<?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 Emirhan Selim Uzun on Medium]]></title>
        <description><![CDATA[Stories by Emirhan Selim Uzun on Medium]]></description>
        <link>https://medium.com/@uemirhanselim?source=rss-73d8bb876e60------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*uCUJMDsQFOTMeaofw3HdeQ.jpeg</url>
            <title>Stories by Emirhan Selim Uzun on Medium</title>
            <link>https://medium.com/@uemirhanselim?source=rss-73d8bb876e60------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 09:01:23 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@uemirhanselim/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[My GitHub Graph Said I Hadn't Coded in a Year. I Commit Almost Every Day.]]></title>
            <link>https://medium.com/@uemirhanselim/my-github-graph-said-i-hadnt-coded-in-a-year-i-commit-almost-every-day-e6b7183b1465?source=rss-73d8bb876e60------2</link>
            <guid isPermaLink="false">https://medium.com/p/e6b7183b1465</guid>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[github]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[git]]></category>
            <dc:creator><![CDATA[Emirhan Selim Uzun]]></dc:creator>
            <pubDate>Wed, 07 Oct 2026 10:32:15 GMT</pubDate>
            <atom:updated>2026-10-07T10:33:52.893Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_xPLXpvOY9xDhn71qT-erA.png" /></figure><h3>My GitHub Graph Said I Hadn’t Coded in a Year. I Commit Almost Every Day.</h3><p><em>Why commits to your company’s private repos don’t show up on your personal profile, and the small script I use to fix it.</em></p><p>I push code to my company’s repositories almost every day. But for a long time my personal GitHub profile looked like I’d quit programming. The contribution graph was empty: no green squares, nothing.</p><h3>It wasn’t my email</h3><p>My first guess was the usual one: commits authored with the wrong email. So I checked:</p><pre>git log --since=&quot;3 months ago&quot; --format=&#39;%an &lt;%ae&gt;&#39; | sort | uniq -c | sort -rn</pre><p>Every commit was authored with my personal Gmail address, which is verified on my personal GitHub account. Then I asked the GitHub API who it thinks wrote one of those commits:</p><pre>gh api repos/&lt;org&gt;/&lt;repo&gt;/commits/&lt;sha&gt; --jq &#39;.author.login&#39;</pre><p>It returned my personal username. So GitHub knew those commits were mine. They were also on the default branch, which is the other common reason commits don’t count.</p><h3>The real reason: my personal account couldn’t see the repo</h3><p>Here’s my setup, and I suspect it’s a common one:</p><ul><li>The company gave me a work GitHub account. That account is a member of the organization.</li><li>My git config uses my personal name and email, so my commits are linked to my personal account.</li><li>My personal account is <strong>not</strong> a member of the organization.</li></ul><p>So I tried reading the repo with my personal account:</p><pre>GH_TOKEN=$(gh auth token --user &lt;personal-username&gt;) gh api repos/&lt;org&gt;/&lt;repo&gt;</pre><ol><li>My personal account had no access to the repository.</li></ol><p>That put me in a gap. The commits belong to my personal account, but that account can’t see the private repo they live in, so they never showed up on its graph. My work account can see the repo, but the commits aren’t authored with its email. Neither profile gets the squares.</p><h3>A quick checklist</h3><p>If your graph is empty, go through these in order:</p><ol><li><strong>Email.</strong> Is the commit’s author email added and verified on the account you expect? Check with git log --format=&#39;%ae&#39;, then look at GitHub → Settings → Emails.</li><li><strong>Branch.</strong> Is the commit on the repo’s default branch? Commits on feature branches or forks only count once they land on the default branch (or gh-pages).</li><li><strong>Access.</strong> Can that account actually see the repo? For a private repo, open it while logged in with that account.</li><li><strong>Private contributions.</strong> Is it turned on? Go to your profile, click <strong>Contribution settings</strong> above the graph, then <strong>Private contributions</strong>. Without it, activity from private repos doesn’t show on your graph at all.</li><li><strong>Time.</strong> New contributions can take up to 24 hours to appear.</li></ol><h3>The proper fix</h3><p>If you can, fix it at the source:</p><ul><li>Ask an org owner to add your personal account to the organization, or as a collaborator on the repos you work on.</li><li>Or do your company work from your personal account in the first place.</li><li>Either way, turn on Private contributions, because company repos are almost always private.</li></ul><p>In my case neither option was available, because my personal account can’t be added to the org. So I built a workaround.</p><h3>The workaround: a private mirror of my activity</h3><p>The idea is simple. My real commits stay where they are. On my personal account I keep a private repo, and for every commit I make at work, I add one <strong>empty</strong> commit with the same timestamp. With Private contributions turned on, those show up as green squares.</p><p>What the script does:</p><ol><li>Uses my work account’s token to list every repo in the organization.</li><li>On each repo’s default branch, finds the commits I authored, skipping merge commits.</li><li>For each new one, creates an empty commit in the mirror repo with the same date, authored with my personal email. The message is just work.</li><li>Pushes the mirror repo using my personal account.</li></ol><p>What it doesn’t do:</p><ul><li>It doesn’t copy any code, file names, repo names or commit messages. The only thing that leaves the company is “I committed something at this time.”</li><li>It doesn’t touch the company repos at all. No hooks, no CI, nothing in shared config.</li></ul><p>To avoid mirroring the same commit twice, each mirror commit carries a short trailer like Ref: 5841a01398609e5f. That&#39;s a SHA-256 hash of the repo name plus the original commit SHA, cut to 16 characters. Nobody can trace it back to the original commit, but the script can tell what it already mirrored.</p><p>People visiting my profile only see the green squares and a line like “N contributions in private repositories”. The repo name isn’t shown.</p><h3>The script</h3><p>You need Python 3 and the GitHub CLI (gh) logged in with both accounts:</p><pre>gh auth login   # work account<br>gh auth login   # personal account<br>gh auth status  # should list both</pre><p>Save this as ~/.activity-mirror/sync.py and fill in the config block at the top:</p><pre>#!/usr/bin/env python3<br>import hashlib<br>import json<br>import os<br>import shutil<br>import subprocess<br>import sys<br>from collections import Counter<br>from datetime import datetime<br><br># --- configure these ---<br>ORG = &quot;your-company&quot;                       # the GitHub org you work in<br>WORK_ACCOUNT = &quot;you-at-work&quot;               # gh account that can read the org&#39;s repos<br>AUTHORS = [&quot;you-personal&quot;, &quot;you-at-work&quot;]  # whose commits to mirror (GitHub usernames)<br>PERSONAL_ACCOUNT = &quot;you-personal&quot;          # where the mirror repo lives<br>PERSONAL_EMAIL = &quot;you@example.com&quot;         # a verified email on your personal account<br>MIRROR_REPO = &quot;work-activity&quot;<br>MIRROR_DIR = os.path.expanduser(&quot;~/.activity-mirror/repo&quot;)<br># -----------------------<br><br>GH = shutil.which(&quot;gh&quot;)<br>GIT = shutil.which(&quot;git&quot;)<br><br><br>def run(cmd, env=None, check=True):<br>    result = subprocess.run(cmd, capture_output=True, text=True, env=env)<br>    if check and result.returncode:<br>        sys.exit(f&quot;{&#39; &#39;.join(cmd[:4])} failed: {result.stderr.strip()}&quot;)<br>    return result<br><br><br>def token(user):<br>    return run([GH, &quot;auth&quot;, &quot;token&quot;, &quot;--user&quot;, user]).stdout.strip()<br><br><br>def api(path, tok):<br>    env = {**os.environ, &quot;GH_TOKEN&quot;: tok}<br>    out = run([GH, &quot;api&quot;, &quot;--paginate&quot;, path, &quot;--jq&quot;, &quot;.[]|@json&quot;], env=env, check=False)<br>    if out.returncode:<br>        return []<br>    return [json.loads(line) for line in out.stdout.splitlines() if line]<br><br><br>def work_commits():<br>    tok = token(WORK_ACCOUNT)<br>    found = {}<br>    for repo in api(f&quot;orgs/{ORG}/repos?per_page=100&quot;, tok):<br>        for author in AUTHORS:<br>            path = f&quot;repos/{ORG}/{repo[&#39;name&#39;]}/commits?sha={repo[&#39;default_branch&#39;]}&amp;author={author}&amp;per_page=100&quot;<br>            for c in api(path, tok):<br>                if len(c[&quot;parents&quot;]) != 1:<br>                    continue  # skip merge commits<br>                ref = hashlib.sha256(f&quot;{repo[&#39;name&#39;]}:{c[&#39;sha&#39;]}&quot;.encode()).hexdigest()[:16]<br>                utc = datetime.fromisoformat(c[&quot;commit&quot;][&quot;author&quot;][&quot;date&quot;].replace(&quot;Z&quot;, &quot;+00:00&quot;))<br>                found[ref] = utc.astimezone().isoformat()  # keep your local timezone<br>    return found<br><br><br>def git(*args, env=None, check=True):<br>    return run([GIT, &quot;-C&quot;, MIRROR_DIR, *args], env=env, check=check)<br><br><br>def ensure_mirror():<br>    if os.path.isdir(os.path.join(MIRROR_DIR, &quot;.git&quot;)):<br>        return<br>    env = {**os.environ, &quot;GH_TOKEN&quot;: token(PERSONAL_ACCOUNT)}<br>    full = f&quot;{PERSONAL_ACCOUNT}/{MIRROR_REPO}&quot;<br>    if run([GH, &quot;api&quot;, f&quot;repos/{full}&quot;], env=env, check=False).returncode:<br>        run([GH, &quot;repo&quot;, &quot;create&quot;, full, &quot;--private&quot;], env=env)<br>    os.makedirs(MIRROR_DIR, exist_ok=True)<br>    git(&quot;init&quot;, &quot;-b&quot;, &quot;main&quot;)<br>    git(&quot;remote&quot;, &quot;add&quot;, &quot;origin&quot;, f&quot;@github.com/{full}.git&quot;&gt;https://{PERSONAL_ACCOUNT}@github.com/{full}.git&quot;)<br>    # gh&#39;s credential helper only hands out the ACTIVE account&#39;s token,<br>    # so this repo gets its own helper that always uses the personal account.<br>    key = &quot;credential.https://github.com.helper&quot;<br>    git(&quot;config&quot;, &quot;--add&quot;, key, &quot;&quot;)<br>    git(&quot;config&quot;, &quot;--add&quot;, key,<br>        f&#39;!f() {{ test &quot;$1&quot; = get &amp;&amp; echo username={PERSONAL_ACCOUNT} &amp;&amp; echo password=$({GH} auth token --user {PERSONAL_ACCOUNT}); }}; f&#39;)<br>    git(&quot;config&quot;, &quot;commit.gpgsign&quot;, &quot;false&quot;)<br>    git(&quot;config&quot;, &quot;core.hooksPath&quot;, &quot;/dev/null&quot;)<br>    if git(&quot;ls-remote&quot;, &quot;--heads&quot;, &quot;origin&quot;, &quot;main&quot;).stdout.strip():<br>        git(&quot;pull&quot;, &quot;origin&quot;, &quot;main&quot;)<br><br><br>def mirrored_refs():<br>    if git(&quot;rev-parse&quot;, &quot;--verify&quot;, &quot;HEAD&quot;, check=False).returncode:<br>        return set()<br>    out = git(&quot;log&quot;, &quot;--format=%(trailers:key=Ref,valueonly)&quot;).stdout<br>    return {line.strip() for line in out.splitlines() if line.strip()}<br><br><br>def main():<br>    found = work_commits()<br>    if &quot;--dry-run&quot; in sys.argv:<br>        days = Counter(d[:10] for d in found.values())<br>        print(f&quot;{len(found)} commits over {len(days)} days ({min(days)} -&gt; {max(days)})&quot;)<br>        return<br><br>    ensure_mirror()<br>    done = mirrored_refs()<br>    pending = sorted((date, ref) for ref, date in found.items() if ref not in done)<br>    for date, ref in pending:<br>        env = {**os.environ,<br>               &quot;GIT_AUTHOR_NAME&quot;: PERSONAL_ACCOUNT, &quot;GIT_AUTHOR_EMAIL&quot;: PERSONAL_EMAIL, &quot;GIT_AUTHOR_DATE&quot;: date,<br>               &quot;GIT_COMMITTER_NAME&quot;: PERSONAL_ACCOUNT, &quot;GIT_COMMITTER_EMAIL&quot;: PERSONAL_EMAIL, &quot;GIT_COMMITTER_DATE&quot;: date}<br>        git(&quot;commit&quot;, &quot;--allow-empty&quot;, &quot;--no-verify&quot;, &quot;-q&quot;, &quot;-m&quot;, &quot;work&quot;, &quot;-m&quot;, f&quot;Ref: {ref}&quot;, env=env)<br>    if pending:<br>        git(&quot;push&quot;, &quot;-q&quot;, &quot;-u&quot;, &quot;origin&quot;, &quot;main&quot;)<br>    print(f&quot;{datetime.now():%Y-%m-%d %H:%M} mirrored {len(pending)} new, {len(done) + len(pending)} total&quot;)<br><br><br>if __name__ == &quot;__main__&quot;:<br>    main()</pre><p>First, see what it would mirror without writing anything:</p><pre>python3 ~/.activity-mirror/sync.py --dry-run</pre><p>For me that printed 338 commits over 139 days, almost a year of work that had been invisible.</p><p>Then run it without --dry-run. The first run creates the private repo, backfills everything and pushes. After that, each run only adds commits it hasn&#39;t mirrored yet.</p><h3>One gotcha: two gh accounts</h3><p>My first real run failed on push with could not read Password. It turns out gh&#39;s git credential helper only returns the token of the <strong>active</strong> account. Mine was the work account, which can&#39;t access my personal repo.</p><p>I could switch accounts with gh auth switch every time, but I&#39;d forget. So the script gives the mirror repo its own credential helper that always asks gh for the personal account&#39;s token. It&#39;s set only inside the mirror repo, so my global git config and the company repos keep using the work account.</p><h3>Run it automatically</h3><p>On macOS I use launchd. Save this as ~/Library/LaunchAgents/com.activity-mirror.sync.plist, replacing YOUR_USER:</p><pre>&lt;?xml version=&quot;1.0&quot; encoding=&quot;UTF-8&quot;?&gt;<br>&lt;!DOCTYPE plist PUBLIC &quot;-//Apple//DTD PLIST 1.0//EN&quot; &quot;http://www.apple.com/DTDs/PropertyList-1.0.dtd&quot;&gt;<br>&lt;plist version=&quot;1.0&quot;&gt;<br>&lt;dict&gt;<br>  &lt;key&gt;Label&lt;/key&gt;&lt;string&gt;com.activity-mirror.sync&lt;/string&gt;<br>  &lt;key&gt;ProgramArguments&lt;/key&gt;<br>  &lt;array&gt;<br>    &lt;string&gt;/usr/bin/env&lt;/string&gt;<br>    &lt;string&gt;python3&lt;/string&gt;<br>    &lt;string&gt;/Users/YOUR_USER/.activity-mirror/sync.py&lt;/string&gt;<br>  &lt;/array&gt;<br>  &lt;key&gt;EnvironmentVariables&lt;/key&gt;<br>  &lt;dict&gt;<br>    &lt;key&gt;PATH&lt;/key&gt;&lt;string&gt;/opt/homebrew/bin:/usr/local/bin:/usr/bin:/bin&lt;/string&gt;<br>  &lt;/dict&gt;<br>  &lt;key&gt;StartInterval&lt;/key&gt;&lt;integer&gt;10800&lt;/integer&gt;<br>  &lt;key&gt;RunAtLoad&lt;/key&gt;&lt;true/&gt;<br>  &lt;key&gt;StandardOutPath&lt;/key&gt;&lt;string&gt;/Users/YOUR_USER/.activity-mirror/sync.log&lt;/string&gt;<br>  &lt;key&gt;StandardErrorPath&lt;/key&gt;&lt;string&gt;/Users/YOUR_USER/.activity-mirror/sync.log&lt;/string&gt;<br>&lt;/dict&gt;<br>&lt;/plist&gt;</pre><p>Load it:</p><pre>launchctl bootstrap gui/$(id -u) ~/Library/LaunchAgents/com.activity-mirror.sync.plist</pre><p>It runs once right away, then every three hours. Check the log with cat ~/.activity-mirror/sync.log. Don&#39;t panic if the log is empty for the first minute. The script only prints when it&#39;s done, and scanning a whole org takes a while. If macOS asks whether a background process can access gh&#39;s keychain item, choose <strong>Always Allow</strong>, or the job can&#39;t read your tokens.</p><p>On Linux, a cron line does the same job:</p><pre>0 */3 * * * python3 $HOME/.activity-mirror/sync.py &gt;&gt; $HOME/.activity-mirror/sync.log 2&gt;&amp;1</pre><h3>Before you do this</h3><ul><li><strong>Be honest about what it is.</strong> The squares reflect real work, but the commits themselves are placeholders. If anyone asks, my work is in private company repos and this mirrors when I committed. I would never use it to invent activity.</li><li><strong>Check your company’s policy.</strong> Only timestamps leave the org, but some companies care about that too.</li><li><strong>Watch out for double counting.</strong> If your personal account ever gets access to the org, your real commits may start counting and every commit could show up twice. Delete the mirror repo then.</li><li><strong>Only merged work counts.</strong> Like GitHub, the script only looks at default branches, so work in open PRs shows up once it’s merged.</li><li><strong>Dates stay correct.</strong> Timestamps are converted to your machine’s timezone, so a commit at 1 AM lands on the right day.</li></ul><p>My profile finally looks like my actual week. If you’ve been staring at an empty graph and wondering what you did wrong, it’s probably not your email. Check whether that account can see the repo.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Yd6KVONpsRevkdOoEV_WPw.png" /></figure><p>Till next time 👋🏻</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e6b7183b1465" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Handling Play Console’s New Media Policy — Android Photo Picker]]></title>
            <link>https://medium.com/@uemirhanselim/handling-play-consoles-new-media-policy-android-photo-picker-ccb3b86f2368?source=rss-73d8bb876e60------2</link>
            <guid isPermaLink="false">https://medium.com/p/ccb3b86f2368</guid>
            <category><![CDATA[policy]]></category>
            <category><![CDATA[google-play-console]]></category>
            <category><![CDATA[flutter]]></category>
            <category><![CDATA[android]]></category>
            <category><![CDATA[photopicker]]></category>
            <dc:creator><![CDATA[Emirhan Selim Uzun]]></dc:creator>
            <pubDate>Wed, 29 Oct 2025 08:09:16 GMT</pubDate>
            <atom:updated>2025-10-29T08:09:16.953Z</atom:updated>
            <content:encoded><![CDATA[<h3>Handling Play Console’s New Media Policy — Android Photo Picker</h3><p>“Avoid policy rejections with simple steps”</p><figure><img alt="https://android-developers.googleblog.com/2023/04/photo-picker-everywhere.html" src="https://cdn-images-1.medium.com/max/1024/1*ssjrsRTaQ860haACim5qBQ.png" /></figure><p>Our Flutter app has been on the Play Store for over <strong>three years</strong>.<br>It already had everything working perfectly — users could upload profile photos, create posts with images, and change their avatars without any issue.</p><p>Then one day, while trying to upload a new build, Google Play threw a warning I hadn’t seen before:</p><blockquote><em>“Your app is requesting broad photo access. Please use the Android Photo Picker to comply with updated privacy policies.”</em></blockquote><p>That’s when I realized — Google had quietly changed how apps should handle <strong>media permissions</strong>.</p><h3>1 — What Changed in Android’s Media Policy</h3><p>Starting with <strong>Android 13 (API 33)</strong>, Google introduced a new <strong>Photo Picker</strong> and stricter privacy rules around media access.</p><p>The goal?<br>To make sure apps can’t access a user’s entire gallery unless absolutely necessary.</p><p>From Google’s own words:</p><blockquote><em>“Improving privacy remains a top priority on Android. We’ve been investing in the platform to give users more control, increase transparency, and reduce the scope of access to private data.”</em></blockquote><p>That means permissions like<br>READ_EXTERNAL_STORAGE,<br>READ_MEDIA_IMAGES,<br>and READ_MEDIA_VIDEO<br>are no longer the right way to go for most apps — especially if you’re just letting users pick photos.</p><h3>2 — Solution Steps</h3><ul><li>2.1 → Update <strong>image_picker</strong> to the latest version. ( 1.1.2 to 1.2.0)</li><li>2.2 → Remove legacy permissions from <strong>AndroidManifest.xml</strong></li></ul><pre> &lt;uses-permission android:name=&quot;android.permission.READ_EXTERNAL_STORAGE&quot; /&gt;<br> &lt;uses-permission android:name=&quot;android.permission.READ_MEDIA_IMAGES&quot; /&gt;<br> &lt;uses-permission android:name=&quot;android.permission.READ_MEDIA_VIDEO&quot; /&gt;</pre><ul><li>2.3 → Keep manual <strong>permission</strong> checks only for <strong>Android 12</strong> and below (<strong>API 32</strong>−)</li></ul><pre>final androidInfo = await DeviceInfoPlugin().androidInfo;<br><br>if (androidInfo.version.sdkInt &lt;= 32) {<br>  var status = await Permission.storage.status;<br>  if (status.isDenied) await Permission.storage.request();<br>} else {<br>  // Android 13+ uses system Photo Picker, no permission needed<br>  postVM.selectMultiImages(context);<br>}</pre><p>That’s it.<br>No rewrites, no headaches — just precise adjustments.</p><h3>3 — Why This Works Perfectly</h3><p>The image_picker package now supports the <strong>Android Photo Picker</strong> under the hood.<br>So if your target SDK is <strong>33 or higher</strong>, the picker runs through the secure system interface automatically, and you no longer need storage or photo permissions.</p><p>For <strong>Android 12 and below</strong>, you still have to request Permission.storage, which is completely fine and compliant with Play Store rules.</p><p><strong><em>Advices from me</em></strong>:</p><ul><li>Always try to keep your flutter plugins up to date.</li><li>Do not ask for unnecessary permission for both android and ios which can cause for possible rejections from stores.</li></ul><p>If you liked my article don’t forget to subscribe and share it with your friends ❤.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ccb3b86f2368" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[NoSQL vs Relational Databases: What You Need to Know]]></title>
            <link>https://medium.com/@uemirhanselim/nosql-vs-relational-databases-what-you-need-to-know-961b8a7b3f59?source=rss-73d8bb876e60------2</link>
            <guid isPermaLink="false">https://medium.com/p/961b8a7b3f59</guid>
            <category><![CDATA[database]]></category>
            <category><![CDATA[nosql]]></category>
            <category><![CDATA[backend]]></category>
            <category><![CDATA[sql]]></category>
            <category><![CDATA[relational-databases]]></category>
            <dc:creator><![CDATA[Emirhan Selim Uzun]]></dc:creator>
            <pubDate>Thu, 21 Aug 2025 07:29:46 GMT</pubDate>
            <atom:updated>2025-08-21T07:29:46.598Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*N4BdoQC1iz007sifQR5oiA.jpeg" /><figcaption>📸 <em>Just a representative me 😅</em></figcaption></figure><p>When I first started working with databases, like many developers, my entry point was the <strong>relational database world</strong> — MySQL, PostgreSQL, SQL Server. Tables, rows, primary keys, foreign keys… everything was structured and well-defined. But as I dug deeper into different types of projects, I realized that relational databases weren’t always the best fit. That’s where <strong>NoSQL databases</strong> came into the picture.</p><p>In this article, I’ll break down the core differences between <strong>Relational Databases (RDBMS)</strong> and <strong>NoSQL databases</strong>, their strengths and weaknesses, and when to choose one over the other.</p><h3>Relational Databases (RDBMS)</h3><p>Relational databases are built around the concept of <strong>tables</strong>. Data is stored in rows and columns, and relationships between data are enforced with <strong>primary keys</strong> and <strong>foreign keys</strong>.</p><p><strong>Key characteristics:</strong></p><ul><li><strong>Structured Schema</strong>: You define the schema (columns, data types) before inserting data.</li><li><strong>Strong Consistency</strong>: ACID compliance (Atomicity, Consistency, Isolation, Durability).</li><li><strong>Great for Complex Queries</strong>: SQL is powerful for joins, aggregations, and reporting.</li></ul><p><strong>Examples</strong>: MySQL, PostgreSQL, Oracle, SQL Server.</p><p><strong>Best Use Cases</strong>:</p><ul><li>Financial applications (banking, transactions).</li><li>Applications that require strong consistency and well-defined data relationships.</li><li>Analytics and reporting systems.</li></ul><h3>NoSQL Databases</h3><p>NoSQL databases were designed to handle the modern challenges of <strong>big data, scalability, and flexibility</strong>. Unlike RDBMS, NoSQL databases don’t rely on fixed schemas or rigid table structures.</p><p><strong>Key characteristics:</strong></p><ul><li><strong>Flexible Schema</strong>: Data can be semi-structured or unstructured (JSON, key-value pairs, documents, graphs).</li><li><strong>Scalability</strong>: Typically designed for horizontal scaling, making them great for large-scale applications.</li><li><strong>Variety of Models</strong>: Key-Value Stores (Redis), Document Stores (MongoDB), Column Stores (Cassandra), Graph Databases (Neo4j).</li></ul><p><strong>Examples</strong>: MongoDB, Cassandra, Redis, DynamoDB.</p><p><strong>Best Use Cases</strong>:</p><ul><li>Real-time analytics and big data applications.</li><li>Content management systems or apps with rapidly changing requirements.</li><li>IoT, gaming, social media platforms that need fast and flexible data handling.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qjy9sPsjM_Bm0H0c5rBXSQ.png" /></figure><h3>So, Which One Should You Use?</h3><p>The truth is, there’s no “one-size-fits-all” answer. If you’re working on a banking app, a medical system, or any environment where <strong>data integrity and complex relationships</strong> are crucial, a <strong>relational database</strong> is the safer choice.</p><p>On the other hand, if your project needs <strong>speed, flexibility, and scalability</strong> — for example, a social media app handling millions of users — then <strong>NoSQL</strong> will likely serve you better.</p><p>In practice, many modern systems actually combine both approaches: a relational database for structured, mission-critical data and a NoSQL database for handling large volumes of semi-structured or unstructured data.</p><p>👉 If you’re just starting out, I recommend learning the fundamentals of both. Understand relational databases deeply first (because SQL is everywhere), then explore NoSQL with projects that push you out of the rigid schema world.</p><p>See you in the next article!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=961b8a7b3f59" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Node.js vs PHP: Which One Should You Learn as a Beginner?]]></title>
            <link>https://medium.com/@uemirhanselim/node-js-vs-php-which-one-should-you-learn-as-a-beginner-c76cb36601d0?source=rss-73d8bb876e60------2</link>
            <guid isPermaLink="false">https://medium.com/p/c76cb36601d0</guid>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[nodejs]]></category>
            <category><![CDATA[php]]></category>
            <dc:creator><![CDATA[Emirhan Selim Uzun]]></dc:creator>
            <pubDate>Fri, 08 Aug 2025 19:07:35 GMT</pubDate>
            <atom:updated>2025-08-08T19:07:35.432Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*g852iKIZFo8PvjpLDjYEGQ.jpeg" /></figure><p>When starting your journey in backend development, it’s common to come across <strong>Node.js</strong> and <strong>PHP</strong>. Both are popular for building web applications, but they take very different approaches.</p><p>If you’re just starting out, let’s compare them in simple terms.</p><h3>🖥 What is PHP?</h3><ul><li>A <strong>server-side scripting language</strong> created in 1994.</li><li>Often used with <strong>Apache</strong> or <strong>Nginx</strong> servers.</li><li>Traditionally processes each request separately.</li></ul><p><strong>Where it’s used:</strong></p><ul><li>WordPress</li><li>Drupal</li><li>Many traditional websites and CMS platforms.</li></ul><p><strong>Pros:</strong></p><ul><li>Easy to start with (especially for beginners).</li><li>Huge amount of tutorials and resources.</li><li>Works out-of-the-box with most hosting services.</li></ul><p><strong>Cons:</strong></p><ul><li>Less suited for real-time applications.</li><li>Can be slower compared to modern event-driven systems.</li></ul><h3>⚡ What is Node.js?</h3><ul><li>A <strong>JavaScript runtime</strong> built on Chrome’s V8 engine.</li><li>Uses a <strong>single-thread, non-blocking</strong> architecture.</li><li>Great for handling many requests at the same time.</li></ul><p><strong>Where it’s used:</strong></p><ul><li>Real-time chat apps</li><li>APIs for mobile apps</li><li>Streaming platforms.</li></ul><p><strong>Pros:</strong></p><ul><li>One language (JavaScript) for both frontend and backend.</li><li>Excellent for real-time and high-concurrency apps.</li><li>Large and active package ecosystem (npm).</li></ul><p><strong>Cons:</strong></p><ul><li>Learning curve for async programming.</li><li>CPU-heavy tasks can be tricky to handle.</li></ul><h3>⚖️ Quick Comparison</h3><p><strong>Language</strong></p><ul><li>PHP: PHP</li><li>Node.js: JavaScript</li></ul><p><strong>Architecture</strong></p><ul><li>PHP: Multi-thread (per request)</li><li>Node.js: Single-thread, event-driven</li></ul><p><strong>Best For</strong></p><ul><li>PHP: Websites, CMS</li><li>Node.js: APIs, real-time apps</li></ul><p><strong>Hosting</strong></p><ul><li>PHP: Widely supported on cheap hosts</li><li>Node.js: Needs specific environment</li></ul><p><strong>Learning Curve</strong></p><ul><li>PHP: Easier at the start</li><li>Node.js: Slightly steeper</li></ul><h3>🏁 Which Should You Choose?</h3><p>If you want to quickly build a <strong>traditional website</strong> or work with WordPress, <strong>PHP</strong> might be the fastest route.<br>If you aim to create <strong>modern, real-time applications</strong> or APIs — and you already know JavaScript — <strong>Node.js</strong> could be a better investment.</p><p>The good news? You can always learn both over time.</p><p>Thanks for reading!<br>Follow me for more beginner-friendly programming tips.<br><strong>See you in the next article! 👋</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c76cb36601d0" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Understanding "let" vs "const" in JavaScript: What Should You Use and When?]]></title>
            <link>https://medium.com/@uemirhanselim/understanding-let-vs-const-in-javascript-what-should-you-use-and-when-49860848453c?source=rss-73d8bb876e60------2</link>
            <guid isPermaLink="false">https://medium.com/p/49860848453c</guid>
            <category><![CDATA[backend]]></category>
            <category><![CDATA[nodejs]]></category>
            <category><![CDATA[clean-code]]></category>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[maintainability]]></category>
            <dc:creator><![CDATA[Emirhan Selim Uzun]]></dc:creator>
            <pubDate>Thu, 07 Aug 2025 21:04:06 GMT</pubDate>
            <atom:updated>2025-08-07T21:04:06.279Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*E2k8kY3-U7OZYEuwToIHQw.png" /></figure><p>When you start writing JavaScript — whether in the browser or with Node.js — one of the first things you’ll run into is variable declarations. And if you’re like me when I started, you might find yourself asking:<br><strong>Should I use </strong><strong>let or </strong><strong>const? What&#39;s the difference anyway?</strong></p><p>Let’s break it down.</p><h3>The Basics</h3><p>JavaScript offers three ways to declare variables: var, let, and const. While var is now mostly considered outdated due to its confusing scoping behavior, let and const are ES6+ features that behave more predictably and are widely used in modern Node.js development.</p><p>Here’s a quick overview:</p><ul><li>let allows you to declare variables that can be reassigned.</li><li>const is for variables that should not be reassigned.</li></ul><p>Sounds simple, right? But there’s more nuance to it.</p><h3>Let’s Talk About let</h3><p>You use let when you know the value of your variable will change.</p><pre>let counter = 0;<br>counter = counter + 1;</pre><p>This works perfectly because let is mutable. You can also declare it without an initial value, and assign later:</p><pre>let name;<br>name = &quot;Alice&quot;;</pre><p>Another important feature of let is <strong>block scoping</strong>. Unlike var, let doesn’t leak outside the block:</p><pre>if (true) {<br>  let message = &quot;Hello!&quot;;<br>  console.log(message); // Hello!<br>}<br>console.log(message); // ReferenceError</pre><h3>Constant Love for const</h3><p>const, on the other hand, means the variable <strong>can’t be reassigned</strong> after it’s declared.</p><pre>const API_KEY = &quot;12345&quot;;<br>// API_KEY = &quot;67890&quot;; ❌ This will throw an error</pre><p>However, <strong>const doesn’t make the value immutable</strong> — it only prevents reassignment.</p><pre>const user = { name: &quot;Alice&quot; };<br>user.name = &quot;Bob&quot;; // ✅ Totally fine</pre><p>So if you’re dealing with objects or arrays, the content inside them can still be changed:</p><pre>const numbers = [1, 2, 3];<br>numbers.push(4); // ✅<br>numbers = [1, 2, 3, 4]; // ❌ Error!</pre><h3>Which One Should You Use?</h3><p>Here’s a general rule that I personally follow, and many devs recommend:</p><blockquote><strong><em>Default to </em></strong><strong><em>const. Use </em></strong><strong><em>let only when you need to reassign.</em></strong></blockquote><p>Why?</p><ol><li><strong>Intentionality</strong>: Using const makes your code more readable. It signals to other developers (and your future self) that this value is not supposed to change.</li><li><strong>Safety</strong>: It prevents accidental reassignment.</li><li><strong>Best practices</strong>: It aligns with functional programming principles, which are growing in popularity.</li></ol><h3>Real-World Example in Node.js</h3><p>Let’s say you’re building a simple API in Node.js:</p><pre>const express = require(&quot;express&quot;);<br>const app = express();<br><br>let requestCount = 0;<br><br>app.get(&quot;/&quot;, (req, res) =&gt; {<br>  requestCount += 1;<br>  res.send(`This endpoint has been hit ${requestCount} times.`);<br>});<br><br>app.listen(3000, () =&gt; {<br>  console.log(&quot;Server running on port 3000&quot;);<br>});</pre><ul><li>We use const for express and app because they’re not going to be reassigned.</li><li>We use let for requestCount because it changes on every request.</li></ul><p>See how that works?</p><h3>Final Thoughts</h3><p>Choosing between let and const isn’t just a syntax decision—it’s about writing clearer, more maintainable code.</p><p><strong>Use </strong><strong>const as your default. Reach for </strong><strong>let only when reassignment is truly necessary.</strong></p><p>In the end, your future self (and your teammates) will thank you.</p><p>See you in the next article…</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/299/1*BY9I47vq9sPxge1RF2HYaw.jpeg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=49860848453c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why You Should Start Building Your Architecture Clean!]]></title>
            <link>https://medium.com/@uemirhanselim/why-you-should-start-building-your-architecture-clean-1f84b7aecec7?source=rss-73d8bb876e60------2</link>
            <guid isPermaLink="false">https://medium.com/p/1f84b7aecec7</guid>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[clean-architecture]]></category>
            <category><![CDATA[clean-code]]></category>
            <dc:creator><![CDATA[Emirhan Selim Uzun]]></dc:creator>
            <pubDate>Mon, 04 Dec 2023 07:14:58 GMT</pubDate>
            <atom:updated>2023-12-04T07:14:58.324Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Why you should write your code clean? How this will effect to your productivity? And more…</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*AaVQbLuo6C4Xp4ge-lStlQ.jpeg" /></figure><p>This is my first article, I want to explain why I decided to write at the beginning, <strong><em>you can skip this part</em></strong> if you wish. I have three reasons:</p><p>1- I want to contribute to community of engineers, developers or whatever you say.</p><p>2- I seek to improve my knowledge of software and programming skills.</p><p>3- I desire to advance my English. Because I always listen and read in English but what I am doing is, I always getting information from outside (input). To improve output I need to write and speak.</p><p>Let’s get back to our topic, <strong>the idea behind the clean architecture</strong> is to build sustainable and maintainable design. Why I used the design keyword here, indeed design and architecture not that different terms. They cannot exist without having eachother. I will not get into so deep on clean architecture, I will simply explain it by giving an example below.</p><p>I will show you 4 graphs of <strong>product size</strong>, <strong>cost per line</strong>, <strong>productivity</strong> and <strong>development payroll</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/569/1*XoyfFkQ1eJOWGmkIGd6WJA.png" /><figcaption>Figure 1</figcaption></figure><p>As you can see in the <em>Figure 1, </em>this product grows with no problem in first 5 years but then it’s <strong>growth size speed is decreasing</strong>. What that means?</p><p>This means this project is not <strong>sustainable</strong>, first few years it is goes well but then it starts slowing.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/575/1*aY8w_czMs0HI0DjuHgZa8w.jpeg" /><figcaption>Figure 2</figcaption></figure><p>At the <em>Figure 2</em>, it shows cost per line of code. Let me explain, when the architecture is not <strong>sustainable</strong> and <strong>maintainable</strong> it will turn into spaghetti by time. Which means <strong>developers will have to spend more time and show more effort</strong> on the project. Eventually this will increase the cost per line of code.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/950/1*8efW5FwqQZHZAqPHl7zgzQ.png" /><figcaption>Figure 3</figcaption></figure><p>In <em>Figure 3</em>, it shows productivity. From developer’s point imagine you are giving all of you effort but in return you are not getting much. That’s causes loss of self-confidence. Developer’s <strong>productiviy</strong> goes down day by day. Let’s see it from Executive’s point.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IL_wlpq8L_vf3F604FNaDQ.png" /><figcaption>Figure 4</figcaption></figure><p>From executive’s point, all of the problems which I mentioned above will cause to lose lots of money to company. Okay, let’s make it clear here when you do not write/build clean code/architecture first it will effect the developer’s productivity then it will cause to going down at the product size growth speed at the and all of these will cost a lot to company.</p><p>In my humble opinion, do not postpone architecture problems. Most of developers (incuding me) delay small problems in code/architecture and say “I can fix it later”. But later developer gets new assignments and do not have time to go back to fix the problems.</p><p>I hope I did not bore you. Take care of yourself, see you until the next article.</p><h3><strong>References:</strong></h3><ul><li>MARTIN, Robert C. Clean Architecture: A Craftsman’s Guide to Software Structure and Design. 1st. ed. USA: Prentice Hall Press, 2017. ISBN 0134494164.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1f84b7aecec7" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>