<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Ritik Patni</title>
    <description>The latest articles on DEV Community by Ritik Patni (@ritikpatni).</description>
    <link>https://dev.to/ritikpatni</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fthepracticaldev.s3.amazonaws.com%2Fi%2F99mvlsfu5tfj9m7ku25d.png</url>
      <title>DEV Community: Ritik Patni</title>
      <link>https://dev.to/ritikpatni</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/ritikpatni"/>
    <language>en</language>
    <item>
      <title>2026 · Week 44</title>
      <dc:creator>Ritik Patni</dc:creator>
      <pubDate>Wed, 04 Nov 2026 03:30:02 +0000</pubDate>
      <link>https://dev.to/ritikpatni/2026-week-44-4i09</link>
      <guid>https://dev.to/ritikpatni/2026-week-44-4i09</guid>
      <description>&lt;p&gt;Hello! This week’s reading keeps circling back to a question: when software gets cheaper to produce, what becomes expensive?&lt;/p&gt;

&lt;p&gt;There’s plenty here about AI and engineering, a look at where CSS could go next, and a bird-sticker collection for a change of scenery.&lt;/p&gt;

&lt;h2&gt;
  
  
  Blogs &amp;amp; essays
&lt;/h2&gt;

&lt;h3&gt;
  
  
  &lt;a href="https://hrih.si/articles/the-compiler-was-supposed-to-end-programming/" rel="noopener noreferrer"&gt;The compiler was supposed to end programming too&lt;/a&gt; · Hatim Rih
&lt;/h3&gt;

&lt;p&gt;In 1954, IBM suggested FORTRAN would virtually eliminate coding and debugging. Debugging appears to have missed the memo.&lt;/p&gt;

&lt;p&gt;Hatim traces that promise through successive generations of programming tools. His argument is that cheaper software makes more problems worth solving, including the ones nobody could previously justify spending a team’s time on.&lt;/p&gt;

&lt;p&gt;The useful tension is between adoption and trust. Developers can rely on AI tools while remaining frustrated by output that is almost correct. The article’s imagined developer workday in 2031 brings that into focus: agents handle implementation, while a person has to decide what a refund policy should do when the customer’s original card has expired. Someone still has to answer the question before the code can be right.&lt;/p&gt;

&lt;h3&gt;
  
  
  &lt;a href="https://hrih.si/articles/the-internets-guano-years/" rel="noopener noreferrer"&gt;The internet is in its guano years&lt;/a&gt; · Hatim Rih
&lt;/h3&gt;

&lt;p&gt;The title takes a little explaining. Bird droppings accumulated for centuries before becoming a valuable commodity. Hatim uses that history to examine the posts, answers, and conversations that accumulated on internet platforms before becoming valuable training data.&lt;/p&gt;

&lt;p&gt;The parallel gives the essay its shape, but the question of ownership is what stays with you. People contribute the writing; platform owners control access and negotiate its value.&lt;/p&gt;

&lt;p&gt;One passage in the week’s saved notes captures the concern:&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;“If your words only live on someone else’s platform, they’re part of someone else’s deposit, and the owner decides what they’re worth and who gets to read them.”&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;The practical suggestion is modest: keep an export of your work somewhere you control. You don’t have to abandon every platform to want a copy of your own words.&lt;/p&gt;

&lt;h3&gt;
  
  
  &lt;a href="https://hrih.si/articles/every-architecture-is-a-receipt/" rel="noopener noreferrer"&gt;Every architecture is a receipt&lt;/a&gt; · Hatim Rih
&lt;/h3&gt;

&lt;p&gt;An architecture records what its designers were short of: memory, machine time, coordination, or attention. The trouble starts when we copy the solution without checking whether we have the same constraint.&lt;/p&gt;

&lt;p&gt;Hatim opens with his own agent-assisted project. Agents write most of the code, but he has to review it, so his architecture is deliberately small enough to keep that review manageable.&lt;/p&gt;

&lt;p&gt;That makes the historical examples useful beyond nostalgia. A decision can have been sensible at the time and still deserve revisiting once its costs change.&lt;/p&gt;

&lt;p&gt;The article ends with a small addition to architecture decision records: write down what is scarce and when to reconsider the decision. That would give the next person considerably more to work with than “this is how we’ve always done it.”&lt;/p&gt;

&lt;h3&gt;
  
  
  &lt;a href="https://www.joshwcomeau.com/css/interop-2027/" rel="noopener noreferrer"&gt;The Future of CSS ✨&lt;/a&gt; · Josh W. Comeau
&lt;/h3&gt;

&lt;p&gt;Josh walks through his ten picks for Interop 2027, including CSS functions and new ways to shape corners and borders. These are his requested focus areas, not a promise that every feature will become universally available next year.&lt;/p&gt;

&lt;p&gt;The examples make this worth opening in a browser. With &lt;code&gt;corner-shape&lt;/code&gt;, for instance, the altered shape can be reflected in borders, outlines, and shadows, which makes it more useful than simply clipping a rectangle into something interesting.&lt;/p&gt;

&lt;p&gt;Josh also makes a point that connects neatly with the earlier engineering reads: knowing modern CSS still matters when you work with coding agents. You need enough familiarity with the platform to recognise when an agent reaches for an old workaround that the browser no longer needs.&lt;/p&gt;

&lt;h2&gt;
  
  
  Resources
&lt;/h2&gt;

&lt;h3&gt;
  
  
  &lt;a href="https://github.com/alchaincyf/huashu-bird-sticker" rel="noopener noreferrer"&gt;Huashu bird stickers&lt;/a&gt; · Huashu / alchaincyf
&lt;/h3&gt;

&lt;p&gt;A collection of bird illustrations covering the 11,064 living species in the project’s chosen taxonomy, accompanied by structured species data and names in up to 43 languages.&lt;/p&gt;

&lt;p&gt;There’s more here than a folder of pictures. The repository includes taxonomy and cross-database identifiers, making it worth exploring for visual projects that need the information behind the illustration too.&lt;/p&gt;

&lt;p&gt;One detail to check before reusing it: the data and artwork have different licensing terms. The stickers restrict competing birding products, and the README flags additional questions around reference-photo rights. A public repository is not the same thing as unrestricted artwork.&lt;/p&gt;

</description>
      <category>css</category>
      <category>programming</category>
      <category>webdev</category>
    </item>
  </channel>
</rss>
