Notices by ben (benjamineskola@hachyderm.io)

@inthehands I might just add (in agreement) that just because someone suggested it before and didn’t win the election doesn’t mean it was wrong then, and especially doesn’t mean it’s wrong now.

You can’t use popularity as a proxy measure for right and wrong.

@dalias that… doesn’t follow at all.

What about people who thought about what it said, agreed, and voted yes?

@JdeBP @soller @neal @feonixrift @wwahammy @artemis

@thomasfuchs Yeah: I haven't been a macos user for as long as some (only since 2011–12) but all of the original appeal (it's unix, but with a polished and reliable UI) has gone.

@inthehands for a while I went about trying to do things the docker way and used one of the docker-oriented hosting services and eventually decided it just doesn’t make sense. It feels like it’s trying to solve problems most people don’t have. At least at the personal-project level; but I also suspect for professional use too.

It’s like microservices or something: not necessarily *never* useful, but people convince themselves it’s useful when it’s not, just because big tech likes it; and/or if it is useful in a particular case it may just be a sign of a deeper problem.

@inthehands no, that’s pretty much it.

IIRC compose was a separately-built tool that got bought by Docker and though the command is now integrated the configuration remained completely separate. So e.g., there’s a ‘port’ option in the Dockerfile that does nothing, it’s just indicative or something? You have to do it in compose, and there’s no requirement for the Dockerfile to have been set up to expect it even.

@thomasfuchs I have a coworker who seemingly can’t take no for an answer on this. As if it’s not bad enough to have to deal with his use of it.

@thomasfuchs @Floppy What people should have and what they do have don’t always match: I’ve heard recently of corporate IT departments banning password managers, for example.

(I assume the same departments also require password changes every three months, etc.)

@inthehands I could get behind a policy by the Nobel committee of just giving it everywhere to the individual or group that would most annoy Donald Trump, though.

@inthehands @julieofthespirits Yes, and I think it makes it difficult to talk about the history (and continuing existence of) concentration camps — because everyone thinks of ‘death camps’. So if you point out that concentration camps were invented by Britain (in the Boer war) it sounds hyperbolic — because we weren’t rounding people up to murder them.

(Or, as pointed out downthread, the US WW2 internment camps.)

But even without being a death camp it’s deeply unjust (and probably also does lead to deaths that otherwise wouldn’t have happened, anyway).

@inthehands my naive interpretation would be that it’s impersonation of a law enforcement officer, or some similarly-termed crime.

But of course it would still depend on other law enforcement being willing/able to enforce it — presumably they’d still have the ‘obstructing a federal officer’ issue?

@thomasfuchs agreed — there have been libraries I’ve stopped using because they were releasing multiple new versions a week, and backwards-incompatible major versions several times a year. Life’s too short to spend that much time thinking about pagination.

It’s not even that the updates were mandatory, but it just gives a strong sense that the developer isn’t stopping to think about what they’re doing.

@inthehands @Bfordham yeah this is what came to my mind too. I don’t think it’s meaningful to think of an LLM as being slave-like at all.

But listing it as a _positive_ about LLMs is deeply, deeply suspect. It’s not something one should want to be true.

@inthehands is an aspect of this also when correctness is hard to verify?

I can think of examples where the verification process would basically involve doing most of the same work over again. (e.g., summarisation, where if the summary needs to be correct, then a human has to read over all the same inputs and ensure that nothing has been missed or invented.)

@thomasfuchs I’ve wondered recently why people claim that sarcasm/irony is hard to detect in writing (in internet communication). Both have been used in literature for centuries.

@thomasfuchs There’s a suggestion that he’s plagiarising (or effectively plagiarising), which might be true — it feels like a risk that might arise from the lack of rigour? But I think that’s a different sort of criticism than “grifter” or whatever.

@thomasfuchs I’m a little puzzled by all the people acting surprised that he works in PR — I don’t read his articles regularly but when I read some in the past he seemed to be pretty transparent about that.

It’s not necessarily a good thing but I don’t see how there’s been any dishonesty about it.

@thomasfuchs I thought this was an SGI workstation at first glance.

@thomasfuchs I switched away from one particular gem because of precisely this sort of update cycle. Multiple releases per week; something like sixty in a single year, including three backwards-incompatible major versions. For a pagination library.

But also I think tools like Dependabot are part of the problem; it’s promoting the idea that installing updates ASAP is the correct and normal approach.

@thomasfuchs I also associate it with startup CEO types on LinkedIn (and yes, I do recognise that there’s a significant overlap).

When it’s not specifically from a programmer I tended to interpret it as someone who wasn’t actually able to explain what they contributed (beyond LinkedIn posts), and so had to adopt a vague waffly title for themselves instead.

@badnetmask yes agreed. Why would “this guy made it” be interesting enough to be a selling point?

(I feel like maybe the very fact that he’s built his own distro suggests his opinions on distros can be dismissed, if nothing else. Feels ego-driven.)