apple can pave the way for Air-Pods.</p><p>But even this won&#39;t cut it, everything has its limits and so does innovation, no matter how many billions they pour into R&amp;D, major corporations are finding it difficult to entice buyers to upgrade, year on year.</p><p>The solution: Subscriptions!!!</p><p>Although the concept of renting a house is not foreign, due to the high costs of real estate, a car is something we usually take for granted. Due to depreciation, anyone with an average income in the west can afford a decent functional vehicle; although this may not be the case elsewhere in the world. That being said, in the not so distant future, Automobile manufacturers like Jaguar, Mercedes, and Volvo, are considering offering their products as a service, pay as you go packages if you will with an annual tiered subscription cost. As a result of this, you will never really own a car, customizations are merely surface depth charades, offered as mass market packages or easy fit add-ons. On the upside though, you will never have to bother about servicing or maintaining your vehicle ever again!</p><p>This is part of a larger trend being adopted by major hardware manufacturers, all over the world. It is all about increasing a customer LTV, or life time value. In a regular transaction, ownership of an asset is transferred. In the subscription model, a customer will never own the asset in question, it is just a commodity loaned by the manufacturer, for a monthly fee. For all intents and purposes, it never will be an asset. Today, this is happening for cars, imagine this happening for all your electronics, household appliances, clothes, etc. etc…. You will never own anything! no assets! Not even the money you earn will be yours. Remember in the age of digital, all currency will be virtual, whatever you think own, will be a reference to a record in some database, or electronic wallet somewhere in the universe that is the virtual world.</p><p>So yes, it will be a rented life. A virtual life, governed partly or entirely by trillion/billion dollar businesses! Will it be a good one? Of course, it will be. It is the future. But the question remains to be answered… are we living life the way we want? or are we just living a lie? are we just pawns living the life someone imagined for us? Is this all a giant computer simulation?</p><p>I think your guess is as good as mine.</p><ul><li><a href="https://www.theverge.com/2018/6/15/17464548/jaguar-land-rover-car-subscription-service-carpe-uk">Jaguar Land Rover&#39;s new subscription service is UK-only for now</a></li><li><a href="https://techcrunch.com/2019/06/03/mercedes-benz-is-expanding-its-luxury-subscription-service/">Mercedes-Benz is expanding its luxury subscription service</a></li><li><a href="https://www.repairerdrivennews.com/2018/06/18/mercedes-jaguar-announce-car-subscriptions-volvo-thinks-model-will-be-be-50-of-cars-next-decade/">Mercedes, Jaguar announce car subscriptions; Volvo thinks model will be be 50% of cars next decade - Repairer Driven News</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=22748d3a188" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to fix macOS Sierra WSO2 Server startup delay]]></title>
            <link>https://ruwanyatawara.medium.com/how-to-fix-macos-sierra-wso2-server-starting-delay-4279b9e69133?source=rss-3c2d4df53380------2</link>
            <guid isPermaLink="false">https://medium.com/p/4279b9e69133</guid>
            <category><![CDATA[wso2]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[apple]]></category>
            <category><![CDATA[macos]]></category>
            <category><![CDATA[sierra]]></category>
            <dc:creator><![CDATA[Ruwan Yatawara]]></dc:creator>
            <pubDate>Fri, 27 Jan 2017 05:20:47 GMT</pubDate>
            <atom:updated>2017-01-27T05:38:42.973Z</atom:updated>
            <content:encoded><![CDATA[<p>If you recently upgraded to macOS Sierra (Version 10.12.X and above), chances are you may have noticed a delay in startup time of your WSO2 Server. Well, don’t fret. You are not alone.</p><p>The OS upgrade messes up your host entries. Adding the following entries to your host configuration file will do the trick, and your WSO2 Server will start as it used to.</p><p><strong>Step 1 : Figure out your machine’s local network name</strong></p><p>Open Sharing preferences (from Apple menu &gt; System Preferences, then click Sharing).</p><p>The local network name should be displayed beneath the computer’s name at the top of the pane. (i.e. Ruwan-MacBook-Pro.local)</p><p><strong>Step 2: Change /etc/hosts file</strong></p><p>Make sure your host file does not already contain the following entries. If so.. hard luck, this solution ain’t for you. Something else is wrong with your computer. :(</p><pre>127.0.0.1       localhost       Ruwan-MacBook-Pro.local<br>::1             localhost       Ruwan-MacBook-Pro.local</pre><p>In order to change your host entries, you need to open it with vim editor. Make sure you are super user, if not you will not be able to save your changes.</p><p>Make sure to change the “Ruwan-MacBook-Pro.local” with your machine’s local network name.</p><p>Once done, Your’s may look like this..</p><pre>127.0.0.1       localhost       Ace-MacBook-Pro.local<br>::1             localhost       Ace-MacBook-Pro.local</pre><p>That’s about it your good to go. Your WSO2 Server should start up like it used to, now.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4279b9e69133" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>