Hacker Newsnew | past | comments | ask | show | jobs | submit | AnotherGoodName's commentslogin

The Houthis also explicitly allow Chinese vessels and Chinese ships have been able to traverse the red sea without issue. Which means China actually has more ability to access oil than other countries right now.

I suspect the Chinese won’t be happy with the pipeline and refinery attacks on Saudi Arabia but i also assume that there will still be plenty of Chinese bound oil for the foreseeable future.


China has a backdoor for oil from Russia and Iran (https://www.atlanticcouncil.org/blogs/new-atlanticist/the-ax...) which is part of what feeds their massive reserves. Russia and Iran have an interest in continuing this trade since they're sanctioned by the US and most crude is priced in USD. It's not a major source of their oil due to threats of sanctions from the US, but in a pinch they can ramp it up if needed.

Fwiw i noticed you absolutely can't reasonably get a US Via consulate appt for some countries without AI right now so the requiremnts already begun.

The consulate is booked out for 3 months and only opens 3 months out. All openings are snapped up before any human could get one. They have a lot of anti-script blocking but AI seems to get around them. So there's an arms race, you pretty much need AI to book it (a basic hand written curl script isn't enough).


Oh wow, that’s aweful. Where I used to live this would happen with driving tests. Driving schools would book out all the tests and then re-assign them to their students. So at some point it got so bad that when you’re buying a driving lesson you’re also effectively buying a right to take the test. What could the government possibly do to stop this? Some highly technical solution? How about just making it so you can’t assign the test to another name? Hmm that might be a pretty obvious solution…

You could require a phone call or mailing a form (in the post) I suppose.

Steam also had a pretty nice idea with the Steamdeck. You book appointments for a month in the e.g. 2 weeks prior, but those who get booked, and when they are booked, are randomly chosen amongst all people/times in those 2 weeks. And then you repeat the same process next month and so on.

Exactly. There are ways if you truly want to avoid that race. It just means the provider doesn’t care about bots

It could be as simple as a new model (astra) was released which takes more resources combined with a surge in usage due to novelty took down OpenAI. Meanwhile everyone at big companies have the ability to switch models and moved to Anthropic pushing it too over the edge.

People keep saying this "everyone can switch" thing, but it's not my experience at $VeryBigCorp. We don't have an Anthropic contract at all. Is this different in other places? The bigger and more bureaucratic an org is, the less I would expect it to have contracts with all the providers. Curious about others' experiences.

At my company (not massive but not tiny either - I think its about 8000 global employees) we get a choice between pretty much all available Google, OpenAI, Anthropic, XAI models.

Using an agent-agnostic harness like Pi switching is trivial - I run into occasional disconnects and slowdown and switch quite easily.


Yeah, I think there are lots of places that have a semi-official preferred AI vendor but also some backup subscriptions floating around. For example, at my workplace, we generally use Claude, but I also have some kind of Codex subscription too, which I'd use if Claude went down.

I actually think they’re hooked on models they can fine tune.

You can’t further train the closed models. The open models can be fine tuned for your company. Big companies fine tune models on all the internal systems and documentation, not just through .md files (you’d blow up the context trying it that way) but actual fine tuning of open weights models. A low tier but open weights model actually beats frontier models when you do this for a specific task.

I think the frontier providers need to have a way to isolate instances (bedrock style?) and allow fine tuning to compete. Big companies are absolutely fine tuning models right now and getting better results than even the best frontier models for their use cases.


The mean and the median wealth differences is the most striking. USA is second on mean wealth and 27th in median (just ahead of Greece). A few trillionaires pulling it up with nothing trickling back down.

Still wealthy either way but I always find it a bit jarring how skewed the US self perception is since people generally experience something closer to the median.


Note that Caplan is focusing on medians and percentiles in his table, not on means.

> A few trillionaires

There has only been one of those in history, and he is no longer one.


One i hate is products that have statements ‘high in vitamins’ and then you read the ingredients and see they just added a crushed vitamin tablet.

Coca-Cola dismissed the allegations as "ridiculous," on the grounds that "no consumer could reasonably be misled into thinking Vitaminwater was a healthy beverage"

https://en.wikipedia.org/wiki/Energy_Brands#Legal_disputes


Is this supposed to be bad? Fortified foods have helped alleviate many nutritional issues like B12 and iron deficiencies. In both instances, supplementing can be the preferred form of ingestion compared to whole food sources:

- https://pubmed.ncbi.nlm.nih.gov/29611716/ - https://pmc.ncbi.nlm.nih.gov/articles/PMC9268692/ - https://pmc.ncbi.nlm.nih.gov/articles/PMC12681447/


I mean I've been seeing some cereals lately that say "high in protein", just for half of their estimated amount owing to milk you'd be presumably adding.

They've been slapping the "high protein" label on pretty much everything that (naturally) contains any protein at all here lately.

I think this occurs due to the prompt. LLMs are actually text completion/translation focused in architecture. We just give them a prompt along the lines of “the context is that you’re a world leading expert now complete the response”.

They need the prompt to encourage expert outputs but unfortunately we also get ‘pretending to be an expert’ outputs since there’s a large amount of polluted training data for this.


Based on what, "lot of people say"?

Seems completely arbitrary to me honestly. Clearly some adjustments to the criteria are needed because this list reads unmaintained vs maintained as opposed to complete vs incomplete.


I'd love to keep improving so if you have some examples that would be great


Cyberpunk 2077 is labeled as "Not yet. Released, but still in active development", but it's clearly a "Done" game, as in they've added a ton of stuff and development slowed since then. Still patches, yes, but should that mean "it's not done yet"?

I agree it seems more to be about "receives patches still or not", which is different compared to if major feature development essentially stopped.


Cyberpunk is a bit of an odd one. They have pretty long stretches between releasing major content. It certainly feels "done" but it had a LONG tail of development after launch. Phantom liberty is most likely the logical place to say its fully finished but cd projekt red haven't definitely stated there is no more content coming (and there might be with edge runners 2 about to release). So I would argue the assessment is fair, its most likely done, but we can't say for sure so we flag it still. Given enough time has passed, with no communication and no more updates, it will eventually flip over to done.


I'm guessing for similar reasons you have The Witcher 3, a game released more than a decade ago, labeled as "Not done yet"? I'm not sure how this is supposed to be helpful, especially not when the labels will arbitrarily change back/forth as stuff is announced or it was a long time ago you heard about anything.


Most games do not change, is the leading principle. But bear this scenario in mind: A game is released, receives post launch support and then the developer posts a blog post stating "This was the final update, thanks for playing my game, there will be no more updates". The system reads that, makes the anaylsis and we mark it done. Then three years later the developre comes back and goes "PSYCHE! HERE IS AN AWESOME NEW UPDATE". Well we can't account for that, so the system will see the update and change it from done to not yet.

This is an imperfect system because its trying to assess a situation that has no actual contract. But for 90% of titles or more, when a game is flipped to done it won't suddenly be revived at a later date. The Witcher 3 being hte exception not the rule.


> "This was the final update, thanks for playing my game, there will be no more updates"

So you're only going to call abandoned and/or end-of-support games "done"? Even if they were abandoned because there wasn't more that needed fixing/balancing, you're still using a definition I have never seen any other human being use. If it would be done at a given version if the dev just left, isn't any game that provides access to run any version better than "done"? (Minecraft, for example.)

For any multiplayer game this seems to just favour yearly release slop over actually good games with good teams behind them.


So a game is only done if you neglect it?


I wouldn't say this is making a judgement on neglect or care. Its just making a judgement on "Do we expect any more content coming for this game". A neglected game over time will become an answer, as will a cared for game with a "We now consider our journey complete" developer post.


The entire page of ‘net yet’.

Counter strike 2, Dota 2, GTA 5, Red Dead Redemption etc. the list goes on. Everyone would consider these games maintained not incomplete. Maybe just merge not yet into complete? The ‘pending’ is more reasonable the ‘not yet’ list is clear nonsense.


But they aren't. They are live service games that are constantly getting updates and new content. They are "Complete" in the sense that you can play them, and live service games fall under a weird area where they are "finished" in the sense that they have active communities and are wildly considered released but they aren't "done" in the sense that they are always evolving, changing, and adding content.

That is why most of those games listed will have the "live service" tag. This isn't really a website for figuring out if live service games are complete. Its more for determining if a singleplayer game you are interested in has released every post launch dlc.


When in Rome (or when on the web)… You need to align with the expectations of users of your site for the site to have value to those users and this doesn’t align with user expectations.

People do wish to be informed if a game is missing core content and features. People in general don’t see it as a negative if new content is added to an otherwise complete game. You have a mismatch here with general consensus.

This mismatch means this is a site that only has worth to yourself. Which is ok if you’re ok with that. If you want others to gain value you have to change to meet their expectations.

So it’s up to you. You can stick to this idea and keep this website as you want it or you can adjust to the feedback given here and accept that these games are considered complete by the vast majority.


i would maybe clarify in the about page that it comes from a campaign lens versus a multiplayer lens


This is also true for HITMAN 2/World of Assassination. This game's whole premise is episodic releases and new targets. At this rate, the game will never be "released".


Pretty sure Elden Ring is done.


Elden ring is still releasing content to the base game, like the Tarnished Pack which literally released today and adds some starting classes and other things. So the arguemnt which is made is that its basically not "done" and there is still more content to come


Thats overdone - almost burnt. Felt "done" after about a year / 9 or 10 patches.


There's a movie "Weird Science" which i always appreciated if only for breaking the rule twice in two words.


i before e except after c or when sounded like a as in neighbor or weigh.

But there are exceptions. English language orthography is a weird science.


Also the endless online debates of ‘is this post made by ai? What about those images, that video or that music?’.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: