<?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">

<channel>

<title><![CDATA[Fezcodex]]></title>

<description><![CDATA[A personal blog by Ahmed Samil Bulbul]]></description>

<link>https://fezcode.com</link>

<image>

<url>https://fezcode.com/logo512.png</url>

<title>Fezcodex</title>

<link>https://fezcode.com</link>

</image>

<generator>RSS for Node</generator>

<lastBuildDate>Sun, 18 Jan 2026 11:33:22 GMT</lastBuildDate>

<atom:link href="https://fezcode.com/rss.xml" rel="self" type="application/rss+xml"/>

<pubDate>Sun, 18 Jan 2026 11:33:22 GMT</pubDate>

<copyright><![CDATA[2026 Ahmed Samil Bulbul]]></copyright>

<language><![CDATA[en]]></language>

<managingEditor><![CDATA[samil.bulbul@gmail.com (Ahmed Samil Bulbul)]]></managingEditor>

<webMaster><![CDATA[samil.bulbul@gmail.com (Ahmed Samil Bulbul)]]></webMaster>

<ttl>60</ttl>

<item>

<title><![CDATA[Dying is Easy, Comedy is Statistically Impossible: An IMDbayes Analysis]]></title>

<description><![CDATA[[object Object]]]></description>

<link>https://fezcode.com/blog/what-genre-should-i-watch</link>

<guid isPermaLink="false">https://fezcode.com/blog/what-genre-should-i-watch</guid>

<dc:creator><![CDATA[Ahmed Samil Bulbul]]></dc:creator>

<pubDate>Sun, 18 Jan 2026 00:00:00 GMT</pubDate>

<content:encoded><![CDATA[<blockquote>

<p>This analysis was built by a Software Engineer relying on 8-year-old university memories of statistics.

If the math looks wrong, just assume it&#39;s a feature, not a bug.

You can always contact me.</p>

</blockquote>

<h2>Deconstructing Hollywood: A Data Science Journey from Raw Data to p99 Insights</h2>

<p>As software engineers, we are used to deterministic systems. If <code>a = b</code>, then a equals b.

Data Science, however, deals with probability, distributions, and noise.

It&#39;s less about &quot;<strong>what is the answer</strong>&quot; and more about &quot;<strong>how confident are we in this trend?</strong>&quot;</p>

<p>Recently, I wanted to bridge my engineering background with data science to answer a simple pop-culture question:

<strong>How do different movie genres actually perform?</strong></p>

<p>Are &quot;<strong>Action</strong>&quot; movies inherently rated lower than &quot;<strong>Dramas</strong>&quot;? Is it harder to make a masterpiece &quot;<strong>Horror</strong>&quot; movie than a masterpiece &quot;<strong>Biography</strong>&quot;?</p>

<p>To answer this, I didn&#39;t just want to run a script; I wanted to build a production-grade Data Science lab?!. (/s)

This post details the entire journey—from choosing the modern Python stack and engineering the data pipeline to defining

the statistical metrics that reveal the &quot;truth&quot; behind average ratings.</p>

<h2>Part 1: The Engineering Foundation</h2>

<p>A data project is only as good as its environment. I wanted a setup that was fast, reproducible, and clean.</p>

<h3>The Stack Decision</h3>

<p>I chose Python because it is the undisputed <strong><a href="/vocab/lingua-franca">lingua franca</a></strong> of data science.

The ecosystem (Pandas for data crunching, Seaborn for visualization) is unmatched.</p>

<h3>The Package Manager: Why <code>uv</code>?</h3>

<p>Traditionally, Python data science relies on Conda because it manages complex C-library dependencies used by

math libraries like NumPy. However, Conda can be slow and bloated.</p>

<p>For this project, I chose <code>uv</code>.</p>

<p><code>uv</code> is a modern, blazing-fast Python package manager written in Rust.

It replaces <code>pip</code>, <code>poetry</code>, and <code>virtualenv</code>. It resolves dependencies in milliseconds and creates deterministic environments instantly.

For a project relying on standard wheels like Pandas, <code>uv</code> provides a vastly superior developer experience.</p>

<pre><code class="language-bash"># Setting up the environment took seconds

$ uv init movie-analysis

$ uv python install 3.10

$ uv add pandas matplotlib seaborn scipy jupyter ipykernel

</code></pre>

<p>Then connected VS Code to this <code>.venv</code> created by <code>uv</code>, giving me a robust Jupyter Notebook experience right in the IDE.</p>

<h2>Part 2: The Data Pipeline (ETL)</h2>

<p>I needed data with genres, votes, and ratings, went straight to the source: the <strong>IMDb Non-Commercial Datasets</strong>.</p>

<p>Then I faced a classic data engineering challenge: these are massive TSV (Tab Separated Values) files.

Loading the entirety of IMDb into RAM on a laptop is a bad idea.</p>

<p>Solution? Build a Python ETL script to handle ingestion smartly:</p>

<ol>

<li><strong>Stream &amp; Filter</strong>: used Pandas to read the raw files in chunks, filtering immediately for <code>titleType == &#39;movie&#39;</code> and excluding older films. This kept memory usage low.</li>

<li><strong>Merge</strong>: joined the <code>title.basics</code> (genres/names) with <code>title.ratings</code> (scores/votes) on their unique IDs.</li>

<li><strong>The &quot;Explode&quot;</strong>: This was the crucial data transformation step. IMDb lists genres as a single string: &quot;Action,Adventure,Sci-Fi&quot;. To analyze by category, I had to split that string and &quot;explode&quot; the dataset, duplicating the movie row for each genre it belongs to.</li>

</ol>

<pre><code class="language-python"># Transforming &quot;Action,Comedy&quot; into two distinct analysis rows

df[&#39;genres&#39;] = df[&#39;genres&#39;].str.split(&#39;,&#39;)

df_exploded = df.explode(&#39;genres&#39;)

</code></pre>

<h2>Part 3: The Science (Beyond Averages)</h2>

<p>With clean data in hand, we moved into a Jupyter Notebook for Exploratory Data Analysis (EDA).</p>

<h3>1. Removing the Noise (The Long Tail)</h3>

<p>If you average every movie on IMDb, your data is polluted by home videos with 5 votes from the director&#39;s family.

In statistics, vote counts often follow a <a href="/vocab/power-law">&quot;Power Law&quot;</a> or long-tail distribution.</p>

<p>To analyze global sentiment, we had to filter out the noise. We set a threshold, dropping any movie with fewer than 100 votes.

This ensured our statistical analysis was based on titles with a minimum level of public engagement.</p>

<h3>2. Visualizing the Truth (The Box Plot)</h3>

<p>A simple average rating is misleading. If a genre has many <code>1/10</code>s and many <code>10/10</code>s, the average is <code>5/10</code> - but that doesn&#39;t tell the story of how polarizing it is.</p>

<p>I used a <a href="/vocab/box-plot">Box Plot</a> to visualize the distribution. It shows the median (the center line), the Interquartile Range (the colored box containing the middle 50% of data), and outliers (the dots).</p>

<p><img src="/images/posts/wgsiw/boxplot.png" alt="The Box Plot"></p>

<p><strong>Initial Observations:</strong></p>

<ul>

<li><strong>Documentary/Biography</strong>: High medians, compact boxes. They are consistently rated highly.</li>

<li><strong>Horror</strong>: The lowest median and a wide spread. It’s very easy to make a bad horror movie.</li>

</ul>

<h3>3. The Metrics: Weighted Ratings &amp; p99</h3>

<p>To get deeper insights, I needed better math than simple means.</p>

<h4>Metric A: The Weighted Rating (Bayesian Average)</h4>

<p>How do you compare a movie with a 9.0 rating and 105 votes against an 8.2 rating with 500,000 votes? The latter score is more statistically significant.</p>

<p>I adopted IMDb&#39;s own <strong>Weighted Rating</strong> formula. This &quot;Bayesian average&quot; pulls a movie&#39;s rating toward the global average (C) if it has few votes (v),

only allowing it to deviate as it gains more votes over a threshold (m).</p>

<p><img src="/images/posts/wgsiw/formula.png" alt="Weighted Rating"></p>

<p>This provided a fair &quot;Quality Score&quot; for every movie.</p>

<h4>Metric B: The p99 Ceiling</h4>

<p>I wanted to know the &quot;potential&quot; of a genre. Even if most Action movies are mediocre, how good are the very best ones?</p>

<p>For this, I calculated the <a href="/vocab/p99">99th Percentile (p99)</a> rating for each genre. This is the rating value below which 99% of the genre falls.

It represents the elite tier, the &quot;Masterpiece Ceiling.&quot;</p>

<h3>Part 4: The Deductions (The Gap Analysis)</h3>

<p>By combining the Average Weighted Rating (the typical experience) and the p99 Rating (the elite potential), we created a &quot;Gap Analysis&quot; chart.</p>

<p>The dark green bar is the average quality. The total height of the bar is the p99 ceiling. The light green area represents the &quot;Masterpiece Gap&quot;.</p>

<p><img src="/images/posts/wgsiw/gap.png" alt="Masterpiece Gap"></p>

<h4>The Data Science Deductions</h4>

<p>This single chart reveals the &quot;personality&quot; of every genre:</p>

<ol>

<li><p><strong>The &quot;Safe Bets&quot; (Documentary, History, Biography)</strong>:

They have very high averages (tall dark bars) and a small gap to the ceiling.

<em>Deduction</em>: It is difficult to make a poorly rated documentary. Audience selection bias likely plays a role here

(people only watch docs on topics they already like).</p>

</li>

<li><p><strong>The &quot;High Risk / High Reward&quot; (Horror, Sci-Fi)</strong>: They have the lowest averages (short dark bars),

indicating the typical output is poor. However, their p99 ceilings remain high.

<em>Deduction</em>: The gap is huge. It is incredibly difficult to execute these genres well, but when it&#39;s done right

(e.g., Alien, The Exorcist), they are revered just as highly as dramas.</p>

</li>

<li><p><strong>The Animation Anomaly</strong>: Animation has a high average and a very high ceiling.

<em>Deduction</em>: Statistically, this is perhaps the most consistently high-quality genre in modern cinema.</p>

</li>

</ol>

<h2>Conclusion</h2>

<p>This project demonstrated that with a solid engineering setup using modern tools like <code>uv</code>,

and by applying statistical concepts beyond simple averages, we can uncover nuanced truths hidden in raw data.

Averages tell you what is probable; distributions and percentiles tell you what is possible.</p>

<h3>Question A: Which genre is &quot;easier&quot; to make? (Action vs. Drama vs. Comedy)</h3>

<p><strong>The Data Verdict:</strong> It is significantly &quot;easier&quot; to make an acceptable <strong>Drama</strong> than an acceptable <strong>Action</strong> or <strong>Comedy</strong> movie.</p>

<ul>

<li><strong>Evidence:</strong> Look at the box plot, kindly.<ul>

<li><strong>Drama</strong> has a high median and a &quot;tight&quot; box (smaller Interquartile Range). This means even &quot;average&quot; Dramas are usually rated around 6.5–7.0. The &quot;floor&quot; is high.</li>

<li><strong>Action</strong> has a lower median. Action movies require budget, stunts, and effects. If those look cheap, the rating tanks immediately.

A bad drama is just &quot;boring&quot; (5/10); a bad action movie looks &quot;broken&quot; (3/10).</li>

<li><strong>Comedy</strong> is arguably the <em>hardest</em> to get a high rating for. Humor is subjective.

If a joke lands for 50% of the audience but annoys the other 50%, the rating averages out to a 5.0.

<strong>Drama is universal; Comedy is divisive</strong>.</li>

</ul>

</li>

</ul>

<h3>Question B: Should I use lower search bounds for Comedy compared to Drama?</h3>

<p><strong>The Data Verdict:</strong> <strong>YES. Absolutely.</strong></p>

<ul>

<li><strong>The &quot;Genre Inflation&quot; Factor:</strong> Users rate genres differently. A <strong>7.0</strong> in Horror or Comedy is effectively an <strong>8.0</strong> in Drama or Biography.<ul>

<li><strong>The Strategy:</strong> If you filter for <code>Rating &gt; 7.5</code>, you will see hundreds of Biographies, but you will filter out some of the funniest Comedies ever made (which often sit at 6.8 - 7.2).</li>

<li><strong>Action/Comedy Filter:</strong> Set your threshold to <strong>6.5</strong>.</li>

<li><strong>Drama/Doc Filter:</strong> Set your threshold to <strong>7.5</strong>.</li>

</ul>

</li>

</ul>

<h3>Question C: The &quot;Blindfold Test&quot; (Documentary vs. Sci-Fi)</h3>

<p><strong>The Data Verdict:</strong> You will be statistically safer picking the <strong>Documentary</strong>.</p>

<ul>

<li><p><strong>The &quot;Floor&quot; Concept:</strong> Look at the &quot;Whiskers&quot; (the lines extending from the boxes) on the box plot.</p>

<ul>

<li><strong>Sci-Fi:</strong> The bottom whisker goes deep down (towards 1.0 or 2.0). There is a significant statistical probability that a random Sci-Fi movie is unwatchable <em>garbage</em>.</li>

<li><strong>Documentary:</strong> The bottom whisker rarely dips below 5.0 or 6.0.</li>

</ul>

</li>

<li><p><strong>The Psychology:</strong></p>

<ul>

<li><strong>Documentaries</strong> are usually made by passionate experts about specific topics. They rarely &quot;fail&quot; completely.</li>

<li><strong>Sci-Fi</strong> is high-risk. It attempts to build new worlds. When that fails, it looks ridiculous, leading to &quot;hate-watching&quot; and 1-star reviews.</li>

<li><strong>Conclusion:</strong> If you are tired and just want a &quot;guaranteed decent watch&quot; (Low Variance), pick <strong>Documentary</strong>. If you want to gamble for a potentially mind-blowing experience (High Variance), pick <strong>Sci-Fi</strong>.</li>

</ul>

</li>

</ul>

<p>You can check the project here: <a href="https://github.com/fezcode/IMDbayes">IMDbayes</a></p>

<p><a href="https://fezcode.com/blog/what-genre-should-i-watch">Read more...</a></p>]]></content:encoded>

</item>

<item>

<title><![CDATA[Upgrading Debian 11 to 13: The Safe Path]]></title>

<description><![CDATA[[object Object]]]></description>

<link>https://fezcode.com/blog/debian-upgrade-path</link>

<guid isPermaLink="false">https://fezcode.com/blog/debian-upgrade-path</guid>

<dc:creator><![CDATA[Ahmed Samil Bulbul]]></dc:creator>

<pubDate>Mon, 12 Jan 2026 00:00:00 GMT</pubDate>

<content:encoded><![CDATA[<h1>Upgrading Debian 11 to 13: The Safe Path</h1>

<p>So, you&#39;re on Debian 11 (Bullseye) and want to jump to Debian 13 (Trixie). Maybe you saw some shiny new package, or you just want to be on the cutting edge (or as cutting edge as Debian gets).</p>

<p>But here&#39;s the catch: <strong>You can&#39;t skip a version.</strong></p>

<p>Debian upgrades are designed to be sequential. Jumping from 11 straight to 13 is a recipe for a broken system (frankstein packages, dependency hell, the works). The safe path is <strong>11 → 12 → 13</strong>.</p>

<p>Here is the quick gist of how to do it properly.</p>

<h2>Phase 1: Bullseye (11) to Bookworm (12)</h2>

<p>First, make sure your current system is fully updated and clean.</p>

<pre><code class="language-bash"># Clean up any broken sources first!

# If you have 404 errors on backports, comment them out in /etc/apt/sources.list

sudo apt update

sudo apt full-upgrade -y

</code></pre>

<p>Now, switch your sources to Bookworm.</p>

<pre><code class="language-bash">sudo sed -i &#39;s/bullseye/bookworm/g&#39; /etc/apt/sources.list

</code></pre>

<p>Run the upgrade. This is the big one.</p>

<pre><code class="language-bash">sudo apt update

sudo apt full-upgrade -y

</code></pre>

<p><strong>Reboot your system.</strong></p>

<h2>Phase 2: Bookworm (12) to Trixie (13)</h2>

<p>Welcome back. You are now on Debian 12. Let&#39;s keep going.</p>

<p>Update your sources to Trixie.</p>

<pre><code class="language-bash">sudo sed -i &#39;s/bookworm/trixie/g&#39; /etc/apt/sources.list

</code></pre>

<p>Run the upgrade again.</p>

<pre><code class="language-bash">sudo apt update

sudo apt full-upgrade -y

</code></pre>

<h2>Phase 3: Cleanup</h2>

<p>You made it. Now clean up the leftovers.</p>

<pre><code class="language-bash">sudo apt autoremove -y

sudo reboot

</code></pre>

<h2>Verification</h2>

<p>When you&#39;re back, check your version:</p>

<pre><code class="language-bash">cat /etc/debian_version

# Should output 13.x (or testing/trixie)

</code></pre>

<p>And that&#39;s it. You have successfully time traveled.</p>

<p><a href="https://fezcode.com/blog/debian-upgrade-path">Read more...</a></p>]]></content:encoded>

</item>

<item>

<title><![CDATA[Steganography: Hiding Secrets in Plain Sight with LSB]]></title>

<description><![CDATA[[object Object]]]></description>

<link>https://fezcode.com/blog/steganography-lsb-deep-dive</link>

<guid isPermaLink="false">https://fezcode.com/blog/steganography-lsb-deep-dive</guid>

<dc:creator><![CDATA[Ahmed Samil Bulbul]]></dc:creator>

<pubDate>Mon, 12 Jan 2026 00:00:00 GMT</pubDate>

<content:encoded><![CDATA[<p>Steganography is the art and science of hiding information within other non-secret data. Unlike cryptography, which scrambles a message so it can&#39;t be read, steganography hides the very existence of the message.</p>

<p>In this deep dive, we&#39;ll explore the implementation of the <strong>Steganography Tool</strong> added to Fezcodex, focusing on the <strong>Least Significant Bit (LSB)</strong> technique.</p>

<h2>The Core Concept: Least Significant Bit (LSB)</h2>

<p>Digital images are made up of pixels. In a standard 24-bit RGB image, each pixel has three color channels: Red, Green, and Blue. Each channel is represented by 8 bits (a value from 0 to 255).</p>

<p>Example of a pixel&#39;s color:</p>

<ul>

<li><strong>Red:</strong> 1011010<strong>1</strong> (181)</li>

<li><strong>Green:</strong> 0110011<strong>0</strong> (102)</li>

<li><strong>Blue:</strong> 1100101<strong>1</strong> (203)</li>

</ul>

<p>The <strong>Least Significant Bit</strong> is the rightmost bit in these binary strings. If we change this single bit, the decimal value of the color channel only changes by 1. For example, changing the Red channel from <code>10110101</code> (181) to <code>10110100</code> (180) is a change so subtle that the human eye cannot detect it in a complex image.</p>

<p>By replacing the LSB of each color channel with a bit from our secret message, we can embed data directly into the image.</p>

<h2>The Protocol: FEZ Steganography</h2>

<p>To make the extraction process reliable, we&#39;ve implemented a simple protocol:</p>

<ol>

<li><strong>Magic Header (<code>FEZ</code>):</strong> The first 24 bits (3 bytes) of the hidden data always spell &quot;FEZ&quot;. This allows the decoder to verify if an image actually contains a hidden message from our tool.</li>

<li><strong>Length (32-bit):</strong> The next 32 bits represent the length of the message in bytes. This tells the decoder exactly when to stop reading.</li>

<li><strong>The Message:</strong> The remaining bits are the actual UTF-8 encoded message.</li>

</ol>

<h3>Tracing the Magic: Encoding &quot;FEZ&quot;</h3>

<p>Let&#39;s look at how the magic header <code>FEZ</code> is scattered across the first few pixels.</p>

<p><strong>Step 1: Convert characters to binary</strong></p>

<ul>

<li><strong>F</strong> (70): <code>0 1 0 0 0 1 1 0</code></li>

<li><strong>E</strong> (69): <code>0 1 0 0 0 1 0 1</code></li>

<li><strong>Z</strong> (90): <code>0 1 0 1 1 0 1 0</code></li>

</ul>

<p>Combined Bitstream: <code>01000110</code> + <code>01000101</code> + <code>01011010</code> (24 bits total)</p>

<p><strong>Step 2: Embed into pixels</strong>

Since each pixel has 3 channels (R, G, B), we need 8 pixels to hide these 24 bits.</p>

<table>

<thead>

<tr>

<th align="left">Pixel</th>

<th align="left">Channel</th>

<th align="left">Original Byte</th>

<th align="left">Bit to Hide</th>

<th align="left">Modified Byte</th>

</tr>

</thead>

<tbody><tr>

<td align="left"><strong>Pixel 1</strong></td>

<td align="left">Red</td>

<td align="left"><code>1011010</code><strong><code>1</code></strong></td>

<td align="left"><strong>0</strong> (from F)</td>

<td align="left"><code>1011010</code><strong><code>0</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Green</td>

<td align="left"><code>0110011</code><strong><code>0</code></strong></td>

<td align="left"><strong>1</strong> (from F)</td>

<td align="left"><code>0110011</code><strong><code>1</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Blue</td>

<td align="left"><code>1100101</code><strong><code>1</code></strong></td>

<td align="left"><strong>0</strong> (from F)</td>

<td align="left"><code>1100101</code><strong><code>0</code></strong></td>

</tr>

<tr>

<td align="left"><strong>Pixel 2</strong></td>

<td align="left">Red</td>

<td align="left"><code>0101010</code><strong><code>0</code></strong></td>

<td align="left"><strong>0</strong> (from F)</td>

<td align="left"><code>0101010</code><strong><code>0</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Green</td>

<td align="left"><code>1111001</code><strong><code>1</code></strong></td>

<td align="left"><strong>0</strong> (from F)</td>

<td align="left"><code>1111001</code><strong><code>0</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Blue</td>

<td align="left"><code>0011001</code><strong><code>1</code></strong></td>

<td align="left"><strong>1</strong> (from F)</td>

<td align="left"><code>0011001</code><strong><code>1</code></strong></td>

</tr>

<tr>

<td align="left"><strong>Pixel 3</strong></td>

<td align="left">Red</td>

<td align="left"><code>1010101</code><strong><code>0</code></strong></td>

<td align="left"><strong>1</strong> (from F)</td>

<td align="left"><code>1010101</code><strong><code>1</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Green</td>

<td align="left"><code>1100110</code><strong><code>1</code></strong></td>

<td align="left"><strong>0</strong> (from F)</td>

<td align="left"><code>1100110</code><strong><code>0</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Blue</td>

<td align="left"><code>0001111</code><strong><code>0</code></strong></td>

<td align="left"><strong>0</strong> (from E)</td>

<td align="left"><code>0001111</code><strong><code>0</code></strong></td>

</tr>

<tr>

<td align="left"><strong>Pixel 4</strong></td>

<td align="left">Red</td>

<td align="left"><code>1011001</code><strong><code>0</code></strong></td>

<td align="left"><strong>1</strong> (from E)</td>

<td align="left"><code>1011001</code><strong><code>1</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Green</td>

<td align="left"><code>0110110</code><strong><code>1</code></strong></td>

<td align="left"><strong>0</strong> (from E)</td>

<td align="left"><code>0110110</code><strong><code>0</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Blue</td>

<td align="left"><code>1110001</code><strong><code>1</code></strong></td>

<td align="left"><strong>0</strong> (from E)</td>

<td align="left"><code>1110001</code><strong><code>0</code></strong></td>

</tr>

<tr>

<td align="left"><strong>Pixel 5</strong></td>

<td align="left">Red</td>

<td align="left"><code>0101010</code><strong><code>1</code></strong></td>

<td align="left"><strong>0</strong> (from E)</td>

<td align="left"><code>0101010</code><strong><code>0</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Green</td>

<td align="left"><code>1111001</code><strong><code>0</code></strong></td>

<td align="left"><strong>1</strong> (from E)</td>

<td align="left"><code>1111001</code><strong><code>1</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Blue</td>

<td align="left"><code>0011001</code><strong><code>1</code></strong></td>

<td align="left"><strong>0</strong> (from E)</td>

<td align="left"><code>0011001</code><strong><code>0</code></strong></td>

</tr>

<tr>

<td align="left"><strong>Pixel 6</strong></td>

<td align="left">Red</td>

<td align="left"><code>1010101</code><strong><code>0</code></strong></td>

<td align="left"><strong>1</strong> (from E)</td>

<td align="left"><code>1010101</code><strong><code>1</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Green</td>

<td align="left"><code>1100110</code><strong><code>1</code></strong></td>

<td align="left"><strong>0</strong> (from Z)</td>

<td align="left"><code>1100110</code><strong><code>0</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Blue</td>

<td align="left"><code>0101111</code><strong><code>0</code></strong></td>

<td align="left"><strong>1</strong> (from Z)</td>

<td align="left"><code>0101111</code><strong><code>1</code></strong></td>

</tr>

<tr>

<td align="left"><strong>Pixel 7</strong></td>

<td align="left">Red</td>

<td align="left"><code>1011001</code><strong><code>1</code></strong></td>

<td align="left"><strong>0</strong> (from Z)</td>

<td align="left"><code>1011001</code><strong><code>0</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Green</td>

<td align="left"><code>0110110</code><strong><code>0</code></strong></td>

<td align="left"><strong>1</strong> (from Z)</td>

<td align="left"><code>0110110</code><strong><code>1</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Blue</td>

<td align="left"><code>1110001</code><strong><code>1</code></strong></td>

<td align="left"><strong>1</strong> (from Z)</td>

<td align="left"><code>1110001</code><strong><code>1</code></strong></td>

</tr>

<tr>

<td align="left"><strong>Pixel 8</strong></td>

<td align="left">Red</td>

<td align="left"><code>0101010</code><strong><code>1</code></strong></td>

<td align="left"><strong>0</strong> (from Z)</td>

<td align="left"><code>0101010</code><strong><code>0</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Green</td>

<td align="left"><code>1111001</code><strong><code>0</code></strong></td>

<td align="left"><strong>1</strong> (from Z)</td>

<td align="left"><code>1111001</code><strong><code>1</code></strong></td>

</tr>

<tr>

<td align="left"></td>

<td align="left">Blue</td>

<td align="left"><code>0011001</code><strong><code>1</code></strong></td>

<td align="left"><strong>0</strong> (from Z)</td>

<td align="left"><code>0011001</code><strong><code>0</code></strong></td>

</tr>

</tbody></table>

<p>By the time we reach Pixel 8, all 24 bits of &quot;FEZ&quot; are woven into the image. If you open this in a hex editor, you might see that the color <code>181</code> became <code>180</code>, but the text &quot;FEZ&quot; is nowhere to be found in the raw bytes!</p>

<h3>Why PNG and not JPEG?</h3>

<p>Our tool works best with <strong>PNG</strong> files. Why?</p>

<ul>

<li><strong>PNG (Portable Network Graphics)</strong> is a <em>lossless</em> format. It preserves every single bit exactly as it was saved.</li>

<li><strong>JPEG (Joint Photographic Experts Group)</strong> is a <em>lossy</em> format. It uses compression algorithms that slightly alter pixel values to reduce file size. These tiny changes are fine for human viewing, but they destroy the data we&#39;ve hidden in the LSBs.</li>

</ul>

<h2>The Implementation (JavaScript/Canvas)</h2>

<p>We use the HTML5 <code>&lt;canvas&gt;</code> API to access and manipulate image data at the pixel level.</p>

<h3>Encoding Logic</h3>

<pre><code class="language-javascript">const imageData = ctx.getImageData(0, 0, canvas.width, canvas.height);

const data = imageData.data; // Uint8ClampedArray [R, G, B, A, R, G, B, A, ...]

// ... transform message to bits ...

let bitIndex = 0;

for (let i = 0; i &lt; data.length &amp;&amp; bitIndex &lt; allBits.length; i += 4) {

for (let j = 0; j &lt; 3 &amp;&amp; bitIndex &lt; allBits.length; j++) {

// Replace LSB of R, G, or B

// (data[i + j] &amp; 0xfe) clears the last bit

// | allBits[bitIndex++] sets it to our secret bit

data[i + j] = (data[i + j] &amp; 0xfe) | allBits[bitIndex++];

}

}

ctx.putImageData(imageData, 0, 0);

</code></pre>

<h3>Decoding Logic</h3>

<p>Decoding is the reverse process. We iterate through the pixels, extract the LSB of each R, G, and B channel, and rebuild the bitstream until we&#39;ve parsed the header, the length, and finally the message content.</p>

<h2>Challenges and Limitations</h2>

<ul>

<li><strong>Capacity:</strong> The amount of data you can hide depends on the image resolution. Each pixel can hold 3 bits (1 for each RGB channel). A 1080p image (1920x1080) can theoretically hold about 777 KB of hidden data.</li>

<li><strong>Robustness:</strong> LSB steganography is very fragile. Resizing, cropping, or re-saving the image as a JPEG will likely corrupt the hidden message.</li>

<li><strong>Security:</strong> Pure LSB is &quot;security through obscurity.&quot; Anyone who knows the technique can extract the message. For true security, you should encrypt the message <em>before</em> hiding it in the image.</li>

</ul>

<h2>Try it out!</h2>

<p>Check out the <strong>Steganography Tool</strong> in the Applications section and start sending your own cryptic signals through the digital aether.</p>

<p><a href="https://fezcode.com/blog/steganography-lsb-deep-dive">Read more...</a></p>]]></content:encoded>

</item>

<item>

<title><![CDATA[The Ultimate Pixel Art Resources Guide]]></title>

<description><![CDATA[[object Object]]]></description>

<link>https://fezcode.com/blog/pixel-art-resources-guide</link>

<guid isPermaLink="false">https://fezcode.com/blog/pixel-art-resources-guide</guid>

<dc:creator><![CDATA[Ahmed Samil Bulbul]]></dc:creator>

<pubDate>Mon, 12 Jan 2026 00:00:00 GMT</pubDate>

<content:encoded><![CDATA[<p>Pixel art is a beautiful and nostalgic medium that has seen a massive resurgence in recent years, especially within the indie game development scene. Whether you&#39;re a seasoned artist or just starting, having the right tools and resources can make a world of difference.</p>

<p>In this post, we&#39;ll explore some of the best resources for pixel art, inspired by the excellent guide by JuniperDev.</p>

<h3>The Ultimate Video Guide</h3>

<p>First and foremost, if you haven&#39;t seen it yet, check out this comprehensive video:

<a href="https://www.youtube.com/watch?v=DKmrBUpd0yw">The ONLY Pixel Art Guide You Need (Beginner to Advanced)</a></p>

<h3>Essential Software: Aseprite</h3>

<p>When it comes to creating pixel art, <strong>Aseprite</strong> is widely considered the industry standard. It&#39;s not just a drawing tool; it&#39;s a specialized environment for sprites and animation.</p>

<ul>

<li><strong>Why it&#39;s great:</strong> Pixel-perfect drawing mode, specialized animation timeline, easy sprite sheet exporting, and a very active community.</li>

<li><strong>Where to get it:</strong> <a href="https://www.aseprite.org/">aseprite.org</a></li>

</ul>

<h3>Mastering Color: Lospec</h3>

<p>Color is everything in pixel art. Since you&#39;re often working with a limited palette, choosing the right colors is crucial. <strong>Lospec</strong> is the go-to resource for this.</p>

<ul>

<li><strong>Palette Database:</strong> Thousands of pre-made palettes that you can filter by color count.</li>

<li><strong>Tools:</strong> They also offer web-based tools for checking contrast and creating your own palettes.</li>

<li><strong>Link:</strong> <a href="https://lospec.com/">lospec.com</a></li>

</ul>

<h3>Assets and Inspiration: itch.io</h3>

<p>Sometimes you need a head start, or you just want to see how other artists tackle specific challenges. <strong>itch.io</strong> is a goldmine for pixel art assets.</p>

<ul>

<li><strong>What you&#39;ll find:</strong> Character sprites, tilesets, UI elements, and full environmental packs.</li>

<li><strong>Learning from others:</strong> Analyzing high-quality asset packs is one of the best ways to learn techniques like dithering, sub-pixeling, and cluster management.</li>

<li><strong>Link:</strong> <a href="https://itch.io/game-assets/free/tag-pixel-art">itch.io/game-assets/free/tag-pixel-art</a></li>

</ul>

<h3>Quick Tips for Beginners</h3>

<ol>

<li><strong>Start Small:</strong> Don&#39;t try to draw a massive 256x256 piece right away. Start with 16x16 or 32x32.</li>

<li><strong>Limit Your Palette:</strong> Using too many colors can make your art look messy. Stick to 4-8 colors initially.</li>

<li><strong>Study Real Life:</strong> Even though it&#39;s stylized, good pixel art is often grounded in real-world lighting and anatomy.</li>

<li><strong>Practice Your Lines:</strong> Learn about &quot;jaggies&quot; and how to avoid them to keep your lines looking clean and intentional.</li>

</ol>

<p>Pixel art is as much about what you leave out as what you put in. Happy pixeling!</p>

<p><a href="https://fezcode.com/blog/pixel-art-resources-guide">Read more...</a></p>]]></content:encoded>

</item>

<item>

<title><![CDATA[Implementing Drag and Drop in React without Libraries]]></title>

<description><![CDATA[[object Object]]]></description>

<link>https://fezcode.com/blog/implementing-drag-and-drop-in-react</link>

<guid isPermaLink="false">https://fezcode.com/blog/implementing-drag-and-drop-in-react</guid>

<dc:creator><![CDATA[Ahmed Samil Bulbul]]></dc:creator>

<pubDate>Sat, 10 Jan 2026 00:00:00 GMT</pubDate>

<content:encoded><![CDATA[<p>When building <strong>Tier Forge</strong>, I needed a flexible way to move items between the &quot;pool&quot; and various &quot;tiers&quot;. While libraries like <code>react-beautiful-dnd</code> or <code>dnd-kit</code> are excellent, sometimes you just want full control without the overhead.</p>

<p>Here is how I implemented a robust drag-and-drop system using only the native HTML5 API and React state.</p>

<h2>The State Architecture</h2>

<p>The key to a good DnD system is centralized state. In <code>TierForge</code>, the state is held in the parent component:</p>

<pre><code class="language-javascript">const [tiers, setTiers] = useState(DEFAULT_TIERS); // The board

const [poolItems, setPoolItems] = useState([]); // The unranked items

const [dragData, setDragData] = useState(null); // What are we dragging?

</code></pre>

<p>We track <code>dragData</code> to know <em>what</em> is moving (<code>itemId</code>) and <em>where</em> it came from (<code>sourceId</code>).</p>

<h2>The Handlers</h2>

<p>We need three main handlers: <code>onDragStart</code>, <code>onDragOver</code>, and <code>onDrop</code>.</p>

<h3>1. Starting the Drag</h3>

<p>When a user grabs an item, we store its ID and source container ID. We also set <code>dataTransfer</code> for compatibility.</p>

<pre><code class="language-javascript">const handleDragStart = (e, itemId, sourceId) =&gt; {

setDragData({ itemId, sourceId });

e.dataTransfer.effectAllowed = &#39;move&#39;;

// Fallback for some browsers

e.dataTransfer.setData(&#39;text/plain&#39;, JSON.stringify({ itemId, sourceId }));

};

</code></pre>

<h3>2. Allowing the Drop</h3>

<p>By default, HTML elements don&#39;t accept drops. We must prevent the default behavior.</p>

<pre><code class="language-javascript">const handleDragOver = (e) =&gt; {

e.preventDefault();

e.dataTransfer.dropEffect = &#39;move&#39;;

};

</code></pre>

<h3>3. Handling the Drop</h3>

<p>This is where the magic happens. When an item is dropped, we:</p>

<ol>

<li>Identify the <strong>Source</strong> (where it came from) and <strong>Target</strong> (where it landed).</li>

<li>If Source === Target, do nothing (or reorder).</li>

<li>Find the item in the Source array.</li>

<li>Remove it from the Source.</li>

<li>Add it to the Target.</li>

</ol>

<pre><code class="language-javascript">const handleDrop = (e, targetId) =&gt; {

e.preventDefault();

const data = dragData || JSON.parse(e.dataTransfer.getData(&#39;text/plain&#39;));

if (!data) return;

const { itemId, sourceId } = data;

if (sourceId === targetId) return;

// ... Logic to find item, remove from source, add to target ...

// This involves setTiers() and setPoolItems() updates.

};

</code></pre>

<h2>The Components</h2>

<h3>Draggable Item</h3>

<p>The item itself needs the <code>draggable</code> attribute and the start handler.</p>

<pre><code class="language-jsx">&lt;div

draggable

onDragStart={(e) =&gt; onDragStart(e, item.id, sourceId)}

className=&quot;cursor-grab active:cursor-grabbing ...&quot;

&gt;

{/* Content */}

&lt;/div&gt;

</code></pre>

<h3>Drop Zone</h3>

<p>The container (Tier or Pool) listens for drag-over and drop events.</p>

<pre><code class="language-jsx">&lt;div

onDragOver={handleDragOver}

onDrop={(e) =&gt; handleDrop(e, containerId)}

className=&quot;...&quot;

&gt;

{/* Render Items */}

&lt;/div&gt;

</code></pre>

<h2>Why Native API?</h2>

<ol>

<li><strong>Zero Dependencies:</strong> Keeps the bundle size small.</li>

<li><strong>Full Control:</strong> I can define exactly how state updates happen.</li>

<li><strong>Performance:</strong> Direct DOM events are highly performant.</li>

</ol>

<p>This pattern powers the entire Tier Forge experience, allowing smooth transitions of assets between the chaotic pool and the structured tiers.</p>

<p><a href="https://fezcode.com/blog/implementing-drag-and-drop-in-react">Read more...</a></p>]]></content:encoded>

</item>

<item>

<title><![CDATA[Fixing gh-pages: Resolving spawn ENAMETOOLONG]]></title>

<description><![CDATA[[object Object]]]></description>

<link>https://fezcode.com/blog/gh-pages-enametoolong-fix</link>

<guid isPermaLink="false">https://fezcode.com/blog/gh-pages-enametoolong-fix</guid>

<dc:creator><![CDATA[Ahmed Samil Bulbul]]></dc:creator>

<pubDate>Thu, 08 Jan 2026 00:00:00 GMT</pubDate>

<content:encoded><![CDATA[<h1>Resolving <code>spawn ENAMETOOLONG</code> in <code>gh-pages</code> Deployment</h1>

<p>If you&#39;ve been using the <code>gh-pages</code> package for a while, especially in projects with large build folders or complex structures, you might have encountered the dreaded <code>spawn ENAMETOOLONG</code> error when running your deploy script.</p>

<h2>The Problem</h2>

<p>When executing the standard deployment command:</p>

<pre><code class="language-bash">gh-pages -d build -b gh-pages

</code></pre>

<p>The process fails with a system error indicating that the argument list or the command path itself has exceeded the operating system&#39;s limits. This is often related to how the underlying <code>globby</code> or <code>async</code> dependencies handle file lists in older versions of the package (like <code>6.3.0</code>).</p>

<p>The issue is documented and discussed in detail here: <a href="https://github.com/tschaub/gh-pages/issues/585">gh-pages Issue #585</a>.</p>

<h2>The Fix</h2>

<p>The specific fix for this issue was highlighted in <a href="https://github.com/tschaub/gh-pages/issues/585#issuecomment-2585457896">this GitHub comment</a>, which explains that the <code>ENAMETOOLONG</code> error occurs on Windows when the <code>rm</code> command receives an excessively long list of files as arguments.</p>

<pre><code class="language-git">diff --git a/lib/git.js b/lib/git.js

index d4c5724272d00bd1f0d76c47dab47d21ccd094d9..d86ac2b0bd7cbc02f34a50dac6980965102ee964 100644

--- a/lib/git.js

+++ b/lib/git.js

@@ -143,7 +143,7 @@ Git.prototype.rm = function (files) {

if (!Array.isArray(files)) {

files = [files];

}

- return this.exec(&#39;rm&#39;, &#39;--ignore-unmatch&#39;, &#39;-r&#39;, &#39;-f&#39;, &#39;--&#39;, ...files);

+ return this.exec(&#39;rm&#39;, &#39;--ignore-unmatch&#39;, &#39;-r&#39;, &#39;-f&#39;, &#39;--&#39;, &#39;.&#39;);

};

/**

</code></pre>

<p>The suggested workarounds included batching the file deletions or simplifying the command to target the current directory (<code>.</code>) instead of individual files. Fortunately, these improvements (including a more robust batching logic and a migration to <code>tinyglobby</code>) have already been merged into the main branch of the repository via PR #607.</p>

<p>While we wait for a stable release on NPM that fully addresses this in all environments, the most effective way to resolve it is to use the latest development version directly from the source.</p>

<p>By updating your <code>package.json</code> to point to the GitHub repository&#39;s main branch, you get the latest fixes (including the migration to <code>tinyglobby</code> and updated <code>commander</code> logic) that bypass these system limits.</p>

<h3>Implementation</h3>

<p>Update your <code>package.json</code> dependencies:</p>

<pre><code class="language-json">&quot;devDependencies&quot;: {

&quot;gh-pages&quot;: &quot;github:tschaub/gh-pages&quot;

}

</code></pre>

<p>Then, refresh your installations:</p>

<pre><code class="language-bash">npm install

</code></pre>

<p>This simple change allowed us to resume our production deployments without hitches, ensuring that our &quot;Brutalist&quot; digital garden stays fresh and accessible.</p>

<p><a href="https://fezcode.com/blog/gh-pages-enametoolong-fix">Read more...</a></p>]]></content:encoded>

</item>

<item>

<title><![CDATA[Git Cheatsheet: From Basics to Time Travel]]></title>

<description><![CDATA[[object Object]]]></description>

<link>https://fezcode.com/blog/git-cheatsheet-gist</link>

<guid isPermaLink="false">https://fezcode.com/blog/git-cheatsheet-gist</guid>

<dc:creator><![CDATA[Ahmed Samil Bulbul]]></dc:creator>

<pubDate>Thu, 08 Jan 2026 00:00:00 GMT</pubDate>

<content:encoded><![CDATA[<h1>Git Cheatsheet: From Basics to Time Travel</h1>

<p>A collection of essential Git commands, from daily workflows to digging through the depths of your repository&#39;s history.</p>

<h2>🔍 Searching History</h2>

<h3>Find when a file existed (even if deleted)</h3>

<pre><code class="language-bash">git log --all -- [path]

</code></pre>

<h3>Search for content changes (Pickaxe)</h3>

<p>Find commits where a specific string was added or removed:</p>

<pre><code class="language-bash">git log -S &quot;your_search_string&quot;

</code></pre>

<h3>Search content with Regex</h3>

<pre><code class="language-bash">git log -G &quot;your_regex&quot;

</code></pre>

<h3>Find a file in any commit/branch</h3>

<pre><code class="language-bash">git rev-list --all | xargs git grep -l &quot;filename&quot;

</code></pre>

<h3>See the history of a specific function/method</h3>

<pre><code class="language-bash">git log -L :function_name:file_path

</code></pre>

<hr>

<h2>🚀 Daily Workflow</h2>

<h3>Stage and Commit</h3>

<pre><code class="language-bash">git add .

git commit -m &quot;feat: descriptive message&quot;

</code></pre>

<h3>Undo last commit (keep changes)</h3>

<pre><code class="language-bash">git reset --soft HEAD~1

</code></pre>

<h3>Fix the last commit message</h3>

<pre><code class="language-bash">git commit --amend -m &quot;new message&quot;

</code></pre>

<hr>

<h2>🌿 Branching &amp; Merging</h2>

<h3>Switch to a new branch</h3>

<pre><code class="language-bash">git checkout -b feature/cool-stuff

# or the newer way:

git switch -c feature/cool-stuff

</code></pre>

<h3>List all branches (including remote)</h3>

<pre><code class="language-bash">git branch -a

</code></pre>

<h3>Safely delete a branch</h3>

<pre><code class="language-bash">git branch -d branch_name

</code></pre>

<hr>

<h2>🛠️ Cleanup &amp; Maintenance</h2>

<h3>Discard all local changes</h3>

<pre><code class="language-bash">git reset --hard HEAD

</code></pre>

<h3>Clean untracked files</h3>

<pre><code class="language-bash">git clean -fd

</code></pre>

<h3>Stash changes for later</h3>

<pre><code class="language-bash">git stash save &quot;Work in progress&quot;

git stash list

git stash pop

</code></pre>

<hr>

<h2>📤 Remote Operations</h2>

<h3>Update local with remote and rebase</h3>

<pre><code class="language-bash">git pull --rebase origin main

</code></pre>

<h3>Prune old remote tracking branches</h3>

<pre><code class="language-bash">git fetch -p

</code></pre>

<p><a href="https://fezcode.com/blog/git-cheatsheet-gist">Read more...</a></p>]]></content:encoded>

</item>

<item>

<title><![CDATA[Aether: Cyberpunk Audio Interface]]></title>

<description><![CDATA[[object Object]]]></description>

<link>https://fezcode.com/blog/aether-music-player</link>

<guid isPermaLink="false">https://fezcode.com/blog/aether-music-player</guid>

<dc:creator><![CDATA[Ahmed Samil Bulbul]]></dc:creator>

<pubDate>Tue, 06 Jan 2026 00:00:00 GMT</pubDate>

<content:encoded><![CDATA[<h1>Aether: The Cyberpunk Music Player</h1>

<p>I&#39;ve just deployed <strong>Aether</strong>, a new cloud-based music player for Fezcodex.</p>

<p><img src="/images/apps/aether.png" alt="Aether Music Player"></p>

<h2>Overview</h2>

<p>Aether isn&#39;t just a music player; it&#39;s an atmospheric audio interface designed to immerse you in the soundscape of the site. It features:</p>

<ul>

<li><strong>Cyberpunk Aesthetic:</strong> A high-contrast, terminal-inspired interface with CRT scanlines, glitch effects, and a generative art background that reacts to the music.</li>

<li><strong>Persistent Playback:</strong> A tiny, &quot;cyber deck&quot; style player docks to the bottom of your screen, allowing you to browse the site without interrupting your tunes.</li>

<li><strong>Generative Art:</strong> If a track lacks cover art, the system generates a unique visual signature based on the track&#39;s title.</li>

</ul>

<p>Check it out here: <a href="/apps/aether">Aether Music Player</a></p>

<p>Enjoy the vibes.</p>

<p><a href="https://fezcode.com/blog/aether-music-player">Read more...</a></p>]]></content:encoded>

</item>

<item>

<title><![CDATA[5 Ways to Pass Arguments in a URL (Beyond the Basic Query)]]></title>

<description><![CDATA[[object Object]]]></description>

<link>https://fezcode.com/blog/5-ways-to-pass-arguments-in-a-url</link>

<guid isPermaLink="false">https://fezcode.com/blog/5-ways-to-pass-arguments-in-a-url</guid>

<dc:creator><![CDATA[Ahmed Samil Bulbul]]></dc:creator>

<pubDate>Wed, 24 Dec 2025 00:00:00 GMT</pubDate>

<content:encoded><![CDATA[<p>When building web applications or designing APIs, understanding how to transfer data is crucial. While <strong>Query Parameters</strong> (the bits after the <code>?</code>) are the most common method, there are four other fundamental ways to pass arguments to a server via a URL or its associated HTTP request.</p>

<p>Here is a quick reference guide to the five main argument passing mechanisms:</p>

<h2>1. Query Parameters</h2>

<ul>

<li><strong>Location:</strong> Appears in the URL after a <strong>?</strong> (question mark) and separated by <strong>&amp;</strong> (ampersand) symbols.</li>

<li><strong>Purpose:</strong> Used for <strong>optional</strong> parameters such as filtering, sorting, searching, or pagination controls.</li>

<li><strong>Characteristics:</strong> Data is highly visible (in the URL, server logs, and browser history). It is typically used with <strong>GET</strong> requests.</li>

<li><strong>Example:</strong> <code>https://example.com/products?**category=1&amp;sort=price**</code></li>

</ul>

<h2>2. Path Parameters</h2>

<ul>

<li><strong>Location:</strong> Directly integrated into the URL&#39;s path structure.</li>

<li><strong>Purpose:</strong> Used to uniquely <strong>identify a specific resource</strong> or define a hierarchical location.</li>

<li><strong>Characteristics:</strong> Essential for defining clear, clean, and meaningful URLs, especially in RESTful API design.</li>

<li><strong>Example:</strong> <code>https://example.com/users/**123**</code> or <code>https://example.com/books/**sci-fi**/dune</code></li>

</ul>

<h2>3. Header Parameters</h2>

<ul>

<li><strong>Location:</strong> Contained within the <strong>HTTP Request Header</strong>, invisible in the URL.</li>

<li><strong>Purpose:</strong> Used for <strong>metadata</strong> about the request, such as authentication (e.g., API keys, tokens), content type, and language preferences.</li>

<li><strong>Characteristics:</strong> Offers better security for sensitive, non-data payload information compared to Query Parameters, as it doesn&#39;t appear in the URL.</li>

<li><strong>Example:</strong> <code>Header: **Authorization: Bearer token**</code> or <code>Header: **Content-Type: application/json**</code></li>

</ul>

<h2>4. Fragment Identifier Arguments</h2>

<ul>

<li><strong>Location:</strong> Appears at the very end of the URL after a <strong>#</strong> (hash symbol).</li>

<li><strong>Purpose:</strong> Used for client-side functionality, like navigating to a specific section (anchor) on a page or managing application state in Single Page Applications (SPAs).</li>

<li><strong>Characteristics:</strong> The browser <strong>does NOT send this part to the server</strong>; it is client-side only. It can still be used to pass data to the front-end application.</li>

<li><strong>Example:</strong> <code>https://example.com/page**#section-name**</code></li>

</ul>

<h2>5. Request Body Arguments</h2>

<ul>

<li><strong>Location:</strong> Contained within the <strong>body</strong> (payload) of the HTTP request, invisible in the URL.</li>

<li><strong>Purpose:</strong> Used for sending large data payloads when creating or updating resources (e.g., submitting a complex form, uploading a file, or sending a JSON object).</li>

<li><strong>Characteristics:</strong> The primary method for data submission using <strong>POST, PUT, or PATCH</strong> HTTP methods. It is an HTTP request argument, not a true URL argument, and it is secure from URL exposure.</li>

<li><strong>Example:</strong> <em>(Data like a user object in JSON format is sent in the hidden body payload.)</em></li>

</ul>

<hr>

<h2>Conclusion</h2>

<p>By strategically selecting among Query, Path, Header, Fragment, or Body arguments, developers can ensure their data is transferred efficiently and securely, leading to a robust and scalable application architecture.</p>

<p><a href="https://fezcode.com/blog/5-ways-to-pass-arguments-in-a-url">Read more...</a></p>]]></content:encoded>

</item>

<item>

<title><![CDATA[Quick Renaming .js to .jsx for React]]></title>

<description><![CDATA[[object Object]]]></description>

<link>https://fezcode.com/blog/renaming-js-to-jsx-gist</link>

<guid isPermaLink="false">https://fezcode.com/blog/renaming-js-to-jsx-gist</guid>

<dc:creator><![CDATA[Ahmed Samil Bulbul]]></dc:creator>

<pubDate>Tue, 23 Dec 2025 00:00:00 GMT</pubDate>

<content:encoded><![CDATA[<h2>The Command (PowerShell)</h2>

<p>I ran this specific script in your terminal. It uses &quot;piping&quot; (<code>|</code>) to pass the result of one command to the next, like a bucket brigade.</p>

<pre><code class="language-powershell">Get-ChildItem -Path src -Recurse -Filter *.js |

Where-Object { $_.Name -notin @(&#39;index.js&#39;, &#39;reportWebVitals.js&#39;, &#39;setupTests.js&#39;) } |

ForEach-Object {

if (Select-String -Pattern &quot;&lt;[a-zA-Z]&quot; -Path $_.FullName -Quiet) {

Write-Host &quot;Renaming $($_.Name) to .jsx&quot;;

Rename-Item -LiteralPath $_.FullName -NewName ($_.Name -replace &#39;\.js$&#39;, &#39;.jsx&#39;)

}

}

</code></pre>

<h3>Deep Dive: Commands &amp; Arguments</h3>

<p>Here is exactly what every part of that spell does:</p>

<p><strong>1. Finding the files</strong>

<code>Get-ChildItem -Path src -Recurse -Filter *.js</code></p>

<ul>

<li><code>Get-ChildItem</code>: The standard command to list files (like <code>ls</code> or <code>dir</code>).</li>

<li><code>-Path src</code>: We only look inside the <code>src</code> folder.</li>

<li><code>-Recurse</code>: We dig deep into every subfolder, not just the top level.</li>

<li><code>-Filter *.js</code>: We ignore everything except files ending in <code>.js</code>.</li>

</ul>

<p><strong>2. Filtering the list</strong>

<code>Where-Object { $_.Name -notin @(...) }</code></p>

<ul>

<li><code>Where-Object</code>: Acts like a bouncer; only lets items through that match the condition.</li>

<li><code>$_</code>: Represents the &quot;current file&quot; being checked.</li>

<li><code>-notin</code>: The condition operator. We are saying &quot;The name must NOT be in this list&quot;.</li>

<li><code>@(&#39;index.js&#39;, ...)</code>: The list of system files we want to skip (leave as <code>.js</code>).</li>

</ul>

<p><strong>3. Processing each file</strong>

<code>ForEach-Object { ... }</code></p>

<ul>

<li><code>ForEach-Object</code>: Runs the code block inside <code>{ ... }</code> for every single file that made it past the filter.</li>

</ul>

<p><strong>4. Checking for React Code (JSX)</strong>

<code>if (Select-String -Pattern &quot;&lt;[a-zA-Z]&quot; -Path $_.FullName -Quiet)</code></p>

<ul>

<li><code>Select-String</code>: Searches for text inside a file (like <code>grep</code>).</li>

<li><code>-Pattern &quot;&lt;[a-zA-Z]&quot;</code>: A Regex pattern. It looks for a <code>&lt;</code> followed by a letter. This catches HTML tags like <code>&lt;div&gt;</code> or React components like <code>&lt;App&gt;</code>.</li>

<li><code>-Path $_.FullName</code>: The full path to the file we are currently reading.</li>

<li><code>-Quiet</code>: Important! This tells the command &quot;Don&#39;t print the matching text, just tell me True or False.&quot;</li>

</ul>

<p><strong>5. Renaming the file</strong>

<code>Rename-Item -LiteralPath $_.FullName -NewName (...)</code></p>

<ul>

<li><code>Rename-Item</code>: The command to change a file&#39;s name.</li>

<li><code>-LiteralPath $_.FullName</code>: We use the full path to ensure we target the exact right file.</li>

<li><code>-NewName ($_.Name -replace &#39;\.js$&#39;, &#39;.jsx&#39;)</code>: We calculate the new name by taking the old name and swapping the ending <code>.js</code> with <code>.jsx</code>.</li>

</ul>

<p><strong>The Result</strong></p>

<p>Now your code editor knows exactly which files contain UI components. You&#39;ll get better autocomplete, better color highlighting, and generally a much happier development experience.</p>

<p><a href="https://fezcode.com/blog/renaming-js-to-jsx-gist">Read more...</a></p>]]></content:encoded>

</item>

<item>

<title><![CDATA[The Corrupted Blood Incident: When a Glitch Taught Us About Pandemics]]></title>

<description><![CDATA[[object Object]]]></description>

<link>https://fezcode.com/blog/corrupted-blood-incident</link>

<guid isPermaLink="false">https://fezcode.com/blog/corrupted-blood-incident</guid>

<dc:creator><![CDATA[Ahmed Samil Bulbul]]></dc:creator>

<pubDate>Sun, 21 Dec 2025 00:00:00 GMT</pubDate>

<content:encoded><![CDATA[<p><em>Okay, gather &#39;round</em>, fellow nerds and accidental epidemiologists.

I need to talk about something that happened in <em>World of Warcraft</em> back in 2005.

It&#39;s called the <strong>Corrupted Blood incident</strong>, and it&#39;s basically the coolest (and most terrifying) accidental science experiment in gaming history.</p>

<iframe width="560" height="315" src="https://www.youtube.com/embed/vj-mz5piYYY?si=RlcxH-t2Q7-2bYIJ" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

<h2>ELI5: What the Heck Happened?</h2>

<p>Imagine you&#39;re playing a game, right? You and your 19 closest friends decide to go punch a giant blood god named <strong>Hakkar the Soulflayer</strong> in the face. This raid boss has a nasty spell called &quot;Corrupted Blood.&quot;</p>

<p>Here&#39;s how it worked:</p>

<ol>

<li><strong>You catch it:</strong> It drains your health. Fast.</li>

<li><strong>It spreads:</strong> If you stand near anyone else, they catch it too. Like a super-flu.</li>

<li><strong>It&#39;s meant for the boss room:</strong> The disease was programmed to disappear when you died or left the dungeon.</li>

</ol>

<p><strong>BUT HERE&#39;S THE GLITCH.</strong></p>

<p>Hunter pets (animals that players control) could catch the disease. If a player dismissed their pet while it was sick, the game &quot;froze&quot; the pet&#39;s state.

When they summoned the pet back in a major city (like Ironforge or Orgrimmar), the pet came back... <em>still sick</em>.</p>

<p><em>Boom. Patient Zero.</em></p>

<h2>The Virtual Apocalypse</h2>

<p>Suddenly, high-level players&#39; pets were nuking entire cities. Low-level players (newbies) were dropping dead instantly just by walking past the auction house.

High-level players were scrambling to keep themselves alive, healing frantically.</p>

<p>It was chaos.</p>

<ul>

<li><strong>The Cities:</strong> Zones of death. Skeletons everywhere.</li>

<li><strong>The Players:</strong> Panic. Some fled to the wilderness (social distancing!). Some deliberately spread it (trolls/bioterrorists). Healers tried to set up triage centers.</li>

<li><strong>Blizzard (The Devs):</strong> They tried quarantines. Failed. They tried warnings. Failed. Eventually, they had to do a hard server reset to scrub the disease from existence.</li>

</ul>

<h2>Why Real Scientists Cared</h2>

<p>Here&#39;s the wild part. Real-world epidemiologists (the doctors who study diseases) looked at this and went, <em>&quot;Holy crap, this is better than our computer models.&quot;</em></p>

<p>Usually, scientific models assume people act rationally. &quot;If there is a plague, people will stay home.&quot; But in <em>WoW</em>, people did <strong>human</strong> things:</p>

<ul>

<li><strong>Curiosity:</strong> &quot;What&#39;s happening over there?&quot; -&gt; <em>Dies.</em></li>

<li><strong>Malice:</strong> &quot;Imma go infect the newbies lol.&quot; -&gt; <em>Spreads plague.</em></li>

<li><strong>Altruism:</strong> &quot;I&#39;ll heal you!&quot; -&gt; <em>Gets infected, spreads it further.</em></li>

</ul>

<p>This accidental glitch provided a perfect, unscripted look at human behavior during a crisis. It showed how fast things spread when people don&#39;t follow rules,

how asymptomatic carriers (pets/high-level players) can destroy vulnerable populations (low-level players), and how hard it is to contain stupidity.</p>

<h2>The GDC Legacy</h2>

<p>This wasn&#39;t just a &quot;remember when&quot; moment. It became a serious case study. At <strong>GDC (Game Developers Conference)</strong>,

this incident is often cited as a prime example of emergent gameplay and complex systems gone wrong (or right, depending on your view).</p>

<p>It taught developers that players will <em>always</em> find a way to break containment.

It taught scientists that &quot;Gamer Behavior&quot; might actually be a decent proxy for &quot;Human Panic.&quot;</p>

<h2>The Rant</h2>

<p>It drives me crazy that we had this perfect simulation in 2005, and when 2020 rolled around, we saw the <em>exact same behaviors</em> IRL.

The deniers, the spreaders, the people fleeing to the countryside. We didn&#39;t learn! We leveled up, but we didn&#39;t put any points into Wisdom!</p>

<p><strong>TL;DR:</strong> A coding bug in a fantasy game predicted modern pandemic behavior better than some government models.

Hakkar the Soulflayer is the ultimate teacher. Wash your hands, dismiss your pets responsibly, and for the love of Azeroth, stop standing in the fire.</p>

<p><a href="https://fezcode.com/blog/corrupted-blood-incident">Read more...</a></p>]]></content:encoded>

</item>

<item>

<title><![CDATA[Building the Knowledge Graph: Visualizing Fezcodex in 3D]]></title>

<description><![CDATA[[object Object]]]></description>

<link>https://fezcode.com/blog/building-the-knowledge-graph</link>

<guid isPermaLink="false">https://fezcode.com/blog/building-the-knowledge-graph</guid>

<dc:creator><![CDATA[Ahmed Samil Bulbul]]></dc:creator>

<pubDate>Sun, 21 Dec 2025 00:00:00 GMT</pubDate>

<content:encoded><![CDATA[<p>The idea was simple: Fezcodex is growing. With hundreds of blog posts, apps, and project logs, a standard list or grid view just wasn&#39;t cutting it anymore. I wanted a way to visualize the <em>connections</em> between everything. To see the &quot;brain&quot; of the website.</p>

<p>Enter the <strong>Knowledge Graph Visualization Protocol</strong>.</p>

<h2>The Concept</h2>

<p>I wanted a 3D, interactive network graph where:</p>

<ul>

<li><strong>Nodes</strong> represent content (Blog Posts, Apps, Projects).</li>

<li><strong>Links</strong> represent relationships (Shared tags, Categories).</li>

<li><strong>Interaction</strong> allows users to fly through the data and navigate to content.</li>

</ul>

<p>It needed to feel like a &quot;cyberspace&quot; visualization from a sci-fi movie—immersive, dark, and slightly chaotic but organized.</p>

<h2>The Tech Stack</h2>

<ul>

<li><strong>React</strong>: The core framework.</li>

<li><strong>react-force-graph-3d</strong>: The heavy lifter. This library uses WebGL (via Three.js) to render force-directed graphs with great performance.</li>

<li><strong>PIML</strong>: My custom markup language for parsing project data.</li>

<li><strong>Tailwind CSS</strong>: For the overlay UI and brutalist styling.</li>

</ul>

<h2>Implementation Details</h2>

<h3>1. Data Extraction (<code>graphDataManager.js</code>)</h3>

<p>The first challenge was aggregating data from three different sources:</p>

<ul>

<li><code>posts.json</code>: A static JSON file containing blog metadata.</li>

<li><code>apps.json</code>: A structured list of all the mini-apps.</li>

<li><code>projects.piml</code>: A custom file format for my project portfolio.</li>

</ul>

<p>I created a utility function <code>fetchGraphData</code> that pulls all three.</p>

<pre><code class="language-javascript">export const fetchGraphData = async () =&gt; {

const nodes = [];

const links = [];

const tagMap = new Map();

// ... fetching logic ...

</code></pre>

<p>For each item, I created a <strong>primary node</strong>. Then, I looked at its <code>tags</code>, <code>category</code>, or <code>technologies</code>. For every tag found, I created a <strong>tag node</strong> (if it didn&#39;t exist) and created a <strong>link</strong> between the item and the tag.</p>

<p>This automatically creates clusters. If five posts are tagged &quot;React&quot;, they all link to the &quot;React&quot; tag node, pulling them together in the 3D space.</p>

<h3>2. The 3D Component (<code>KnowledgeGraphPage.js</code>)</h3>

<p>I used <code>&lt;ForceGraph3D&gt;</code> to render the data.</p>

<pre><code class="language-jsx">&lt;ForceGraph3D

ref={fgRef}

graphData={graphData}

backgroundColor=&quot;#050505&quot;

nodeLabel=&quot;name&quot;

nodeColor=&quot;color&quot;

onNodeClick={handleNodeClick}

// ...

/&gt;

</code></pre>

<h3>3. Camera Controls</h3>

<p>The &quot;cool factor&quot; comes from the camera movement. When you click a node, I didn&#39;t want a hard jump. I wanted a smooth flight.</p>

<pre><code class="language-javascript"> const handleNodeClick = useCallback((node) =&gt; {

// Calculate a position slightly &quot;outside&quot; the node

const distance = 40;

const distRatio = 1 + distance/Math.hypot(node.x, node.y, node.z);

if (fgRef.current) {

fgRef.current.cameraPosition(

{ x: node.x * distRatio, y: node.y * distRatio, z: node.z * distRatio }, // new pos

node, // lookAt

3000 // ms duration

);

}

}, []);

</code></pre>

<p>This calculates a vector from the center (0,0,0) to the node, extends it by a fixed distance, and moves the camera there while focusing on the node.</p>

<h2>Challenges</h2>

<ul>

<li><strong>PIML Parsing</strong>: My custom language parser needed to be robust enough to handle the varying structures of the <code>projects.piml</code> file.</li>

<li><strong>Performance</strong>: Rendering hundreds of text labels in 3D can be heavy. I kept the UI minimal and only showing detailed info on hover.</li>

<li><strong>Theme</strong>: Matching the &quot;Brutalist/Cyberpunk&quot; aesthetic required careful tuning of node colors (Emerald for Apps, Red for Posts) and link opacity.</li>

</ul>

<h2>The Result</h2>

<p>The result is a living, breathing map of Fezcodex. It reveals patterns I didn&#39;t explicitly plan—like the massive cluster around &quot;React&quot; or the isolated islands of specific game experiments. It&#39;s not just a navigation tool; it&#39;s a piece of generative art powered by my own work.</p>

<p>Go check it out at <code>/graph</code> and fly through the system.</p>

<p><a href="https://fezcode.com/blog/building-the-knowledge-graph">Read more...</a></p>]]></content:encoded>

</item>

<item>

<title><![CDATA[Routing Revolution: SSG, BrowserRouter, and the SEO Fix]]></title>

<description><![CDATA[[object Object]]]></description>

<link>https://fezcode.com/blog/routing-revolution-ssg-and-seo</link>

<guid isPermaLink="false">https://fezcode.com/blog/routing-revolution-ssg-and-seo</guid>

<dc:creator><![CDATA[Ahmed Samil Bulbul]]></dc:creator>

<pubDate>Sun, 21 Dec 2025 00:00:00 GMT</pubDate>

<content:encoded><![CDATA[<h1>Routing Revolution: SSG, BrowserRouter, and the SEO Fix</h1>

<p>For a long time, Fezcodex lived behind the &quot;Hash Gap.&quot; If you looked at your address bar, you’d see that familiar <code>/#/</code> slicing through every URL. While functional, this was the primary reason social media thumbnails were failing and search engines were only seeing the home page.</p>

<p>Today, I’ve completed a total migration to <strong><a href="/vocab/spa">BrowserRouter</a></strong> combined with <strong><a href="/vocab/ssg">SSG (Static Site Generation)</a></strong>. Here is the technical breakdown of why this was necessary and how it works.</p>

<h2>The Problem: The Hash Black Hole</h2>

<p>We originally used <code>HashRouter</code> because Fezcodex is hosted on GitHub Pages. Since GitHub Pages is a static file host, it doesn&#39;t know how to handle a request for <code>/apps/markdown-table-formatter</code>. It looks for a folder named <code>apps</code> and an <code>index.html</code> inside it. When it doesn&#39;t find them, it throws a 404.</p>

<p><code>HashRouter</code> solved this by putting everything after the <code>#</code>. The server ignores the hash, always serves the root <code>index.html</code>, and React handles the rest. </p>

<p><strong>The SEO Cost:</strong> Most <strong><a href="/vocab/crawler">crawlers</a></strong> (Twitter, Facebook, Discord) do not execute JavaScript and ignore the hash entirely. To them, every single link you shared looked like <code>fezcode.com/</code>—resulting in generic &quot;Fezcodex - Personal Blog&quot; thumbnails instead of page-specific content.</p>