The Changelog: Software Development, Open Source - Good ideas in computer science (News)

Episode Date: April 29, 2024

Daniel Hooper lists out all the good ideas in computer science, Jeff Geerling declares 2024 the year corporate open source dies, Jared Turner says all kinds of works-in-progress are waste, Daroc Alden covers the leadership crisis in the Nix community & John Hawthorn explains why Ruby may be faster than you think.

What up, nerds? I'm Jared, and this is Changelog News for the week of Monday, April 29th, 2024. After a Friday afternoon tornado knocked out the power to my home office for over 50 hours, I planned to record this episode from my parents' basement, like the nerd that I am. But, thankfully, the lights flipped back on last night around 8pm and the series of tubes pushed that sweet, sweet internet back into our lives. Okay, let's get into the news. Good ideas in computer science. Daniel Hooper writes,

Programmers love arguing for their favorite technologies. C++ vs. Rust, Mac vs. PC. These arguments overshadow the victories of computer science, the ideas that we all agree on. To unearth these ideas, I recently asked a simple question on Twitter slash X. What ideas in computer science are universally considered good? End quote. The resulting list includes 17 ideas, not implementations,

just ideas that are universally considered good, which means not really debated. What's interesting, in addition to the ideas themselves, is that they were all thought of or discovered prior to 1974. Where are all the new good ideas in computer science? Perhaps we are still debating them. Daniel concludes with, by 1974, 50 years ago, we had most of what we call modern computing. Today's fundamentals are the same. A C programmer from 1974 would feel at home on a modern computer except for the alien-like speed. I hope we have new ideas that in 50 years will be universally considered good.

Corporate open source is dead. Instead of linking to an IBM press release announcing their purchase of HashiCorp for $6.4 billion, let's enjoy our friend Jeff Geerling's analysis. Jeff says, quote, as free money dries up and profits slow, companies slash headcount almost as fast as community trust, end quote. Can you smell what Jeff is cooking? Quote, open source culture relies on trust, trust that companies you and I help build, even without being on the payroll, wouldn't rug pull. But time and time again, that trust is shattered, end quote. He goes on to declare that 2024 is the year that corporate open source dies.

But it ain't all bad, not in Jeff's eyes. Quote, in fact, this could be a huge opportunity. What happened to the spunky startups like Ansible, HashiCorp, Elasticsearch, or Redis? They were lighting their industries on fire with great new software. What happened to building up communities for developers, crossing cultural and economic barriers to make software that changed the world? There are still projects doing that, but so many succumb to enterprise money, where eye-watering amounts of revenue puts profit over philosophy. Maybe it's time for a new open source rebellion. Maybe this time, money won't change company culture as new projects arise from the ash heap.

Maybe not, but at least we can try. Whip is waste. Jared Turner writes on ThoughtBot's blog, a blog which I've linked to seldomly in recent years, but I'm tickled at the opportunity. ThoughtBot did a lot of good for me in my early consulting career, but I digress. Jared says, quote, work in progress has zero value. Ship. Before a task is shipped, it provides zero value. Any work in progress is pure cost. Two tasks in progress adds cost for no value. Only after shipping do you create value. Always ship. One task shipped is infinitely better than four tasks. Almost done. Ship something of value first, then begin something new. You get the drift. This is a good reminder for me that

our super cool custom feeds feature that currently only I'm using provides zero value to the people I'm building it for until I actually ship the sucker. Always be shipping. It's now time for sponsored news. Join Tauri's DevTools Premium waitlist. If you're building with Tauri, this might be the best news you hear all week. Their DevTools Premium offering is right around the corner. Here's what you can expect. One, a native desktop app built with Towery. Two, first class support for Towery 2.0. Three, powerful collaboration features that allows teams to inspect and debug their applications together. Four, it can be used

as an embedded drawer in a towery application or popped out for a bigger view, just like in Chrome DevTools. The Crab Nebula team is even experimenting with more advanced features like record slash replay of debugging sessions. This isn't just about finding and fixing issues. It's about understanding, optimizing, and perfecting the application development process. So join the waitlist today by following the link in the newsletter and chapter data. And thank you to Daniel and our friends at Crab Nebula for sponsoring this week's ChangeLog News. A leadership crisis in the Nix community. Derek Alden writing for LWN.

Quote, On April 21st, a group of anonymous authors and non-anonymous signatories published a lengthy open letter to the NYX community and NYX founder Ilko Dostra calling for his resignation from the project. They claimed ongoing problems with the project's leadership, primarily focusing on the way his actions have allegedly undermined people nominally empowered to perform various moderation and governance tasks. Since its release, the letter has gained more than 100 signatures. End quote. That open letter isn't merely lengthy, I'd call it meandering and difficult to follow. Thankfully, the good communicators at LWN grokked it so we don't have to. Quote, The open letter has several related complaints, but the most central one is that they allege DOLSTRA has repeatedly strong-armed the board

and members of the other community teams to overrule their decisions. End quote. The crux of the matter seems to be that Nix is governed by the NixOS Foundation, but DOLSTRA is acting like a BDFL. The authors and signatories of the letter who are calling for DOL Dolstra's resignation from the board

say if he doesn't, a fork is planned. Ruby might be faster than you think. After running a crystallized version of Fibonacci, which embeds crystal code directly in Ruby, in order to confirm that it executed four times faster than a pure Ruby version, John Hawthorne noticed something. Quote,

Something is a bit off here. The Ruby implementation has a subtle mistake which causes significantly more work than it needs to. End quote. Turns out by tweaking the Ruby implementation and enabling the built-in JIT compiler, he flipped the benchmark on its head. Quote,

Now it's Ruby that's five times faster than Crystal, and 20 times faster than our original version. I thought it was notable that by making some minor tweaks to Ruby code, it can now outperform a pre-compiled statically typed language in a purpose-built example of when it is slow. I'm hopeful that someday, with future advancements in the Ruby JIT, even the small tweaks might not be necessary. End quote. My takeaway from this post isn't that Ruby is finally web scale now.

It's that we do well to know our tools deeply. Whichever tools we decide to use. That's the news for now. But be sure to scan the companion changelog newsletter for more stories, including MS-DOS 4.0 going open source, the end of the free tier, Google laying off its Python team, and a sweet list of new tools for the old toolbox. If you are audio only, pop your email address in at changelog.com slash news, or use a service like Feedbin or Kill the Newsletter to pipe it into your RSS reader. We have some great episodes coming up. Dustin Block, the new owner of Castro, on Wednesday, and Ron Evans from TinyGo on Friday.

Have a great week. Email me your five-star review if you want some free stickers. And I'll talk to you again real soon.