This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
I'd give a different issue: regardless of how good or bad his ideas are, they're clearly unrelated to the actual goals he's claiming to champion. Twitter and YouTube and Discord and almost every company of relevance here are not market leaders due to the strength of their intellectual property; it's trivial to implement one-off examples of their functionality, and building a decent many-to-many implementation is a small business, not a large one. Their strengths come from their scaling capabilities and, to a far greater extent, the absolutely massive network advantages. The division from LibreOffice or GIMP to MSOffice and Photoshop isn't a massive, deep moat of algorithmic design or CPU optimizations, but a shallow one of user interface and user training. Individual people can build cell phones. It's just only a rounding error of people wants that done, to fund it, or to use it once manufactured.
It might be more relevant for specialized software (operating systems, CAD work, simulation software), but notably none of these spaces are things Doctorow focuses on. He talks about iOS in the sense of jailbreaking iPhones, a matter where legal constraints have never been the primary limit. He never mentions Linux, and only mentions Microsoft to say they "bricked" the International Criminal Court's outlook server due to sanctions (real world: cut access to Karim Khan's e-mail account). The ICC's moving to openDesk (also not mentioned, wouldn't have been my first choice)... and having it run by B1 Systems GmbH, a contractor in Germany. A quick google estimates <150 IT staff; having tried OpenDesk, I'd expect <20 full-time staff equivalent for the ICC, mostly tech support.
That is not a moonshot. It's definitely not the moonshot Doctorow's theory would need.
The only place they might be relevant is AI models (hmmm), and then only to the point where there are closed-source, high-capability models that could be cloned and run from EU services. That's not coherent to Doctorow's whole view - "Because even though the AI can't do the 's job, an AI salesman can convince the 's boss to fire them and replace them with an AI that can't do their job", that's the text - but he's not pretending to be coherent so much as tell his readers what he needs to get his goals, so whatever.
((Presumably they only ignore the copyright requests Doctorow dislikes, not artist and writer intellectual property, but to be fair, it's not like anyone without a hundred million dollar business can get an inter_state_ copyright lawsuit, nevermind an international one.))
How's that supposed to work? Okay, the model leaks, quickly. That I can buy, I've been a proponent of the theory that 'the leak always gets through' even if it hasn't always applied in practice. The EU companies are able to clone the graphics cards or ASICs, probably. Can they make them? The current best fab is 18nm, and while they're planning to build a 2nm-ish plant, the current timeline is 2030 and also kinda a joke. Okay, well, over long enough the hardware and training costs get amortized, it's the landscape and inference cost. Is EU power going to be cheap? Regulatory compliance? Legal overhead?
What's the business plan, here? Be annoying?
Mistral's been suffering for a while. It had some sizable influence in low-parameter models a year ago - and to an extent, still has: Cydonia is a Mistral-3.1-24B-derived model that's popular for roleplay, even if it introduces a lot of world consistency issues as context scale - but it's ranged from middling to actively bad since.
One complication here is that there are clear spaces that OpenAI and Anthropic are unlikely to want to explore, that would leave a niche for not-quite-frontier models that don't excel at things like coding but do focus well on other career spaces ... but that is likely to be more regulated in the EU, in ways that impact the ability of providers to provide decent models. And that's particularly overt for Mistral: one of the suspected causes for (some of the many) problems in Mistral 4 was the repeated 'safety' failures in Mistral3 variants. Ideally, they'd be able to avoid regulatory failures without harming core capabilities, but so far the degree models seem to suffer from overcorrection correlates pretty heavily with regulatory exposure.
(Caveat: they could have also just found some local minima. Things are moving so fast in these spaces that they could well turn around quick.)
More options
Context Copy link