Nolan Lawson

toot.cafe admin, dev at Socket. Creator of https://pinafore.social. Formerly Salesforce and Microsoft Edge. The web is for everyone.

Website: https://nolanlawson.com
Pronouns: he/him
Patreon: https://www.patreon.com/user?u=5535137
Header: https://flic.kr/p/owsn7U

Last Notes

2025-05-10 00:21:46 UTC

Every time I look at the new CSS APIs coming down the pike, I get excited about all the dumb code I can delete.
Today's unreasonable excitement comes from `field-sizing`, which aims to solve the problem of auto-expanding/auto-shrinking <textarea>s. Can't come soon enough! https://developer.mozilla.org/en-US/docs/Web/CSS/field-sizing

2025-03-14 14:44:45 UTC

This post from @npub1fdr…lvhs captures some of the dread dredged up by my recent thread on using LLMs for coding: https://pluralistic.net/2025/03/13/electronic-whipping/
> "Code reviewer" is a much less fulfilling job than "programmer." Code reviewers are also easier to replace than programmers. A code reviewer is a reverse-centaur, a servant to the machine. Every time you hear "AI-assisted programmer," you should substitute "programmer-assisted AI."

2025-03-09 14:16:32 UTC

- reply

These are good points, thank you. Maybe that's the heart of my inner conflict: these tools "work" (for some definition of "work") but they are inherently dehumanizing and insulting. They make a mockery of my ~20 years of hard-won expertise.
Ezra Klein in his podcast once said that the core "message" of GenAI (in the McLuhan sense) is: "You are derivative." I totally agree with that.

2025-03-08 23:29:14 UTC

- reply

I'm familiar with all the arguments against GenAI, and I'm a big fan of authors like @npub1hr7…jxyc whose excellent book, The Intelligence Illusion, I've read twice. It's highly recommended. https://illusion.baldurbjarnason.com/
I have a degree in computational linguistics, and that's partly why I was so skeptical of these tools for so long. I still am! But more and more I felt like the world was moving on without me. To this day I feel deeply conflicted.

2025-03-08 23:22:59 UTC

- reply

Of course, this is a choice: I *could* choose not to use LLMs. When I ride my bike, I don't bemoan the fact that a car could do it faster – the goal is exercise.
But nobody's paying me to ride my bike. If I were a delivery driver, it'd be pretty unprofessional to show up an hour late with a cold pizza just because I like biking.
With LLMs, it almost feels like malpractice *not* to use them at this point. I can't justify taking ~3x longer to ship a feature just because I don't enjoy using them.

2025-03-08 23:20:17 UTC

- reply

The other issue of course is that I'm not learning as much when I use LLMs. Even if it's an iterative, back-and-forth process, the tool is doing ~80% of the thinking for me.
It feels like I either need to spend more time learning outside of coding, or just accept at some level that I'm "cashing in my chips" and relying on ~20 years of actual coding experience.
Either way, I feel less excited by these tools than defeated. They're incredible magic wands, but I kind of liked doing my own sorcery?

2025-03-08 23:19:13 UTC

I held out for a long time, but I finally started using LLMs in my regular coding work. I couldn't deny the utility of tools like Claude any longer, when it can do in 2 minutes what would take me an hour.
There are problems: hallucinations, verbosity, bad/unperformant/inaccessible code. I have to correct it a lot. That's OK.
The main problem I find is that it's sucked a lot of the fun out of coding for me. It's like I've been pushed into a management role when I just wanted to stay a coder.

2025-02-03 20:35:14 UTC

"Offlinin' Aint Easy" by @npub1t0w…zgs2 https://blog.codepen.io/2025/02/03/chris-corner-offlinin-aint-easy/
Just to further illustrate how complicated Service Workers can be, the linked minimal examples are missing nav preloads, which means it can actually make your site _slower_ for non-cached requests: https://web.dev/blog/navigation-preload
Service Workers are great, but I think they're still too low-level a lot of the time. Would be great to have a higher-level, more ergonomic API.

2024-10-07 07:18:18 UTC

- reply

Honestly I don't hate this. I sympathize with JS engine authors trying to balance security/perf with ergonomics. Given the CVE count and dumb perf bugs like "var is faster than const/let", it seems like a losing battle.
JS is already quite good, and it would be sad IMO if it became like Perl where there are 20 ways to do everything. And of course if perf/security suffer. JS0/JSSugar seem totally reasonable to me.

2024-09-09 05:35:29 UTC

- reply

Totally randomly, I opened a PR on Tachometer to add support for cross-origin isolation… need to check on why it wasn't merged: https://github.com/google/tachometer/pull/234
For the Firefox error, I managed to fix it on Ubuntu by pointing to the snap binary: https://gist.github.com/nolanlawson/c9d42d0a49130ae6922718608c921efb

2024-09-01 16:35:34 UTC

The goal of Dependabot seems to be to remind me that my Node projects from ~2 years ago are utterly unmaintainable

2024-07-04 05:09:35 UTC

- reply

Great episode. Very even-handed, and a lot deeper than a lot of the web components discourse out there.
At Salesforce yes the whole UI is web components. Main benefit for us is that we build an evergreen platform – we change a component, we break a customer. So same constraints as the browser with <video controls> et al, so shadow DOM fits. It's not perfect, but working pretty well.

2024-06-04 14:31:00 UTC

Good news for #webcomponents: Chromium fixed a perf bug for repeated stylesheets in shadow DOM: https://issues.chromium.org/issues/341327461