No. It's not your identity, it's a piece of software. "I use ____ for as long as the benefits outweigh the drawbacks" is what you should be thinking.

OpenAI offers that, or at least used to. You can batch all your inference and get much lower prices.

This makes no sense. It's not like they have a "slow it down" knob, they're probably parallelizing your request so you get a 2.5x speedup at 10x the price.

What they are probably doing is speculative decoding, given they've mentioned identical distribution at 2.5x speed. That's roughly in the range you'd expect for that to achieve; 10x is not.

It's also absolute highway robbery (or at least overly-aggressive price discrimination) to charge 6x for speculative decoding, by the way. It is not that expensive and (under certain conditions, usually very cheap drafter and high acceptance rate) actually decrease total cost. In any case, it's unlikely to be even a 2x cost increase, let alone 6x.

All of these systems use massive pools of GPUs, and allocate many requests to each node. The “slow it down” knob is to steer a request to nodes with more concurrent requests; “speed it up” is to route to less-loaded nodes.

Right, but that's still not Anthropic adding an intentional delay for the sole purpose of having you pay more to remove it.

Oh, of course. That’s just conspiratorial thinking. Paying to be in a premium pool makes sense, all of this “they probably serve rotten food to make people pay for quality food” nonsense is just silly.

Yeah I don't understand. Is it actually saying that fast mode is ten times more expensive than normal mode? I cannot be reading this right.

But that's the killer feature for me! I always forget the little commands I've written over the years, whereas a leading comma will easily let me list them.

Their incentive is to keep the browsers good enough to not lose market share. Other than that, the incentive is to either close the web down, or to make the experience as shitty as possible without leading users to switch away, so they can steer users towards the more closed-down native apps.

Unfortunately, companies have an incentive to put us into walled gardens, so the only company that actually cares about the web is the company whose only business model is selling a browser.

I wish it weren't. I would have gotten a lot more mileage out of "force a yawn, see what your mouth does, and do that" rather than "more space, more space, open up!".

Have a look at Complete Vocal Technique.

https://completevocalinstitute.com/complete-vocal-technique/

Their work includes pedagogical research to develop a consistent terminology which abandons lots of outdated and confusing terms such as you mention. No more ambiguous words like "project" or "space" or "support".

Their research also includes using endosciopic cameras to directly observe the vocal tracts of professional singers.

I've not actually trained with them, I just like their research and approach.

The ambiguous word “support” is listed on the page you linked as the first of their principles

That looks really useful, thanks!

I'm danish. CVI is the source of my inspiration

Seconding both points. I'm not one of those cases, as I could already sing decently, but I've seen people go from "terrible" to singing professionally.

I also agree that the linked page isn't useful, it's more of a glossary than anything, but then again, I'm not convinced that a distinction between head voice and chest voice actually exists. I've never been able to tell any qualitative difference, as opposed to, for example, falsetto, and the community can't really agree on whether they actually are a thing or not.

This is a lot of cryptography, but how is it better than the hundred previous attempts, that simply hashed the input?

Most prior attempts reduce to hash(master || site). Bastion treats password generation as a cryptographic protocol with explicit invariants, not a convenience function.

An important note is Hashing ≠ memory-hard Hashing ≠ unbiased sampling Hashing ≠ domain separation Hashing ≠ rotation without storage

As I understood it, the "issues" were more like todo list items of "look into whether this is an actual problem" than "this should be fixed".