zoul

But it’s true in the general idea of Bluesky being highly centralized and dependent on a single for-profit entity.

Catching bugs in CI is orders of magnitude slower and more expensive than catching them as type errors while writing the code.

Static type checking greatly affects code design. That’s why converting from JS to TS doesn’t give you full benefits of the type system – the code doesn’t lean into types to model invariants.

For me the problem wasn’t the cultural change, it was depending on language whose leadership doesn’t seem to give a fuck about real-world problems the language users face in production. And then being constantly gaslighted about it on the official channels like the language forum.

They did the best for society, not for their child. Yes, vaccines are the best choice, there is no doubt about that. But the society in question must take much better care of those who sacrifice so much for the whole.

As a programmer, I don’t feel threatened by the technology itself, but I do feel threatened by the second-degree effects such as what the technology does to our field, especially in the wrong hands.

This! There is a “Bluesky outage”. There is no “Mastodon outage”.

I think we are all saying that Mastodon is meaningfully decentralized and there is no single party that could pull the rug from beneath the whole network. This is what instances bring to the table. Bluesky may be architecturally beautiful, but it’s not decentralized in practice. (See https://arewedecentralizedyet.online for example.) And we know what tends to happen to important centralized services.

New services don’t start with the same userbase because users don’t use protocols, they use apps. And for Bluesky that’s predominantly a single centralized app. If that demotes content from the “competitors” or even switches the protocol completely, you are stuffed. The apps are very much not equal.

On the other hand, the currently overwhelmingly dominant Bluesky app and infrastructure can control what users see, presenting a centralized chokepoint. It’s nice to own your data but if your social circle uses a single centralized app to view it, shrug.

Any instance will work I think, just start following someone like https://eupolicy.social/@bert_hubert and you should discover relevant people and hashtags soon enough.

Mastodon works fairly well for that I think.

Very nice developer experience. A lot of batteries included, like CDN, incremental page regeneration, image pipeline or observability. Not having to maintain a server.

I’m still planning to move elsewhere though, the vendor lock-in is not worth it and I’d like to keep our infra in the EU.

“The vision was always compelling. A social web that no single company controls, where users own their identity and their relationships,” says the new CEO of the single company that can control what millions of Bluesky users see in their apps.

Well, right now a foreign government can do that to us, so shifting the risk to our own politicians would be an improvement.

Is there or do you plan to add some kind of webhook to notify interested parties that a vault has changed? I think it would go together nicely with the headless sync.

Yes, you can add an enormous amount of complexity to your app in 3 clicks. Should you? The proverbial night crashes are, in my experience, often much better problem to have than the complexity and cost of AWS.

My argument works just as well if you replace “perfecting” with “improving”.

Would you keep fixing the underlined spelling mistakes on your “watch out for holes in the pavement” sign while people are already walking there?

I would even say that from a project management perspective, zero technical debt is undesirable. It means you have invested resources into perfecting something that, almost by definition, could have waited a while, instead of improving some more important metric such as user experience. (I do understand tech debt makes it harder to work with the codebase, impacting all metrics, I just don’t think zero tech debt is a good target.)

Solar power is very cheap and still getting cheaper:

https://www.statista.com/chart/35117/levelized-cost-of-energ...

In what way is Mastodon a failed experiment? Low number of active users or something else?

Believe it or not, many people just don’t care about the stock market. But they may still care about the economy that could crash badly if the AI bubble gets too big before it pops.

Oh yes. As opposed to the “kiss the ring or else” market freedoms of the US.

I hope it’s not that black-and-white, that it’s possible to have a sane social network with algorithmic feed, only we need to design the algorithms around users’ needs first.

It’s arguably much easier to solve the UX issues when you are designing a centralized service, which is what Bluesky is.

A safe rollback on a Friday afternoon is a nice thing for sure :)

That would get us back to the “bigger is better” hole though.

This is not about finding the most effective solution, it’s about showing that they “understand” the problem. Could they write the algorithm if it were not in their training set?

I love it. It feels great to have a search engine built for users, not advertisers or investors. Trivial stuff like being able to remove some domains from search results. Or not having to wade through AI bullshit to get to the results. Or, you know, not having your search queries sold to whatever advertising partners that care about my privacy.