4bpp

I assume your interlocutor was referring to how he dismissed those who complained about the move as "Russian trolls" more than the original delisting itself.

That's interesting, but these seem like they are just a slightly more structured form of the ways in which the CCP has been known to keep track of their own nationals abroad for many years. Not only is there no evidence or reason to expect that they would interact with people who are not PRC nationals, they presumably don't have guns and certainly have no actual policing powers either. If these "Chinese police stations" were to dispatch someone to my door, I could just call the actual police to have them removed. Meanwhile, I doubt I could call the "Chinese police stations" to protect me from the police of the country I live in, if they were to act upon a friendly request from the US like the Swedes and British did with Assange or the New Zealanders did with Kim Dotcom.

I doubt this particular move will complicate Russia's war efforts in any appreciable way. There is no non-fanciful way in which the named, well-established and closely watched kernel maintainers could have leveraged their status for it, and the tech scene has been one of the remaining bastions of pro-Western sentiment in Russian society. How many young programmers will be disenchanted with the West after hearing of this, and finally relent and answer their government's call to go write software for military drones or whatever?

Why should this be a problem for anyone outside of China? It's only when the same people can read your messages and send dudes with guns to your doorstep if they don't like what they see that things actually get dangerous.

"as we see in figure 4" is simply incorrect, though; expressions such as "Figure 4" or "3rd Street" are treated as proper names and must be capitalised, as would the non-nominal "third" in "Dritte Straße" if any German-speaking city were to have American-style numbered streets.

If I understand the timeline correctly here, it seems that gorhill overreacted, and I say that as someone who is usually harshly critical of everything Mozilla has done in the past 5+ years. It's hardly practical for Mozilla to manually review every add-on revision for safety in a timely manner, so they had the choice between automation and delays that would make add-on development a slog; automation though inevitably will cause false positives.

What's the alternative? No pre-release review at all? As a user I would hope that this will not be the case, especially now that we have confirmation that flashy supply chain attacks are being executed in the wild. In fact the review policy protects gorhill himself too, since it makes him a bit less attractive as a target for a rubberhose attack (no point in blackmailing him to put in spyware if the spyware would be caught before release).

In my experience the prototypical response is closer to an assertive "this was not made for you, you are part of a tiny minority of users who always have been catered to anyway" - if you were actually "handicapped", making the UI work for you would be a higher priority on political grounds.

In fact, it is sometimes entertaining to imagine that somewhere in the depths of the GNOME cult compound, they keep a single individual, who is legally blind, physically incapable of using a keyboard, simultaneously belongs to all demographic groups that are traditionally underrepresented among computer users, and has in fact never used a computer themselves but was shown a few things on an iPhone by their Gen Alpha great grandnephew. All these factors combine to turn this person (affectionately referred to as Mother Gnome by those in the know) into a utility monster of UI design, so that it becomes an absolute moral imperative to design things to appeal to them, at whatever expense it takes.

I'm sure that if Xwitter dispatched a "legal representative" to North Korea who has full moderation powers and will agree to be subjected to anything that a North Korean court orders, even an unblocking of it in NK could be negotiated. Does that mean that it is Musk's fault that his site remains blocked in that country, too? Conversely, is it actually normal for countries to personally punish legal representatives for actions that their employer performed, even if they did not have the power to prevent those actions?

Well, for what it's worth, your machine-translated sentence is basically incomprehensible in Japanese, let alone in a position to go viral (except in the capacity in which exquisite gibberish sometimes does). I could only infer what I think you intended to say ("This only works as long as translation is expensive"...?) by translating it back literally coupled with guesswork about what English text could have given rise to this "translation".

Not that your point isn't actually a good one - we do now often see the worst of US-made culture war toxoplasma (in all stages of its lifecycle) being imported wholesale into European cultures where it is barely relevant to local circumstances, which I would squarely blame on sufficient English skills and superior production value of anglophone politics.

Intriguing (and surprising to me that they offer E2EE at all), but there is seemingly no Linux build. I can't seem to find source code either (Telegram Desktop's is released under the GPL).

They used to have a page wittily named "feature matrix", which made it apparent that only Element was really kept up to date, with other clients missing features ranging from channel search to embedding images. I don't know if this situation has improved and whether the original page still exists somewhere.

EU countries prosecuted Assange, the Pirate Bay guys and now Durov. People in countries like Britain appear to frequently get persecuted for political posts. I'm sure that I've said things online that could get me in trouble now or would at some point in the future when the Overton window shifts in some other direction.

For one, it's the only major messenger that has an actually lightweight, well-written and full-featured desktop client rather than yet another boxed-up web browser. I might be more enthusiastic about using the alternatives if I could use the Telegram client.

In a way, Durov's arrest retroactively vindicates every EU citizen's decision to use Telegram (up until now), as it proves that they haven't been getting what they want from him. I am not nearly as concerned about Durov himself or the government of Dubai getting to read my messages as I am about the EU or one of its member states doing so, as there simply isn't much I can see the former doing with that data. The real danger only arises when the people who can read your messages and the people who can dispatch dudes with guns to your house are in cahoots. (For the same reason, I tend to roll my eyes at warnings about various forms of Chinese spyware.)

Well, except you were forced to use Nightly if you wanted to use any unapproved extensions (i.e. the vast majority of them).

I would fully expect no improvements before this lands in stable. To add insult to injury, they might even say that they trialled it in Nightly and found it to be a great success.

As yet another entry in the long cavalcade of inexplicable user-hostile changes, Firefox for Android has stopped displaying anything but the TLD at the top of the screen, unless you explicitly edit it. The bugtracker thread makes it clear that this is a deliberate change, though none of the deliberation seems to have been public. My best non-conspiracy guess would be that it is rationalised with an argument along the lines of hypothetical users being fooled by phishing websites along the lines of http://www.fake.com/1/2/3/http/www.real.com/, but surely this would have been prevented by the default left-justification and could have been reinforced with colour if they were so concerned.

Neither you nor the other first-order responses seem to acknowledge that the two actions are not exactly independent - it's not like the US government did a bad thing, and then Assange went and did a completely unrelated bad thing. The problem (as also hinted at in kevinventullo's response to a different subthread) is that the USG set up a situation in which anyone revealing their misdeeds would basically have to reveal the identities of the translators, or at least pay a great cost (in effort, credibility (as they no longer can release quite "original documents") and possibly liability (if the "editorialising" can be spun by lawyers somehow)) to not do so. The situation is thus a lot more similar to someone shooting the hostage (except of course in this case the hostage was not even actually shot!) along with the terrorist that took it and was threatening to blow up the building, then someone shooting a random person because they are a psychopath.

What part of the US government or its decision-makers spent 10 years stewing in prison for bribing Afghans to expose themselves to that same risk of death, or straight up killing many more Afghans for no other reason that they happened to be at the wrong place at the wrong time? Does the US government have some special natural right to toy with the lives of Afghans that Assange does not have?

Unless you believe it to be so, it seems quite strange to assign any significant share of the blame to Assange for any hypothetical deaths that may occur as a result of him taking actions to reduce the US government's ability to kill people abroad, akin to blaming police who stop a hostage-taker because this might have prompted the hostage-taker to kill the hostage but holding the hostage-taker himself blameless.

> But EU is in a different position, mostly oriented around trade, and notably lacking in security and military bodies. But the EU has absolutely stood up against US interests, especially their predatory corporations.

I think this understanding of the EU's behaviour may be insufficiently cynical. There's one pattern in politics that is very hard to not see everywhere once you have been primed to, which is "high + low against middle": the faction that is in power allies with one(s) that is so far away from power as to never become a credible threat to it, in order to put the squeeze on a third faction that is actually a serious contender for the position of the first.

A canonical example that's sufficiently historical that it hopefully won't be too incendiary was the practice of early communist states to elevate individuals of peasant/worker background into positions they were unqualified for, as in the famous case of Lysenko - here, high (party brass) supported low (peasants/workers) at the expense of the middle (bourgeois intellectuals, represented in that particular instance by academia, who could have been organised and experienced enough to orchestrate a palace coup).

Within the US, the federal government/military/foreign policy complex and tech-based New Money are widely recognised as two distinct power centres, with it at times being unclear if the former can actually fully dictate terms to the latter. Under normal circumstances one would expect the former to champion the interests of its industries on the international stage, and indeed the US is known to have very sharp elbows in this regard (from the famous oil wars in the Middle East via the slightly less famous fruit ones in Central America to the backdoor arm-twisting in copyright matters). The picture for the tech industry looks quite different - far from starting a war or even merely successfully lobbying the EU to drop its regulation, the USG is looking away and whistling. As it happens, out of the four industries mentioned (oil, fruit, media, tech), the tech industry happens to be the one that is by far the most autonomous and misaligned with federal government interests (Apple randomly grandstanding on privacy, everyone wanting to keep their Chinese supply contracts and market access, general abundance of politically engaged progressives and libertarians...). Wouldn't it make sense if what happened was that the USG (high) actually gave the EU (low) a tacit go-ahead for their anti-US-tech measures, and perhaps even indicated to everyone involved that they may let them crack down even harder if the tech industry (contender for high) keeps falling out of line?

The problem is that PHP still has no credible competition whatsoever in core use cases like "take an HTML document and replace static part with something fetched from a database", or "take multiple HTML documents and factor out a common structure that you might want to change everywhere at once later". Of course there exist solutions in "modern, well-designed languages" for these, but I will confidently take a challenge from any node.js, Elixir or Ruby developer on who can get it done faster.

Sophistry about how my solution won't scale to a million users per day or I should have used their framework from the start rather than having a bunch of HTML files won't matter to me nor the <10k people I expect on my website. In many cases, modern languages and frameworks drive adoption for themselves by tricking you into solving a problem that is not actually the problem you have, but they happen to be adapted to.

The funny thing is that while the "big move" population likes to say that they wanted to get away from The Algorithm, what actually convinced them to move was the threat that the Algorithm might change (due to Musk's takeover).

There is nothing particularly ironic about that - it's in the Chinese government's interest to control its own citizens and subvert its adversaries' control over theirs, and vice versa. See also Russia granting asylum to Snowden, and the myriads of Chinese, Russian, Middle Eastern etc. dissidents sheltered by the Western bloc. I imagine we would see more of the converse were it not for the language barrier, too.

If anything, we should be grateful for this situation; in a worse world all these governments would abide by gentlemen's agreements to not act against each other when it comes to controlling their subjects, notwithstanding their differences.

From what I can tell, this is essentially a complaint about a reviewer gently pushing back when Timnit Gebru (& less-famous coauthor) tried to instantiate Godwin's Law in an academic paper, padded to article length with applause lights for those who are concerned about AI, academia and/or capitalism and references to postmodernist literature.

To experience the actual ideologies of AI and consolidation of power, one may try to submit an AI paper that dismisses the Gebru and Torres conception of safety as a goal, or, if feeling especially cheeky, push back against Foucauldian discourse being appropriate for deciding questions of scientific ethics and purpose, for example in one of those "how can we improve review and grow our community" sessions every other CS conference seems to have these days. I doubt the pushback will be confined to a few side remarks from a single stray reviewer (no doubt since reprimanded and/or blacklisted). This camp has won comprehensively, and yet here they are talking up a tiny pocket of sheepish anonymous resistance as if all the powers that be were aligned against them.

I didn't have any direct interactions with JT then or now - while it was hard to ignore the discussion as an onlooker, it did not touch upon any parts of the code that I was involved with. This seems to be one of the topics where everyone who is even tangentially involved is under a default suspicion of being biased in one direction or another.

These all postdate the discussions that I linked (from March 31st). By April 1st JT themselves seems to have stopped making/boosting the claim about low memory usage.

The loading time improvements largely held up, and on the balance the mmap contribution was ultimately good (though the way it was implemented was really quite problematic, as a matter of process and communication). However, as I point out in https://news.ycombinator.com/item?id=39894542, JT quite unambiguously did try to cash in on the "low memory usage" claim - uncritically reprinting positive claims by others about your own work that otherwise would have been largely invisible should really not be treated differently as making those claims yourself.

I do think that there is a real risk that the numbers are wrong (not necessarily "fabricated", as this implies malfeasance, but possibly based on an erroneous measurement insufficiently questioned due to an excess of trust from themselves and others, as the mmap ones were). This is also in part based on the circumstance that at the time (of the mmap story, and myself being more involved in the project) I was actually involved in trying to optimise the SIMD linear algebra code, and unless llama.cpp has since switched to a significantly less performant implementation the proposition that so much more performance could be squeezed out strikes me as quite surprising. Here, your intuitions may say that Justine Tunney is just so brilliant that they make the seemingly impossible possible; but it was exactly this attitude that at the time made it so hard to evaluate the mmap memory usage claims rationally and turned the discussion around it much more dysfunctional than it had to be.

The very opening line says

> I'm glad you're happy with the fact that LLaMA 30B (a 20gb file) can be evaluated with only 4gb of memory usage!

The line you quoted occurs in a context where it is also implied that the low memory usage is a fact, and there might only be a bug insofar as that the model is being evaluated incorrectly. This is what is entailed by the assertion that it "is" sparse: that is, a big fraction of the parameters are not actually required to perform inference on the model.

My sense is that the part about credit/collaboration was actually somewhat overblown among the detractors. What roughly happened as far as I can remember is that JT and another person worked on mmap together with about equal contribution, though the other person might have been the one to have initiated the idea (and solicited help to push it through); then at some point JT decided to make a PR to the main repository in their own name, but crediting the other collaborator as a coauthor, which may or may not have been coordinated with the other person. After that, though, in a fairly characteristic fashion, JT started fielding adulatory questions from their fans (on Github, but also on HN, Twitter and possibly other media) about the change, and quickly switched to simply referring to it as their own, with no mention of the other contributor. The other contributor expressed some misgivings about having their contribution erased, which were picked up by a growing set of people who were generally resentful about JT's conduct in the project. As far as I can tell, when confronted about it, JT at no point explicitly denied what the other person did (and I think the commit logs should all still be there in the fork), but at some point the other person just decided to stop pushing the issue due to being uncomfortable with becoming a playing ball in the fandom war between JT fans and antis.

My personal main gripe with JT really was the tone they adopted in the Github discussions, and the effect of the large numbers of drive-by supporters, who were often far less restrained in both unfounded claims about Justine's accomplishments and attacks on any critics. (At this point I'd also like to note that I consider some sibling comments to be uncomfortably hostile in a personal way, like the "hit piece" one.) I think that as a public persona, especially one who actively pursues publicity, you have some responsibility to restrain your followers - Justine, I get the sense, instead uses them as deniable proxies, as also seen with the instances where instead of straight up putting their signature on the "RAM usage reduced to 6GB" claim they instead choose to post a collage of screenshots of supporters making it.

Plenty of claims about it, e.g. here as a "fact": https://github.com/ggerganov/llama.cpp/discussions/638#discu.... I don't think occasional expressions of lingering doubt (still couched among positive language like calling it a "miracle") can offset all the self-promotion that clearly seeks to maximise visibility of the implausible claim, even as it is attributed to others, as for example in https://twitter.com/JustineTunney/status/1641881145104297985... . A cereal manufacturer would probably be held responsible for package text like "Fruity Loops cured my cancer! - John, 52, Kalamazoo" too.

I don't know, my first (and main) impression of them was actually in the context of the llama.cpp mmap story, as I was somewhat involved in the project back then, and there I thought their impact on the project was predominantly negative. While they introduced a mildly beneficial change (mmap-based model loading), the way in which this was done was not healthy for the project - the changes were rammed through with little regard for concerns that existed at the time about backwards compatibility and edge cases that might be broken by the half-baked patch, Justine came across as self-aggrandizing (in the sense of "acting as if they ran the place", presenting their proposals as a plan that others must follow rather than suggestions) and overly eager to claim credit (epitomized by the injection of their own initials into the magic number file format identifier next to those of the project originator's, and the story of the hapless other author of the mmap changeset who was at first given a token acknowledgement but then quickly sidelined). Arguments for the inclusion of the patch seemed to be won by a combination of half- and untruths like those about memory savings and the sudden participation of a large number of previously uninvolved sycophants. It is fortunate that Georgi handled the fallout as well as he did, and that he in fact had amassed the social capital necessary to survive his heavy-handed solution (soft-banning both JT and their most prominent detractor). A less-successful project would probably have found itself captured or torn apart by the drama.

There is nothing wrong with holding people in esteem for their achievements, but in this case the degree of esteem really seems to be excessive. This is not a matter of simply being annoyed that people like "the wrong thing" - the mmap situation was significantly exacerbated by the presence of irrational/excessive supporters of Justine's as well as the irrational/excessive detractors that emerge wherever the former exist.