Need to let loose a primal scream without collecting footnotes first? Have a sneer percolating in your system but not enough time/energy to make a whole post about it? Go forth and be mid: Welcome to the Stubsack, your first port of call for learning fresh Awful youāll near-instantly regret.
Any awful.systems sub may be subsneered in this subthread, techtakes or no.
If your sneer seems higher quality than you thought, feel free to cutānāpaste it into its own post ā thereās no quota for posting and the bar really isnāt that high.
The post Xitter web has spawned soo many āesotericā right wing freaks, but thereās no appropriate sneer-space for them. Iām talking redscare-ish, reality challenged āculture criticsā who write about everything but understand nothing. Iām talking about reply-guys who make the same 6 tweets about the same 3 subjects. Theyāre inescapable at this point, yet I donāt see them mocked (as much as they should be)
Like, there was one dude a while back who insisted that women couldnāt be surgeons because they didnāt believe in the moon or in stars? I think each and every one of these guys is uniquely fucked up and if I canāt escape them, I would love to sneer at them.
(Semi-obligatory thanks to @dgerard for starting this)
Interesting post and corresponding mastodon thread on the non-decentralised-ness of bluesky by cwebber.
https://dustycloud.org/blog/how-decentralized-is-bluesky/
https://social.coop/@cwebber/113527462572885698
The author is keen about this particular āvision statementā:
Preparing for the organization as a future adversary.
The assumption being, stuff gets enshittified and how might you guard your product against the future stupid and awful whims of management and investors?
Of course, they donāt consider that it cuts both ways, and Jack Dorseyās personal grumbles about Twitter. The risk from his point of view was the company he founded doing evil unthinkable things like, uh, banning nazis. Heās keen for that sort of thing to never happen again on his platforms.
note that cwebber wrote the ActivityPub spec, btw
how come every academic I have worked with has given me some variation of
they already have all of my data, I donāt really care about my privacy
iām in computer science š
When people start going on about having nothing to hide usually it helps to point out how thereās currently no legal way to have a movie or a series episode saved to your hard drive.
I suspect great overlap between the nothing-to-hide people and the people who watch the worst porn imaginable but think incognito mode is magic.
whatās wild is in the ideal case, a person who really doesnāt have anything to hide is both unimaginably dull and has effectively just confessed that they would sell you out to the authorities for any or no reason at all
people with nothing to hide are the worst people
the marketing fucks and executive ghouls who came up with this meme (that used to surface every time I talked about wanting to de-Google) are also the ones who make a fuckton of money off of having a real-time firehose of personal data straight from the source, cause thatās by far whatās most valuable to advertisers and surveillance firms (but I repeat myself)
Never thought Iād die fighting alongside a League of Legends fan.
Aye. That I could do.
You just know Netflixās inbox is getting flooded with the absolute worst shit League of Legends players can come up with right now
And having played more LoL than I care to admit in high school, thatās some truly vile shit. If only it actually made it through the filters to whoever actually made the relevant choices.
Dude discovers that one LLM model is not entirely shit at chess, spends time and tokens proving that other models are actually also not shit at chess.
The irony? Heās comparing it against Stockfish, a computer chess engine. Computers playing chess at a superhuman level is a solved problem. LLMs have now slightly approached that level.
For one, gpt-3.5-turbo-instruct rarely suggests illegal moves,
Writeup https://dynomight.net/more-chess/
HN discussion https://news.ycombinator.com/item?id=42206817
I remember when several months (a year ago?) when the news got out that gpt-3.5-turbo-papillion-grumpalumpgus could play chess around ~1600 elo. I was skeptical the apparent skill wasnāt just a hacked-on patch to stop folks from clowning on their models on xitter. Like if an LLM had just read the instructions of chess and started playing like a competent player, that would be genuinely impressive. But if what happened is they generated 10^12 synthetic games of chess played by stonk fish and used that to train the model- that aināt an emergent ability, thatās just brute forcing chess. The fact that larger, open-source models that perform better on other benchmarks, still flail at chess is just a glaring red flag that something funky was going on w/ gpt-3.5-turbo-instruct to drive home the āeMeRgEnCeā narrative. Iād bet decent odds if you played with modified rules, (knights move a one space longer L shape, you cannot move a pawn 2 moves after it last moved, etc), gpt-3.5 would fuckin suck.
Edit: the author asks āwhy skill go down thoā on later models. Like isnāt it obvious? At that moment of time, chess skills werenāt a priority so the trillions of synthetic games werenāt included in the training? Like this isnāt that big of a mysteryā¦? Itās not like other NN havenāt been trained to play chessā¦
Particularly hilarious at how thoroughly theyāre missing the point. The fact that it suggests illegal moves at all means that no matter how good itās openings are the scaling laws and emergent behaviors havenāt magicked up an internal model of the game of Chess or even the state of the chess board itās working with. I feel like playing games is a particularly powerful example of this because the game rules provide a very clear structure to model and itās very obvious when that model doesnāt exist.
LLMs sometimes struggle to give legal moves. In these experiments, I try 10 times and if thereās still no legal move, I just pick one at random.
uhh
Battlechess both could choose legal moves and also had cool animations. Battlechess wins again!
Here are the results of these three models against Stockfishāa standard chess AIāon level 1, with a maximum of 0.01 seconds to make each move
Iām not a Chess person or familiar with Stockfish so take this with a grain of salt, but I found a few interesting things perusing the code / docs which I think makes useful context.
Skill Level
I assume ālevelā refers to Stockfishās Skill Level option.
If I mathed right, Stockfish roughly estimates Skill Level 1 to be around 1445 ELO (source). However it says āThis Elo rating has been calibrated at a time control of 60s+0.6sā so it may be significantly lower here.
Skill Level affects the search depth (appears to use depth of 1 at Skill Level 1). It also enables MultiPV 4 to compute the four best principle variations and randomly pick from them (more randomly at lower skill levels).
Move Time & Hardware
This is all independent of move time. This author used a move time of 10 milliseconds (for stockfish, no mention on how much time the LLMs got). ā¦ or at least they did if they accounted for the āMove Overheadā option defaulting to 10 milliseconds. If they left that at itās default then 10ms - 10ms = 0ms so š¤·āāļø.
There is also no information about the hardware or number of threads they ran this one, which I feel is important information.
Evaluation Function
After the game was over, I calculated the score after each turn in ācentipawnsā where a pawn is worth 100 points, and Ā±1500 indicates a win or loss.
Stockfishās FAQ mentions that they have gone beyond centipawns for evaluating positions, because itās strong enough that material advantage is much less relevant than it used to be. I assume it doesnāt really matter at level 1 with ~0 seconds to produce moves though.
Still since the author has Stockfish handy anyway, itād be interesting to use it in itās not handicapped form to evaluate who won.
@gerikson @BlueMonday1984 the only analysis of computer chess anybody needs https://youtu.be/DpXy041BIlA
Stack overflow now with the sponsored crypto blogspam Joining forces: How Web2 and Web3 developers can build together
I really love the byline here. āKindest view of one anotherā. Seething rage at the bullshittery these āweb3ā fuckheads keep producing certainly isnāt kind for sure.
Strap in and start blasting the Depeche Mode.
When the reporter entered the confessional, AI Jesus warned, āDo not disclose personal information under any circumstances. Use this service at your own risk.
Do not worry my child, for everything you say in this hallowed chamber is between you, AI Jesus, and the army of contractors OpenAI hires to evaluate the quality of their LLM output.
a better-thought-out announcement is coming later today, but our WriteFreely instance at gibberish.awful.systems has reached a roughly production-ready state (and you can hack on its frontend by modifying the
templates,pages,static, andlessdirectories in this repo and opening a PR)! awful.systems regulars can ask for an account and Iāll DM an invite link!The mask comes off at LWN, as two editors (jake and corbet) dive in to frantically defend the honour of Justine fucking Tunney against multiple people pointing out sheās a Nazi who fills her projects with racist dogwhistles
Not the only trans NRXer to pull this Iām afraid. I could say things but I really canāt I think.
Is Google lacing their free coffee??? How could a woman with at least one college degree believe that the government is even mechanically capable of dissolving into a throne for Eric Schmidt.
fuck me that is some awful fucking moderation. I canāt imagine being so fucking bad at this that I:
- dole out a ban for being rude to a fascist
- dole out a second ban because somebody in the community did some basic fucking due diligence and found out one of the accounts defending the above fascist has been just a gigantic racist piece of shit elsewhere, surprise
- in the process of the above, I create a safe space for a fascist and her friends
but for so many of these people, somehow thatās what moderation is? fucking wild, how the fuck did we get here
See, youāre assuming the goal of moderation is to maintain a healthy social space online. By definition this excludes fascists. Itās that old story about how to make sure your punk bar doesnāt turn into a nazi punk bar. But what if instead my goal is to keep the peace in my nazi punk bar so that the normies and casuals keep filtering in and out and making me enough money that I can stay in business? Then this strategy makes more sense.
Centrists Donāt Fucking Be Like This challenge not achieved yet again
fwiw this link didnāt jump me to a specific reply (if you meant to highlight a particular one)
It didnāt scroll for me either but thereās a reply by this corbet person with a highlighted background which I assume is the one intended to be linked to
Post by Corbet the editor. āWe get it: people wish that we had not highlighted work by this particular author. Had we known more about the person in question, we might have shied away from the topic. But the article is out now, it describes a bit of interesting technology, people have had their say, please letās leave it at that.ā
So you updated the article to reflect this right? padme.jpg
Seems like theyāve actually done this now. Thereās a preface note now.
This topic was chosen based on the technical merit of the project before we were aware of its authorās political views and controversies. Our coverage of technical projects is never an endorsement of the developersā political views. The moderation of comments here is not meant to defend, or defame, anybody, but is in keeping with our longstanding policy against personal attacks. We could certainly have handled both topic selection and moderation better, and will endeavor to do so going forward.
Which is better than nothing, I guess, but still feels like a cheap cop-out.
Side-note: I can actually believe that they didnāt know about Justine being a fucking nazi when publishing this, because I remember stumbling across some of her projects and actually being impressed by it, and then I found out what an absolute rabbit hole of weird shit this person is. So I kinda get seeing the portable executables project, thinking, wow, this is actually neat, and running with it.
Not that this is an excuse, because when you write articles for a website that should come with a bit of research about the people and topic you choose to cover and you have a bit more responsibility than someone whoās just browsing around, but what do I know.
Well, at least they put down something. More than I expected.
And doing research on people? In this economy?
so is corbet the same kind of fucker thatāll complain āeverything is so political nowadaysā? it seems like they are
@dgerard @BlueMonday1984 also, and I know this is way beside the point, update the design of your website, motherfuckers
I donāt run any websites, what are you coming at me for
most of the dedicated Niantic (Pokemon Go, Ingress) game players I know figured the company was using their positioning data and phone sensors to help make better navigational algorithms. well surprise, itās worse than that: theyāre doing a generative AI model that looks to me like itās tuned specifically for surveillance and warfare (though Niantic is of course just saying this kind of model can be used for robotsā¦ seagull meme, āwhat are the robots for, fucker? why are you being so vague about whoās asking for this type of model?ā)
Quick, find the guys who were taping their phones to a ceiling fan and have them get to it!
Jokes aside Iām actually curious to see what happens when this one screws up. My money is on one of the Boston Dynamics dogs running in circles about 30 feet from the intended target without even establishing line of sight. Theyāll certainly have to test it somehow before it starts autonomously ordering drone strikes on innocent peopleās homes, right? Right?
Pokemon Go To The War Crimes
Pokemon Go To The Hague
Peter Wattsās Blindsight is a potent vector for brain worms.
Watts has always been a bit of a weird vector. While he doesnāt seem a far righter himself, he accidentally uses a lot of weird far right dogwhistles. (prob some cross contamination as some of these things are just scientific concepts (esp the r/K selection thing stood out very much to me in the rifters series, of course he has a phd in zoology, and the books predate the online hardcore racists discovering the idea by more than a decade, but still odd to me)).
To be very clear, I donāt blame Watts for this, he is just a science fiction writer, a particularly gloomy one. The guy himself seems to be pretty ok (not a fan of trump for example).
Thatās a good way to put it. Another thing that was really en vogue at one point and might have been considered hard-ish scifi when it made it into Rifters was all the deep water telepathy via quantum brain tubules stuff, which now would only be taken seriously by wellness influencers.
not a fan of trump for example
In one the Eriophora stories (I think itās officially the sunflower circle) I think thereās a throwaway mention about the Kochs having been lynched along with other billionaires on the early days of a mass mobilization to save whatās savable in the face of environmental disaster (and also rapidly push to the stars because a Kardashev-2 civilization may have emerged in the vicinity so an escape route could become necessary in the next few millenia and this scifi story needs a premise).
Huh. Say more?
Oh man where to begin. For starters:
- Sentience is overrated
- All communication is manipulative
- Assumes intelligence has a āvalueā and that it stacks like a Borderlands damage buff
- Superintelligence operates in the world like the chaos god Tzeench from WH40K. Humans canāt win, because all events are ājust as plannedā
- Humanity is therefore gormless and helpless in the face of superintelligence
It just feeds right into all of the TESCREAL nonsense, particularly those parts that devalue the human part of humanity.
Sentience is overrated
Not sentience, self awareness, and not in a parĻicularly prescriptive way.
Blindsight is pretty rough and probably Wattās worst book that Iāve read but itās original, ambitious and mostly worth it as an introduction to thinking about selfhood in a certain way, even if this type of scifi isnāt oneās cup of tea.
Itās a book that makes more sense after the fact, i.e. after reading the appendix on phenomenal self-model hypothesis. Which is no excuse ā cardboard characters that are that way because the author is struggling to make a point about how intelligence being at odds with self awareness would lead to individuals with nonexistent self-reflection that more or less coast as an extension of their (ultrafuturistic) functionality, are still cardboard characters that you have to spend a whole book with.
I remember he handwaves a lot of stuff regarding intelligence, like at some point straight up writing that what you are reading isnāt really whatās being said, itās just the jargonaut pov character dumbing it way down for you, which is to say he doesnāt try that hard for hyperintelligence show-donāt-tell. Echopraxia is better in that regard.
It just feeds right into all of the TESCREAL nonsense, particularly those parts that devalue the human part of humanity.
Not really, there are some common ideas mostly because tesrealism already is scifi tropes awkwardly cobbled together, but usually what tescreals think is awesome is presented in a cautionary light or as straight up dystopian.
Like, thereās some really bleak transhumanism in this book, and the view that human cognition is already starting to become alien in the one hour into the future setting is kind of anti-longtermist, at least in the sense that the utilitarian calculus turns way messed up.
And also I bet thereās nothing in The Sequences about Captain Space Dracula.
I got a really nice omnibus edition of Blindsight/Echopraxia that was printed in the UK, but ultimately, the necessarily(?) cardboard nature of the vampire character in Echopraxia was what left me cold. The first chapter or two are some of the most densely-packed creative sci-fi ideas Iāve ever read, but I came to the book looking for more elaboration on the vampires, and didnāt really get that. Valerie remains an inscrutable other. The most memorable interaction she has is when sheās breaking her arm and making the POV character guy reset it, seemed like she was hitting on him?
I hear you. I should clarify, because I didnāt do a good job of saying why those things bothered me and nerd-vented instead. I understand that an author doesnāt necessarily believe the things used as plot devices in their books. Blindsight a horror/speculative fiction book that asks āwhat if these horrible things were trueā and works out the consequences in an entertaining way. And, no doubt thereās absolutely a place for horror in spec fic, but Blindsight just feels off. I think @Soyweiser explained the vibes better than I did. Watts isnāt a bad guy. Maybe itās just me. To me, it feels less Hellraiser and more Human Centipede i.e. hereās a lurid idea that would be tremendously awful in reality, now buckle up and letās see how it goes to an uncomfortable extent. Thatās probably just a matter of taste, though.
Unfortunately, the kind of people who read these books donāt get that, because media literacy is dead. Everyone Iāve heard from (online) seems to think that it is saying big deep things that should be taken seriously. It surfaces in discussions about whether or not ChatGPT is āaliveā and how it might be alive in a way different from us. Eric Schmidtās recent insane ramblings about LLMs being an āalien intelligence,ā which donāt call Blindsight out directly, certainly resonate the same way.
Maybe Iām being unfair, but it all just goes right up my back.
It might be just the all but placeholder characters that give it a b-movie vibe. Iād say itās a book thatās both dumber and smarter that people give it credit for, but even the half-baked stuff gets you thinking. Especially the self-model stuff, and how problematic it can be to even discuss the concept in depth in languages that have the concept of a subject so deeply baked in.
I thought that at worst one could bounce off to the actual relevant literature like Thomas Metzingerās pioneering, seminal and terribly written thesis, or Sackās The Man Who Mistook His Wife For A Hat.
Blindsight being referenced to justify LLM hype is news to me.
I, too, have done the āall communication is manipulativeā, but in the same way as one would do a bar trick:
all communication is manipulative, for any words I say/write that you perceive instantly manipulate (as in the physical manner / modifying state) your thoughts, and this is done so without you requesting I do so
itās a handy stunt with which to drive an argument about a few parts of communication, rhetoric, etc. because it gives a kinda good handle on some meta without getting too deep into things
(although there was one of my friends who really, really hated the framing)
Explaining in detail is kind of a huge end-of-book spoiler, but āAll communication is manipulativeā leaves out a lot of context and personally I wouldnāt consider how itās handled a mark against Blindsight.
predictions for the trump admin?
Hot Take: the damage from RFK Jr will be limited by the fact that heās messing with the money for several large industries, particularly agriculture and pharmaceuticals. They have bottomless pockets and arenāt afraid to bribe the bribable. There will be damage, but heāll be crushed like a bug in the end.
Also, he clearly annoys the orange guy, can offer him nothing in return now that the election is over, and has already been the victim of a ritual humiliation (e.g. being forced to partake in a McDonaldās meal for the camera), which is the first sign of a Trump guy being de-emphasized.
Please sneer at this article. I thought it was pissweak myself.
Edit: āHear me out: RFK Jr could be a transformational health secretaryā
Will a Republican-controlled Congress allow for more government regulation ā even if it saves lives?
No. Next question.
should RFK Jr be able to abandon his numerous conspiracy theories about vaccines, he can be the most transformative health secretary in our countryās history
this is exactly the sort of shit centrists were writing about trump in 2016. I guess they canāt get away with doing that now so theyāre just writing the same pieces but about his underlings
Itās very āshould RFK Jr become a completely different person to the person he is and has been for years, he could really do some great stuffā
Jesus Christ.
This guy is a classic example of kook magnetism with a side of Dunning-Kreuger.
Job history is sus as hell too: https://en.wikipedia.org/wiki/Neil_Barsky
Some dweeb with a journalism degree gets a job at a hedge fund right out of grad school, bombs out, āgoes back toā journalism. Newspapers canāt get enough of this kind of guy; not surprised TERF Islandās paper of record picked him up nor am I surprised that they think RFK Jr might be good actually.
all your barsk are belong to us
With a touch of, and maybe Iām being mean, āthis particular bad thing shouldnāt happen to someone like me! it must be that no one else tried hard enough to fix it till nowā
I thought it was particularly telling that he says someone else in the āmetabolic healthā world warned him not to be mean about RFK. Take that as a sign that youāre in kooksville!
āā¦this particular bad thing shouldnāt happen to someone like me! it must be that no one else tried hard enough to fix it till nowā¦ā
Ugh, but also, good point. Itās like Engineerās Disease fucked The Secret and made something even worse.
I think it happens often enough when someone rich and/or famous becomes disabled or has a disabled kid. See the whole world of alternative treatments for autism.
Hopefully the established capitalists will protect us from the fascistsā worst excesses hasnāt been much of a winning bet historically.
oh no, nothing is protecting us, youāre 100% right there. Eating food is about to become a much more dicey proposition.
It is still safe to assume that the ghouls who run Pfizer and ConAgra will bend their resources to protecting the bag from a disposable nutjob.
mine:
prediction 1: he dies halfway through. funniest way would be another pandemic gets him
prediction 2: he doesnāt die. it will be exactly the same as the first admin but infinitely worse. everyone will hate and backstab each other, they will constantly get fired and rehired and fired like reality tv, there will be a constant dribble of horrible things happening, then in four years thereās a coup attempt
prediction 3: elon doesnāt last a year, possibly doesnāt even make it six months
prediction 1: he dies halfway through. funniest way would be another pandemic gets him
Iām anticipating an Elvis re-enactment.
If they do press conferences this time around, ever question should just be ādoes Elon approve of decision ____ ?ā Will drive Trump fkn insane.
Despite worrying my brains out about getting deported from my home of 14 years because I wasnāt born in this godforsaken place, Iām extremely excited that Elon will get fired in the next 6 months or less. Gives me life to think about him getting very publicly humiliated by an even greater piece of shit than he is.
At work, Iāve been looking through Microsoft licenses. Not the funniest thing to do, but thatās why itās called work.
The new licenses that have AI-functions have a suspiciously low price tag, often as introductionary price (unclear for how long, or what it will cost later). This will be relevant later.
The licenses with Office, Teams and other things my users actually use are not only confusing in how they are bundled, they have been increasing in price. So I have been looking through and testing which licenses we can switch to a cheaper, without any difference for the users.
Having put in quite some time with it, we today crunched the numbers and realised that compared to last year we will saveā¦ (drumroll)ā¦ Approximately nothing!
But if we hadnāt done all this, the costs would have increased by about 50%.
We are just a small corporation, maybe big ones gets discounts. But I think it is a clear indication of how the AI slop is financed, by price gauging corporate customers for the traditional products.
Thereās got to be some kind of licensing clarity that can be actually legislated. This is just straight-up price gouging through obscurantism.
now seeing EAs being deeply concerned about RFK running health during a H5N1 outbreak
dust specks vs leopards
The way many of the popular rat blogs started to endorse Harris in the last second before the US election felt a lot like an attempt at plausible deniability.
Sure weāve been laying the groundwork for this for decade, but we wanted someone from our cult of personality to undermine democracy and replace it with explicit billionaire rule, not someone with his own cult of personality.
Anyone here read āWorld War Zā? Thereās a section there about how the health authorities in basically all countries supress and deny the incipient zombie outbreak. I think about that a lot nowadays.
Anyway the COVID response, while ultimately better than the worst case scenario (Spanish Flu 2.0) has made me really unconvinced we will do anything about climate change. We had a clear danger of death for millions of people, and the news was dominated by skeptics. Maybe if it had targetted kids instead of the very old it would have been different.
Itās not just systemic media head-up-the-assery, thereās also the whole thing about oil companies and petrostates bankrolling climate denialism since the 70s.
When I run into āClimate change is a conspiracyā I do the wide-eyed look of recognition and go āYeah I know! Have you heard about the Exxon files?ā and lead them down that rabbit hole. If they want to think in terms of conspiracies, at least use an actual, factual conspiracy.
If H5N1 does turn into a full-blown outbreak, part of me expects itāll rack up a heavier deathtoll than COVID.
My professor is typing questions into chat gpt in class rn be so fucking for real
gentlemen, this means war
-me imagining myself paying to sit through that
Heās using it to give examples of exam question answers. The embarrassment
Iād pipe up and go āuhhh hey prof, arenāt you being paid to, like, impart knowledge?ā
(I should note that I have an extremely deficient fucks pool, and do not mind pissing off fuckwits. but I understand itās not always viable to do)
āSo, professor sir, are you OK with psychologically torturing Black people, or do you just not care?ā
It was there and gone fairly quickly and I wouldnāt say Iām a model student so I didnāt say anything. Iāve talked to him about Chat GPT before thoughā¦
I mean, that kind of suggests that you could use chatGPT to confabulate work for his class and he wouldnāt have room to complain? Not that Iād recommend testing that, because using ChatGPT in this way is not indicative of an internally consistent worldview informing those judgements.
Weāre going to be answering two essay questions in an in-class test instead of writing a paper this year specifically to prevent chat gpt abuse. Which he laughed and joked about because he really believes chat gpt can produce good results !
Iām pretty sure you could download a decent markov chain generator onto a TI-89 and do basically the same thing with a more in-class appropriate tool, but speaking as someone with dogshit handwriting Iām so glad to have graduated before this was a concern. Godspeed, my friend.
