EverythingIsFine
Well, is eventually fine
I know what you're here for. What's his bias? Politically I at least like to think of myself as a true moderate, maybe (in US context) slightly naturally right-leaning but currently politically left-leaning if I had to be more specific.
User ID: 1043
EverythingIsFine
Well, is eventually fine
12hr ago
But why now, when during Trump I and even the Biden Interregnum he was dealt quite a few defeats? I mean I'm well aware of what Trump means to the GOP and how he's exerted sustained pressure over the last decade but typically you'd at least expect recent events to provide more of a counterbalance, right?
Take Massie. His most notable stances are anti-Israel and holding administration feet to the fire about budgets and Epstein stuff. These are all issues where Republicans are, theoretically, quintessentially sympathetic (small government, anti-secret liberal cabals, non-interventionism). All of which are basically more popular now than any time in the past 10 years, right? Well, maybe not small-government spending priorities, but you get the idea.
EverythingIsFine
Well, is eventually fine
12hr ago
I mean, generally speaking a child is an 18 year commitment (and socially this is not usually passed on to fathers); the most we asked from any drafted Vietnam soldier was what, a 13 month tour? Sure, WW2 if you were unlucky it was maybe 4 years at most, WWI was only like a year, and the Civil War was a 3 year commitment (which ended early). This is not a fair comparison at all and I cannot really get past that so I don't understand what use the rest of the conversation is. They also all involved a pretty serious national or international emergency.
EverythingIsFine
Well, is eventually fine
12hr ago
Can anyone explain to me this chain of Trump primary victories? Normally I find myself pretty in the loop and things make sense, but I'm having trouble here. Trump as we all know has approval ratings in the doldrums and that extends even to a decent amount of historical loyalist, electorally - recent surveys show his endorsement is a drag in general elections in battlefield states. He also has a mixed at best record of picking primary winners. Yet he's scored several notable wins recently.
He has endorsed former Texas AG Paxton (and dogged by significant simmering corruption allegations), endangering the Texas Senate seat and going against sitting incumbent Sen. Cornyn. His pick for Kentucky Senate seat won the primary despite opposition from both Rep. Massie and retiring incumbent Sen. McConnell (notably, opposite wings of the party despite being somewhat anti-Trump). Rep Massie himself, they are reporting, has lost a primary as well (the most expensive House primary in history, in fact, drawing both Trump and AIPAC opposition) despite drawing support from other somewhat Trump-skeptic but influential right-wingers such as Tucker Carlson, MTG, and Boebert. Trump-opposed incumbent Louisiana Sen. Bill Cassidy finished third and didn't even make the runoff. In Georgia, perennial enemy (of 2020 election fame) Brad Raffensperger lost the primary for governor. Trump even took out five state senators in Indiana merely over their refusal to jump in the redistricting fight!
So why amid generalized disaster is Trump scoring so many primary victories?
EverythingIsFine
Well, is eventually fine
12hr ago
I'll go a step further and confess what I likely never will overtly say IRL: I don't like dogs. I don't hate dogs, I tolerate them to some extent, but I think they are smelly and gross, generally parasitical on the finances of most owners, annoying to be around due to barking/leaping at you/bothering guests, enormously disruptive for personal travel and to a lesser extent daily routine that I think I would resent, their QoL is often quite poor, and although somewhat useful for generalized companionship/mindless adoration and they empirically seem to fill child-sized emotional holes in people with decent efficiency, I don't really think the comparison is fundamentally sound.
I have yet to find the right balance of bringing this up when dating as sadly all too many women my age are dog-crazy and frankly it is actually a deal-breaker to me. Call me selfish but I don't want to be eternally second fiddle.
EverythingIsFine
Well, is eventually fine
13hr ago
People are not in fact OK with the current method of "bribes" but more to the point I've literally never had a conversation about modern bribery in US politics that didn't end with someone conceding "oh yeah well actually those rules on the books actually make at least a half decent amount of sense". People's perception of how bribery works (even a good chunk of otherwise smart and informed people) almost always involves a pretty inaccurate mental model that doesn't represent the facts as we know them.
EverythingIsFine
Well, is eventually fine
13hr ago
everyone to the left of Le Pen considers themselves interchangeable
It's incredible to me that you can write this with (presumably) a straight face. Really? Really? The infamously prone to squabble center, left, and far left? Consider themselves interchangeable? What? I presume you're talking about the assassination attempt but I not only don't see what specifically you'd be referring to but also fail to see how this translates to "there is only pro- and anti-migration" as if the two things are directly linked.
EverythingIsFine
Well, is eventually fine
13hr ago
I mean, the Trump crypto scandal and conflict of interest web related to it alone dwarfs literally each and every similar corruption scandal of the Biden administration put together, just on the basis of plain facts of the people and timing involved...
EverythingIsFine
Well, is eventually fine
22d ago
It doesn’t seem like the security was bad but I did see reporting done that the security level for the event was not proportionate to the number of high level leaders there (President and VP and a number of Cabinet members) and so the claim that security wasn’t of the appropriate quality is almost certainly true (and admittedly so, according to the government’s own rules and policies for this sort of thing). So it’s hard to judge the execution itself other than to say that there’s clearly a planning weakness that we can clearly see across all three assassination attempts.
Not to take away from what are probably legitimate complaints about the modern day, but:
More realistically they were just inherently stupid to begin with, which is why they never got around to agriculture or more advanced social organization.
Historically this is pretty much straightforwardly false. Like almost as flagrantly false as you can get. A study of early human history is painfully clear that there is an immensely strong causal arrow from “excess food production” to advanced social organization and accelerated technological development.
Virtually every single modern scholar that I’m aware of seems to think that Australia is pretty much impossible for early human society to do any better. The soil quality is not good partially because of the geological history (no volcanoes or glaciers, bad phosphate content, etc), so you need artificial phosphate imports to boost food production (not a thing until like 1900ish). No domesticable beasts of burden, for plowing, or even for food. Native grains are shit. Rainfall patterns are laughably inconsistent.
Opinions can be debated about more broad topics but the quote in question is not an opinion, it is simply a false claim. There’s room for other HBD type arguments I suppose but even the anti-woke scholars will tell you that the Aborigines had hit pretty much a hard ceiling in civilizational development and it wasn’t their fault. You can correct me if I’m wrong but I’m pretty sure even the early settlers who brought all sorts of stuff with them had a shitty time at agriculture until they could mass import fertilizers and stuff of that nature.
Now, can culture be maladaptive still? Sure. Just because the Aborigines did all right optimizing for their shitty environment doesn’t mean that the society they formed was logical, just, or fair or anything like that. Cultural selection pressures are mostly survival-based.
EverythingIsFine
Well, is eventually fine
22d ago
Eh, partially but not entirely. For one, these Named technically have allegiance to the “Gods Above or Below”. So there’s a bit of divine pressure. In that sense Good is more “do what you are told” and Evil is more “spit in the eye of the heavens” (also tolerate dirty tricks and blackmail and stuff in combat for example) and so there’s technically a hard divide there and in terms of the Roles that arise (narratively self reinforcing too which is part of the point/problem). So evil and capital-E Evil in this setting are overlapping but distinct.
Also the Dread Empire still will do stuff like assassinations, collateral damage, even massacres, that kind of stuff, and to some extent the main character participates in that too; part of the broader setting, kind of cleverly, is that the Empire doesn’t have good farmland or rather, much of their land is ruined, so they turn to massive blood and sacrifice rituals to magically sustain crop output and avoid starvation (and invasions of course for food plunder). There’s some plot threads that try to connect the macroeconomics to the political conflicts IIRC. The predominant human ethnic group in the Empire (there are several) have a culture of backstabbing and poisoning and such. Did I also mention that despite having many more mages, they deal in necromancy and diabolism as very prominent magical disciplines?
With that said yes, the whole overall arc of the series is an attempt towards pragmatism on the side of Evil but also seeing if Good can work together with them sometimes or even come to a kind of accord rather than be a constant kill on sight cycle of violence. And a central tension is in order to get that, you have to gain raw power first, but not so much that a better future becomes impossible. But it’s a series with several books so we see some variety and detail come out over time.
EverythingIsFine
Well, is eventually fine
23d ago
I don't really see what you describe seeing. I see some jokes about Patel, some commentary about increasing apathy around assassination attempts on Trump, a few threads related to today's news cycle that insult Trump or call him a pedophile, a few left-wing conspiracy theories (par for the course for assassinations + people ignorant of Halon's Razor), a bit of "if so many people want to kill him there must be a reason", a fair amount of criticism about Trump using it to promote his WH ballroom project, none of which I would characterize as "celebration".
Especially given that it's hard to take the assassination attempt seriously qua assassination attempt given how obviously idiotic the attempt was.
I think all the discussion about an identity-based lens downthread is extremely misguided. This is more neutral life crisis, I would argue. So the guy, clearly fairly smart, graduates CalTech in 2017 with a degree in mechanical engineering. Seems like he works for a bit over a year, then I assume is laid off or maybe he absolutely hated it, in the two months after he tries to work on an indie game, publishes one crappy Flash-era looking game that got probably double-digit sales at most (2 reviews on Steam in 6 years before the shooting led to a handful more sales), goes quiet and probably lives on savings for a bit, at some point I think moves home, eventually after a year or two picks up tutoring as a part-time gig for a few years. Highly likely here, reading between the lines, he gets some FOMO and feels in a dead end career-wise, goes back to school for a CS master's degree, then doesn't get a job in the big slump right as he graduates. He's Christian, probably somewhat typical left-leaning politics, by all accounts fairly mild-mannered, and like many Californians including Christians really dislikes Trump and his movement on a moral basis. Weirdly, he seems to like Kash Patel as the one exception (didn't include on list of targets). Moral enough to not target bystanders, but morally misguided enough to still commit murder despite Christianity being pretty clear about how that's bad.
This describes a fair amount of people, including actually myself to a loose extent biographically/politically speaking (if you squint a bit, but not too too hard). More specifically to him, smart enough to realize that political assassinations do have some kind of national political impact, not smart enough to realize these impacts are almost impossible to predict and often are counterproductive, smart enough to get guns and go cross-country without getting caught, but not smart enough to actually plan almost anything meaningful (though to be fair this at least partially checks out, since as we know being too detailed in your planning raises the chances of getting caught before you go, so it's not 100% stupid). If you wanna kill administration officials, usually you would have to limit yourself to 1 official at a smaller event to have a decent chance, so I assume this is an overambitious plan here. Clearly he didn't expect to live. I assume poor tactical awareness.
To circle back briefly and reiterate, it's not that political assassinations "don't work" it's that they don't consistently work in the ways shooters want them to. Often a stupid idea.
This doesn't tell us much other than the memetic influence of copycat shootings is starting to spread to more 'normie' people who aren't extremely mentally ill ('only' depression or similar here doesn't count). If history is any guide we'll see a few more assassination attempts of various types for the next 5-8 ish years and then it will slowly fade again, again purely based on the epidemiological-like traits the spread of acts like these are well documented to have (think for example certain modes of domestic terrorism over the years). I think this also slightly re-contextualizes a few of the incidents of violence like the Kirk shooter or Luigi Mangione as potentially less identity-based than initially suspected, when viewed in context, although the sequential nature of the memetic spread of this kind of political violence makes it hard to be super sure about that. I think it's a mistake to view these through primarily identity-based lenses, because memetic spread of this kind of thing naturally travels in subgroups but doesn't inherently correspond to groups themselves. Frankly I think nearly any loose political grouping is susceptible to this kind of thing, so the recurring "the lefties do this more" or "the righties do this more" that both left and right are constantly obsessed with in the last decade in particular is just a stupid argument.
If there IS a culture war angle to this? It's more about how fulfilling men in modern society seem to find life. That is, not very fulfilling, especially when you encounter job, social, or severe mental difficulties in the 15-35 age range or so. If this guy had more to live for, this kind of thing doesn't happen. At least that's my take.
EverythingIsFine
Well, is eventually fine
23d ago
Nike is much better at marketing than Adidas, that's pretty much it. Also I think most people don't actually care about the nitpicky rules and so the barrier was already broken, does breaking it a second time "but actually" really say anything new? However, marketing can only get you so far and Nike should be worried overall about losing such a large part of the running market, there's a reason the stock price is in the dumpster. Arguably, however, this isn't really a technology problem so much as an execution problem; Nike's loss in market share isn't really because rivals invented something Nike didn't, but rather Nike's shoes haven't had the right mix of quality control, durability, and comfort that hobby-runners and semi-pros want.
Somewhat recently I finished the series A Practical Guide to Evil. Pretty fun. I wish I had time to put it into better detail, but the world and setup is interesting. It's a sort of fantasy-with-superpowers type world? I don't know if I've really seen a book do it quite this way, however. The core concept is that "character archetypes" called "Named" periodically form in the world and with powers to match. For example, there's a Black Knight that shows up every few decades in the Evil-aligned nations, and a Paladin or whatnot who will show up in a Good-aligned nation. Roles can and do evolve if you survive long enough and are successful (e.g. you might start as a Squire and later evolve to e.g. a Mirror Knight). They have 3 limited-use (recharging) powers that are partly personal partly role-based. Fights between them tend to loosely follow some meta-narrative type rules, like starting a 'Rule of Three' set of conflicts between two rivals, or how Parties of Five tend to naturally form and are more powerful. It's a typical fantasy world (magic, nonhuman races, vaguely medieval) but with better than average worldbuilding IMO. The Evil-aligned empire on the continent exploits Orcs and Goblins to fight in their armies. You get periodic crusades against them, and periodic bouts of world-conquering too.
The context for the books however is that the main character is born and orphan in a middle, traditionally Good-aligned kingdom that usually gets the worst end of the stick when the Good nation of city-states to their West and the Empire to their East rampage all over their land. She is recruited as an Evil Squire, at first, to the Black Knight. However, said Black Knight and the Empress have conspired to "break" the typical cycle of Good vs Evil. They carefully try to avoid narrative traps in their fighting, stamp out Good heroes before they can get enough experience to start winning, treat the conquered middle Good nation abnormally well with expanded autonomy and economic prosperity, and develop the Empire's army into a more egalitarian and deadly fighting force, with expanded rights for nonhumans. This means that the MC's home nation is slowly turning... Evil! Mostly. But "Evil" in a, well, "practical" way as the title suggests. The idea is to be juuuust not-Evil and competent enough to prevent Fate spawning too many Good heroes from ruining everything. In their eyes, Good are jerks who are overly rigid in their thinking, while Evil has the potential to be pragmatic and level-headed about the greater good, paired with a resentment that narrative usually blindly favors the Good. Throughout the series the main character slowly adopts more and more of this attitude, but also tries to look out for her home nation and eventually grows quite powerful both personally and politically.
There is some character stuff of course, starting a bit tropey but gaining depth as you go on, some inventive fights (the main character often has to resort to tricks and cunning to win against the often narratively stronger Good heroes), and a surprising amount of politics and political maneuvering. And yes, the meta-narrative impact on fights is pretty interesting to see, especially among the more-experienced Named. You might get a hero who deliberately sets up a noble sacrifice as a giant fake-out, or deliberately as part of their fight strategy sending someone to wander around and thus rely on divine providence to guide them to the exact right spot, or a villain who tries to avoid their monologuing tendencies which inevitable backfire, but sometimes leaning into As the series expands you do eventually visit most parts of the continent, other nations' politics and alliances often become highly relevant. You've got a surprisingly deep and fleshed-out history of the nations involved. Which I've always really appreciated in series, like for example Wheel of Time was great in part because you ended up actually using the map over the course of the series with a nice sense of scale. A fair amount of the series is mostly war-stuff, though, which you either love or hate.
And you've got some comedy too. There's a city-state to the south that is an exaggerated democracy, where everything is put to a vote and the bureaucracy is intense and they almost never agree to do anything, but also has secret police who are constantly trying to guard against Tyranny. We get periodic epigraphs from some of the Named former Emperors, from Emperor Irritant, the Oddly Successful (the best unexpected quotes), Emperor Traitorous (infamous for several quadruple-crosses and such), etc. that occasionally give Hitchhiker's Guide vibes. Anyways, it's originally a web serial and that shows at times but nonetheless was a very fun read if pulp fantasy is your jam.
As an aside, in the original serialization, this world does apparently have a reason that the world is stuck at a certain technological level. Apparently there is a race of "gnomes" which are implied to be super-advanced, flying or space-faring or something, that will deliver a warning if an invention happens or line of research is pursued they don't want. If the warning is ignored they basically nuke the city from orbit. Is this elaborated on anywhere else in the novel aside from a few random mentions? No. They in no way affect the plot. I guess that's one way to set up a fantasy world's tech level... (IIRC in the published, edited novelization which is in progress, the second of ~6 currently about to come out, which I do recommend as an improvement over the original, this idea was dropped in favor of some kind of Fate hand-wavy thing, but IMO the gnomes are more funny)
Haven't really tried the other two much myself, high costs kinda scare me off where I find it hard to believe the value is there for my use case, but I'm looking quite closely at Deepseek V4 Flash specifically, because the cost-to-performance seems to be pretty insane.
I mean, the official API price is $0.0028 per million input (cached), $0.14 (cache miss), $0.28 per million output. Has a 1M context window, and per OpenRouter, because their stats have above a 90% cache hit rate, this means the effective weighted price per million input is only $0.015 per million. That's really crazy. I need to test a bit more to figure out where I'd place the intelligence exactly, but...
NONE of the current frontier 'cost-efficient' models come even remotely close to that. For comparison Gemini 3.1 Flash Lite is 0.25/1.50 input/output per 1M, Gemini 3 Flash is 0.50/3.00, GPT 5.4 Nano is 0.20/1.25, GPT 5.4 Mini is 0.75/1=4.50, Claude Haiku 4.5 is 1.00/5.00. Sure, it's not as good at coding as Gemini Flash, allegedly, but also allegedly it's better at agentic workflows. Those are some pretty significant gaps, approaching an order of magnitude in some cases.
So yeah, Pro is also very cheap and that might make some waves, but contextually Flash is SUPER cheap. Like, obscenely so.
This to me is a big deal because part of what makes AI so compelling is the cost/benefit ratio. With a model like V4 Flash, especially input-heavy workflows, there are plenty of scenarios where it's literally cheaper to throw 5 different approaches at the wall and pick the best than to make a single attempt with a model that's just a hair smarter. We'll see how well it does when encountering actual codebases and such, but I find that it might potentially enable a slightly different type and set of workflows than we're used to.
It's hard to say for sure these days because especially with agentic coding the harnesses are so important (and often what works for certain setups doesn't transfer that well, including across generations of models). I'm curious if someone will figure out a good way to leverage this new cost-benefit balance, because it potentially changes e.g. how you might spin up subagents quite a bit. Although possibly as I mentioned the model is just a bit too stupid to do a large enough range of useful work. We'll see, gotta figure out how much is benchmaxxing vs inherent quality.
At the same time, the Claude shift in tokenizer probably long-term helps efficiency and intelligence, but short term you're looking at a 10-30% flat increase in costs on higher token costs alone, before you get into the token efficiency of the models themselves, at least per the numbers I was looking at initially.
EverythingIsFine
Well, is eventually fine
23d ago
It seems to me that internships with placement opportunities not only signal "I'm aiming for a higher job" but are also harder to get relative to full-time IT support type positions. So not only is the opportunity cost higher if you were to not go with the internship, but the signals align with what you likely want for your career. I'd go with the internship unless the finances don't allow it. Sure, it sucks to then go around and job-hunt a second time, but there's nothing stopping you from job-hunting a bit again especially towards the end of the internship (plus if you find another opportunity, it also inflates your value a bit if they want to bid to keep you).
Although where exactly you are education/career-wise might matter a bit. The one caveat here is how good the reputation of the internship company is. 17 seems pretty low, so I'd wonder if that stinginess extends to the rest of their org.
Just my 2 cents, gut reaction
EverythingIsFine
Well, is eventually fine
24d ago
There's a small subset of commenters here that are experts in what is covert, effectively bad-faith argumentation, perfectly calibrated to the rules. You will post a substantive comment arguing/claiming/observing A B C D and they will nitpick B, and respond to D with their most full-force argument. Of course, ignoring entirely A and C. If you're interested in light and truth of course, this is a most strange way to respond, but many of these people aren't. To many of them forums are about WINNERS and LOSERS and they are determined to be 'winners'. Tragically, most people in this forum seem inclined to let them get away with it. Who cares if someone ignored A and C? Disagree = downvote, rebuttal to D was strong = upvote. "Hey you ignored my point" = crickets. Then, for bonus points a week later, a toplevel post with a gish-gallop of blue-links chosen for maximum effect. It's effectively the loophole here.
Personally I'm not interested in playing the game back, I value my own integrity too highly, so I'll begrudgingly accept reddit-hivemind downvote patterns, block a few people who are particularly egregious, and accept that I'm here purely as a mix of fun and intellectual hobbyism. I'm not here on a crusade to change minds because the personal cost is too high. I'll say my piece and maybe people will find it interesting. But if you're wondering if this community lives up to its billing the answer is no. Is that depressing? Yes, a bit. Is it ironic considering some of these bad-faith people are some of the very same lamenting the slide of society into a "low-trust" paradigm, which is, of course, allegedly the fault of progressives and minorities? Yes. It's not even really the mods' fault because a) their job is really hard and b) some of these people are just way too good at skirting the rules in just the right way.
EverythingIsFine
Well, is eventually fine
24d ago
No, insiders are still visible even if small in number because their relative confidence is so out-of-distribution. They will be almost always betting larger amounts of money on more-rare events in a way that is noticeable. Partially due to the nature of the thing, but also due to the amount of reward necessary to outweigh the large inherent risk.
OK, sure, maybe the CIA might use this on purpose as psyop stuff, but that's a classic case of shooting yourself in the foot in a more broad sense. It's not in the US government's general interest to have tons of ammo for conspiracy theorists (mainstream or niche)
EverythingIsFine
Well, is eventually fine
24d ago
Somewhat (in)famously just recently in Utah they found guilty a woman who tried the fentanyl thing exactly. And failed. Partially because the husband literally suspected her of poisoning him and told others as much, partially because obtaining the fentanyl when you don't previously have a drug network is nontrivial (the housekeeper who the lady knew to have had a prior drug conviction was hit up for some), and partially because more broadly speaking even smart people (idk if Kouri was though) don't kill people without a motive and motives are pretty legible. That's kind of a big one.
EverythingIsFine
Well, is eventually fine
1mo ago
Good one. I'm totally capable but if there was something easy and pre-existing it could save me some tokens, you know?
EverythingIsFine
Well, is eventually fine
1mo ago
Except crypto was almost always purely in the realm of theory-applications.
With AI, right now, I can do things like generate custom flashcards for subjects I'm learning (job interview prep). I can get more in-detail answers about random questions without spending hours on Google piecing things together (just yesterday, asking for details about how stomachs process different macronutrient profiles). I can generate custom mini-apps for a wide variety of tasks (recently I made a custom task-selection spinner for my todo list that weights the important tasks more than smaller tasks, while occasionally mandating a break). It can make sure an email I send to a recruiter doesn't have obvious mistakes or commit a faux pas. I can get personal advice of at least middling quality without friction on a wide variety of topics. Obviously, it can code really well, and that touches my field very directly in a lot of ways. There are plenty of other use cases, too. These aren't "lines of code" type accomplishments, they are concrete deliverables of various scopes. Some of which were previously high-friction or even impossible.
Sure, some of these are gratuitous or busywork, but they are all real. Crypto stuff was like, "what if the government keeps track of property listings on the blockchain" which is a) something the government already does mostly just fine and b) obviously never happened and c) would have required very significant network effects. And currently, crypto is extremely useful for pretty much exactly two types of people: those who treat it like digital gold (it does OK at that) and criminals who can move money around that's difficult to track. Nothing else. So sure, in that sense it was real, but AI plainly can do more than two things and will continue to do more than two things even as hype dies out.
And sure, my IRL friend will give me better advice than Claude will, but there are some things that are so low-stakes that it would be disrespectful of their time to ask or discuss. Paradigms like that are all over, because of the speed and cost AI offers. In that sense, it's more like the Industrial Revolution, where speed and cost enable things to happen that previously were functionally impossible at scale. In fact most of the Industrial Revolution was about things that were already feasible to do, but were cost-prohibitive (or took too long). This in turn generated new industries that were previously only theory. Now, I don't think AI will have that level of impact on society, and I'm also not sold on it 'creating new industries' at all, but probably it's somewhere on the level of the impact similar to the invention of Google at least?
EverythingIsFine
Well, is eventually fine
1mo ago
Somewhat an aside, but I consider that first link to be a first-degree chart-crime. First of all radar plots are inherently iffy, since we pay close attention to the "area" and the area is highly dependent on how the categories are organized (a "spiky" radar plot has much less total area than if you sort the axes to create a "lopsided" plot, despite showing the same information). This is a little bit defensible if the adjacency of the categories is obvious and inherent, but they frequently are not. For example, "Occupational: Writing Literature and Language" is NOT next to "Text: Creative Writing" for no good reason at all. And furthermore, what is the scale of the chart? It's "Arena rank"... which is NOT equally spaced. The chart implies that the difference between #1 and #2 is the same as (or even slightly bigger than, considering how the radar chart "expands") that between #3 and #4, but this is plainly not the case. They should be using some kind of actual score instead, perhaps a scaled one. Sure, it allows consistency across axes, but if we are comparing a model to its successor, the rating scale definitely shouldn't be implicitly including other models like it does now (in one spot it drops from rank 2 to rank 5, does this mean in that category some other model class does abnormally well, or that did Claude truly degrade?). Even worse, the center of the plot, usually a natural "zero", is not a zero at all - it's rank 6. There are, as you know, dozens and dozens of models in the rankings, so rank 6 being a zero score is totally nonsensical.
EverythingIsFine
Well, is eventually fine
1mo ago
It's not about profitability, it's that they got a giant wave of users but not enough compute to fill that demand. So, it's pretty obvious what must happen next, you do some mix of increased mandatory token efficiency (adaptive reasoning) + stricter limits (across the board, free and paid, but mostly targeting the super-user hogs who theoretically will pay for extra API usage after limits run out).
I will say though this probably bodes poorly for Claude in the near-medium term, because ChatGPT had the same thing more or less happen with their 5.0 launch (forced adaptive model selection for mandatory token efficiency) and it definitely took the wind out of their sails for at least 4-5 months.
At any rate, however, I strongly, strongly disagree about this empowering the skeptics (or being evidence of a shift against AI adoption). The fact that people are whining about problems with their tools is selection bias. It's kind of like the classic armoring spots on the airplane that didn't have holes (because they didn't survive to be examined), in that people wouldn't complain so vociferously if they weren't so needy for the tool in the first place. The complaints to me are evidence of a generalized latent enthusiasm, not pessimism. In the grand scheme of things, it's far, far better for a company to have complaints that users can't get enough of their product, than it is for the product to be simply ignored. In the near term, I expect a decent chunk of users to swing back toward the OpenAI offering, Codex (which is undergoing a PR blitz of sorts right now)
EverythingIsFine
Well, is eventually fine
1mo ago
protip: the smaller bananas are almost always the most flavorful ones. Maybe the same is true of strawberries, I dunno, but I have found the banana tip to be pretty consistent.
I know this is a bit more off-topic, but it's been extremely hilarious to see various right-wing communities essentially suggest or re-invent the American left-wing commune (that actually has a very rich and deep history that doesn't even have much to do with communism exactly) from first principles. It's uncanny. Understandable, but still uncanny. I guess in some sense communitarian counterculture homesteading has intrinsic human appeal, but in another sense they don't spring up out of nowhere and so their presence I think usually says something about moral and socio-political climates, beyond just "horseshoe theory is correct".
More options
Context Copy link