<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Writing · Josh Spratt</title><description>Posts by Josh Spratt</description><link>https://jspratt.dev/</link><item><title>My finished side projects are all dead</title><link>https://jspratt.dev/writing/the-graveyard/</link><guid isPermaLink="true">https://jspratt.dev/writing/the-graveyard/</guid><description>Every project I’ve shipped lately is dead. That doesn’t mean I didn’t finish them.</description><pubDate>Fri, 13 Nov 2026 00:00:00 GMT</pubDate><content:encoded>&lt;p&gt;“Why do I never finish projects?”&lt;/p&gt;
&lt;p&gt;I sat down and did a bit of soul searching and looked at what I’d built over the last few months:&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;&lt;code&gt;twit.rs&lt;/code&gt;, a Twitter clone written in Rust&lt;/li&gt;
&lt;li&gt;&lt;code&gt;spill&lt;/code&gt;, a web app for family interaction (where you spill your guts anonymously about a fun/silly topic each day)&lt;/li&gt;
&lt;li&gt;&lt;code&gt;doorchime&lt;/code&gt;, an AOL Instant Messenger clone&lt;/li&gt;
&lt;li&gt;A historically accurate Titanic sinking simulation game&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;They were all fun to build. Every single one of them is dead now. I built them, hosted them, and then could not get anyone to try them out. But here’s the thing: I did finish them. Every one of them shipped.&lt;/p&gt;
&lt;figure&gt;
  &lt;img src=&quot;https://jspratt.dev/writing/the-graveyard/twit-rs.webp&quot; width=&quot;1600&quot; height=&quot;1088&quot; loading=&quot;lazy&quot; alt=&quot;twit.rs: a green-on-black terminal-style Twitter timeline, with my own post at the top: “shipped a twitter clone in rust. now i just need someone to tweet at.”&quot; /&gt;
  &lt;figcaption&gt;&lt;code&gt;twit.rs&lt;/code&gt;&lt;/figcaption&gt;
&lt;/figure&gt;
&lt;figure&gt;
  &lt;img src=&quot;https://jspratt.dev/writing/the-graveyard/spill.webp&quot; width=&quot;1160&quot; height=&quot;1000&quot; loading=&quot;lazy&quot; alt=&quot;Spill: a newspaper-style page titled The Daily Spill, with today’s prompt “What made you laugh today?” and four family members’ answers revealed.&quot; /&gt;
  &lt;figcaption&gt;&lt;code&gt;spill&lt;/code&gt;&lt;/figcaption&gt;
&lt;/figure&gt;
&lt;figure&gt;
  &lt;img src=&quot;https://jspratt.dev/writing/the-graveyard/doorchime.webp&quot; width=&quot;1600&quot; height=&quot;902&quot; loading=&quot;lazy&quot; alt=&quot;Doorchime: a retro desktop with a buddy list, one buddy away at band practice, and an open chat window where a friend asks “omg is this aim”.&quot; /&gt;
  &lt;figcaption&gt;&lt;code&gt;doorchime&lt;/code&gt;&lt;/figcaption&gt;
&lt;/figure&gt;
&lt;figure&gt;
  &lt;img src=&quot;https://jspratt.dev/writing/the-graveyard/twenty-boats.webp&quot; width=&quot;1600&quot; height=&quot;1000&quot; loading=&quot;lazy&quot; alt=&quot;Twenty Boats: the Titanic at 01:25, listing by the bow, with a deck plan of every lifeboat’s status and the starboard boats loading under Murdoch.&quot; /&gt;
  &lt;figcaption&gt;The Titanic game, Twenty Boats&lt;/figcaption&gt;
&lt;/figure&gt;
&lt;p&gt;If you read &lt;a href=&quot;https://jspratt.dev/writing/the-lego-set-that-builds-itself&quot;&gt;my last post&lt;/a&gt;, you know I struggle to get the same fulfillment out of building with AI. But AI is the reason most of these projects got finished at all. Back in the old days, they would have died a half-baked repo on my GitHub profile. These projects still got built. We started with an idea, and shipped a real working version of that idea. That counts for something.&lt;/p&gt;
&lt;p&gt;Building used to be the hard part. It was also the fun part. Now, building is fast and easy. Most of the fun is gone. Finding the right thing to build is the new hard part.&lt;/p&gt;
&lt;p&gt;I loved building the Titanic game. I loved playing it too. I’m a huge history and shipwrecks nerd. But you can only play your own game so many times. If I can’t get anyone else interested in it, what’s the point? I used to feel rewarded twice: the work, and then the people using it. AI took the first. Nobody showed up for the second.&lt;/p&gt;
&lt;p&gt;So, where does this leave me? What were my takeaways?&lt;/p&gt;
&lt;ol&gt;
&lt;li&gt;Building something and then tearing it down is not a failure. You still had a fun idea and brought it to life. That’s worth being proud of.&lt;/li&gt;
&lt;li&gt;The hard part has shifted (or maybe always was) to identifying the right thing to build next, and how to get anyone to care. I want to get better at that, and find a way to enjoy it as much as I used to enjoy the building.&lt;/li&gt;
&lt;/ol&gt;
&lt;p&gt;Everyone has a project graveyard. What’s in yours?&lt;/p&gt;
</content:encoded></item><item><title>The Lego set that builds itself</title><link>https://jspratt.dev/writing/the-lego-set-that-builds-itself/</link><guid isPermaLink="true">https://jspratt.dev/writing/the-lego-set-that-builds-itself/</guid><description>Why I&apos;m writing, for every engineer who misses building things by hand.</description><pubDate>Sun, 08 Nov 2026 00:00:00 GMT</pubDate><content:encoded>&lt;p&gt;Engineering has never felt more isolating.&lt;/p&gt;
&lt;p&gt;Think back to what first got you into this field. For a lot of us, it was being a bright-eyed kid building something out of a pile of Lego bricks. That feeling lit a fire, a passion to build.&lt;/p&gt;
&lt;p&gt;Landing in a career where you got paid to build things and solve hard problems was wonderful. That’s exactly why adapting to AI is hard for some of us. Are we still building? Of course. But is a Lego set that builds itself as fun as one you designed and assembled by hand? No.&lt;/p&gt;
&lt;p&gt;Add the pressure from executives, and even tech leaders, to go all in on AI. (“Why aren’t you a 10x developer yet?”) And God forbid your algorithm picks up on any of this on X or TikTok. You’ll be living in AI-psychosis doom and gloom for the rest of your days.&lt;/p&gt;
&lt;p&gt;Here’s the isolating part. Most of the people around me seem unfazed. I felt like I was just getting started, like I was finally becoming a good engineer. Now it feels like that doesn’t matter anymore. And when I try to explain this to leaders, they just don’t get it.&lt;/p&gt;
&lt;p&gt;Don’t get me wrong. I use these tools every day. I even used them to build this site. If we’re not growing, we’re dying, and I know this is where I need to grow. It’s just not where my heart is.&lt;/p&gt;
&lt;p&gt;So why am I writing these posts?&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;To flex the ol’ brain muscle and see if I can still string a coherent sentence together.&lt;/li&gt;
&lt;li&gt;Because a lot of engineers feel alone and out in the cold right now. You’re not alone! If even one person reads these and feels a little less isolated, it’s worth it.&lt;/li&gt;
&lt;li&gt;Because it’s fun to write something and throw it up on the public internet!&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;So follow along. I’ll keep posting here, along with whatever I’m building. You can grab the &lt;a href=&quot;https://jspratt.dev/writing/rss.xml&quot;&gt;RSS feed&lt;/a&gt; or just check back now and then.&lt;/p&gt;
</content:encoded></item></channel></rss>