Notices by Jenniferplusplus (jenniferplusplus@hachyderm.io), page 6

Ok, let's do this. The Harris campaign should not be on mastodon. It would be bad for both the campaign, and mastodon.

🧵

#UsPol #FediMeta #MastodonForHarris, for muting purposes

@mattly Yeah, project with budgets should really be blazing trails for other projects to follow.

But then, for that matter, so should the FSF, the linux foundation, or the apache foundation

@mattly there used to be other realistic options, and I hate that they all just collapsed.

Http needs a status code for grudging acceptance. We have 202 Accepted. We need 222 Tolerated.

For when the client is doing it wrong, but you've decided to be the bigger person and deal with it anyway.

Actually, what I *really* want is a 3xx with a body.

322 Translated
Here's how you should have written that request, now go back to your seat and do it over

So, crowdstrike

The thing is, I still don't see the kind of attitude or behavior *from crowdstrike* that makes me think their culture or process made an outsized contribution to the weekend outage. Notably, they have not done the kind of labor discipline or stock manipulation that's been so common the last couple of years. And yeah, this has happened under the CEO before. But how many times did Bezos preside over disruptions that took out half the internet?

May I offer a different view?

People keep talking about what crowdstrike did or didn't do. What they should or shouldn't have done. Disappointingly, people who should know better. This event happened in a certain environment.

Public policy and capital malfeasance have created an extremely brittle and vulnerable computing environment (also in basically all other respects).

That's the thing we need to fix.

Other companies have been gutting their operations and security teams. I don't care what they say, the large majority of jobs they eliminated were actually necessary. They've been outsourcing to vendors to cover the gap.

Sometimes bad things happen. And knowing this, the entire field has been shredding their own adaptive capacity and crowding into a shrinking number of oversized blast zones.

Cory Doctorow has a piece about good fires vs wildfires. This is a wildfire.
https://doctorow.medium.com/let-the-platforms-burn-6fb3e6c0d980

@inthehands if we're just comparing package managers, then I guess gems and the bundler is closer to just working than pip is. But that ignores pythons venv solutions, and it also ignores rake which is like make but worse in every possible way, and also rails and the incredibly arcane bullshit that's involved with getting it to do anything at all. With python, once you've actually gotten your dependencies, it all just works.

@inthehands oh I have. That is awful. Easily the worst thing about python, and it's baffling to me that it's not considered a deal breaker for large projects.

This is worse

jfc ruby has the most god awful tool chain in the fucking world

@norootcause honestly, I'm really hoping we eventually learn what mitigations they put in place. It's all well and good for us to say you should have secondary access channels or automated rollbacks or whatever else. But that's a hard enough problem in userspace application code. How do you even do these things in the bootloader?

@J12t what would you hope to achieve using that proof?

@J12t Ok, then we'll go up a level and ask the question again. What claims have you made? Who did you make them to? What mission or criteria would they follow for funding decisions?

This question just isn't answerable until you define good. Good for whom? Good in what way?

@J12t
I'm obviously not in these conversations. I'm just going based on what you said, "They are asking, in so many words: you are saying all sorts of good things about the Fediverse, where is the evidence that what you say is true?"

Someone told someone else that the fediverse was good. Without knowing who any of these people are, or in what context this was said, the question is unanswerable.

@mattly no, I don't think that's a thing in mastodon

#HugOps for crowdstrike's on-call, and um, everyone I guess

Seriously, I want you all to stop snarking on crowdstrike . Accidents happen, and your smug armchair hindsight architect routine is not helping. If you can't muster up some sympathy for the teams scrambling to get their systems back online, then just shut up and let the adults work.

@mattly ah WTF, it used to be much longer.

that or ATT's dogshit cable modem is just dropping dns lookups to half the internet?

God, why the fuck do we let private companies own public infrastructure