I disagree with the centralization thought. I'm working on a project to create a central AI platform for a very large corporation. As you can imagine, such a large company already has several different platforms made by different branches within it. Each branch has something they are relatively happy with. But corporate sees this redundancy as waste, so they want to consolidate it into one.

The problem with one is the same with any other single source: you are stuck with whatever they offer on whatever timelines they provide. You are sharing this with a hundred thousand other people, so it's not tailored to your use case, but either made as generic as possible, or full of irrelevant features. Both of these cases make it worse for you.

I have often made in house libraries either based on or completely replacing a free equivalent because the free one doesn't meet our needs and has extra complications we don't need. Yes, it comes with its costs, but sometimes it's worth it for something that genuinely meets your needs.

By forcing everybody to use the same common platform, you are either forcing them to work around the mismatch between the platform and their needs, or your are forcing the platform to provide for everyone's needs. Neither is efficient and it harms every other user.

If you are going to make your own, you had better have a good reason for it, but you shouldn't be forced into a single solution for everything - it can end up being more expensive than making a few special purpose applications.

As end user of one such platform, thank you giving me hope that at least some people are at least putting in an effort to make decent. We currently have a very weird mix of features applied to everyone and we are not all quite the same almost by definition.

I'm almost certain it was parody.

That was my take too:)

There is a big difference in the cost of a 5-nines up time system in a heavily space constrained environment compared to a home hobby white box used for some coding. The GPUs alone cost 10x for the data center versions compared to the gaming versions even with similar specs.

The cost of online services is also largely a result of the cost of training (though hard to say exactly what that number is). Assuming you are using open weight models at home, you aren't paying for the training - someone else is.

> The cost of online services is also largely a result of the cost of training

OpenRouter prices are somewhat simmilar to Antrhopic/OpenAI API prices. So I conclude that the hardware plus operating margin alone can genuinely produce prices way above what you'd pay if you had a subscription. Of course the primary unkown factor is average token use per subscription. Without that it's all wild speculation.

There's a fine line, as with alcohol. A single cup of tea might feel refreshing. A monster energy drink is going to feel like a fight or flight response. Some people are also more sensitive than others and some are addicted more really than others.

I guess I am a little bias here too - if my caffeine intake goes near 200 mg for the day then I get shaky. I have a friend who can toe the line at 700-800 mg! Which I read the other day that is the 'danger zone'.

Especially when there are many ways to do the quantization. You can get quite different results from different methods.

They are also an order of magnitude difference in number of parameters and that matters a lot.

I could perhaps understand ending the reporting rule, but deleting all existing records? That's very suspicious.

Govt should not have information unless there is a very specific and "in the law" reason for govt to have that information.

If there's no reporting rule, there's no reason for govt to have that information.

Each person takes turns. One person sings alone, then the other sings alongside their recording.

I store my belongings with a storage company. They rent facilities from a property owner. The storage company goes under while my stuff is still in the storage unit. I go to the property owner and say my stuff was in there. They say, no, that's owned by the storage company. Incorrect - the belongings were owned by me the entire time, the storage company was just holding it for me. I need to prove it was mine, but I don't lose property rights just because it's in someone else's possession.

Likewise if I store data with a cloud provider. Sure, Iron Mountain shouldn't just offer it up without proof, but if I give them notice of a legal case they should be sure to hold on to it, and if I win the case they need to hand it over to me. All I need to do is provide proof that I am the rightful owner of those goods.

> I store my belongings with a storage company. They rent facilities from a property owner. The storage company goes under while my stuff is still in the storage unit. I go to the property owner and say my stuff was in there. They say, no, that's owned by the storage company. Incorrect - the belongings were owned by me the entire time, the storage company was just holding it for me. I need to prove it was mine, but I don't lose property rights just because it's in someone else's possession.

You are a subtenant. The tenant went bust. The case law on this is settled. Depending on state, you either have a period of time to vacate the premises or you can/do become a tenant of the property owner. People have been subleasing rooms/apartments for hundreds of years, so every possible scenario you could imagine in this space has been litigated to death.

That is real estate.

American case law about data stored on computers is that whosoever owns that computer owns all the data on that computer. Which is why US privacy law is far weaker than EU privacy law (where you own the data about yourself). I think pets.com was one of the very few exceptions where the TOS said "we won't sell your data" and the bankruptcy liquidator said "it is an asset, we sure will sell it" and the courts said "no you can't sell it".

In the edge case that it wasn't Nine PBS requesting the Nine PBS data, you've now opened your company to a huge can of worms. Only OSS knows who Nine PBS is in their database, hence why, from a liability standpoint, it makes sense to say it's OSS's problem.

Also your storage company case is actually it's own can of worms, see the Lego drama from a couple months ago about that. It seems obvious, but in the eyes of the law there's more paperwork than you'd think.

You know how they say "possession in 9/10s of the law"? I.e., "9/10s of the law is just devoted to figuring out who owns what."

That is to say - I really doubt anything about this is remotely simple or straightforward, but I'm sure it's all lucrative for the lawyers involved.

I don't want them to. I'm not giving them that data myself and I don't consent to McDonalds giving them that data either, where that data is something I have to McDonalds. It's about information asymmetry.