That thing where you change your mind (and links)
For the past few weeks, I’ve been preparing to launch a new blog. But now I think I probably shouldn’t.
No, really.
I have this idea that there’s been a vibe change in web development in the past few months. There’s less money to waste. Many of the people laid off by big tech companies went to smaller companies which have much smaller teams and require very different approaches. Harsher regulations in terms of accessibility are coming.
It feels like we’re ready for new approaches. Or, more specifically, combining old approaches that have been unpopular, with new additions to the web platform.
- CSS Cascade Layers.
- Richer colour features in CSS.
- New features for offline storage.
- Deno’s maturity and faithfulness to web standards, which has prompted the node team to improve its compatibility.
- Firefox finally shipping module support in workers, which should make it simpler for us to use workers in JS.
- Transitions are finally on their way. They’ve already shipped in Chrome and other browser vendors sound positive.
- Navigation API looks very promising. Firefox is positive about it. And although the WebKit team is silent on it, I really hope that they end up supporting it as well.
These new features are an opportunity to reassess and rethink how we approach web dev. Moreover, I think they are an opportunity to find new ways to teach web dev. A clean slate. Help people get into a web dev practice that’s free of the cruft of the past and the complexities that have weighed web dev down in recent years.
I still think that’s a good idea.
So why have I decided against launching the blog? I even had a nice design and theme ready for it and everything, which I’ve put a lot of work into.
Because I already have a blog.
To be honest, this blog has never been that technical. My impression has always been that the audience for my blog and my newsletter has always been a bit mixed. Not just developers but also people who are interested in how technology and software affects us. This has often given me pause in the past and caused me to hesitate to broaden or shift the topic of the blog itself. But, you know, I think the benefit of focusing your efforts on a single platform outweights the audience I might lose by being a little big more eclectic and varied in my choice of topics.
I’ve also always been a little bit concerned about, well, my name. It’s a question that’s nagged me since my early days on the web. “Would my blog do better if the name sounded less foreign to Anglophones?”
I don’t know. But I do know that just because English is the de facto language of web development, that doesn’t mean it’s only ready by English-speakers, so the impact of my name probably isn’t as big as I’ve worried in the past.
Instead of launching a separate site and wasting time tweaking and updating a new theme, my plan is to start to post more on web dev on my blog and newsletter.
And maybe I’ll find some use for the almost-ready blog theme for a microsite or something later on.
At least, that’s what I think today. As always, I reserve the right to change my mind in the future.
Links #
AI #
- "The value-destroying potential of AI". “A lot of the people trying to deploy AI as a business solution are doing it because they don’t know how to measure what’s valuable about their business”
- "Visiting OpenAI – jill/txt". Seems like OpenAI has started a charm offensive. Wining and dining people who have the ears of policy makers.
- "Ugly Numbers from Microsoft and ChatGPT Reveal that AI Demand is Already Shrinking". “A huge amount has been invested into AI but consumers aren’t taking the bait. They’re treating it like those American Online startup disks in the mail.” My guess is that OpenAI’s biggest single user base are cheating students. Hence the summer lull.
- "The Rise and Fall of ChatGPT? - by Gary Marcus". Some of us have been saying from the start that generative models have a functionality problem.
- "Pluralistic: Supervised AI isn’t (23 August 2023) – Pluralistic: Daily links from Cory Doctorow". “The story of AI being managed by a “human in the loop” is a fantasy, because humans are neurologically incapable of maintaining vigilance in watching for rare occurrences.”
- "Reclaiming AI as a theoretical tool for cognitive science". In this paper, we have shown that (re)making human-like or human-level minds is computationally intractable (even under highly idealised conditions).
Dev #
- "iCloud sucks and it really shouldn’t". It really does and it really shouldn’t.
- "Pants on Fire – Pixel Envy". “But Evans does not give nearly enough weight to how often big industry players and their representatives simply lie.” I’m shocked, shocked etc. etc.
- "GitHub - orlp/polymur-hash: The PolymurHash universal hash function."
- "Use web components for what they’re good at". This is a solid and practical overview of what web components are good at.