chemicalnovae

I think _small_ is still doing a lot of work here. Yes, an MoE model doesn't "use" all of it's weights at once _but_ what it does use is still a hell of a lot heavier than the 20K toy shown in the paper.

The parent said exercise though right, not weight loss? Exercise can/does lead to weight loss but it's certainly not the only benefit and might not even be the most important one.

Sure, and most small organic molecules are an order of magnitude smaller than that definition.

> And? That span is within an individual's lifetime, which is not very long in the context of human history. As of now there's zero sign any entity with regulatory power is doing anything about microplastics.

Primarily, I _am_interested with health outcomes within my and my children's lifespan so that's the sort of time span I'm primarily concerned about. If the comparison to asbestos hold's true then we still have a _long_ time (long enough that any potential deleterious effects will be felt by all currently living and soon to be living members of my family) before any sort of regulatory action will be taken regardless of the health impacts.

> Also, why are you trying to deliver a point without looking up most of the examples I've listed? Do you expect that to be a convincing argument?

Because I'm _not_ your fact-checker. You're other examples may well follow a much quicker time-frame between discovery and strong regulatory action; I don't really care one way or the other, since at least one example shows a course of history which would play out poorly for those of us alive _now_ and exposed to increasing levels of environmental plastics.

I didn't go and look up the others but this argument by similarity, at least when applied to asbestos says the exact opposite to the claim you're trying to make. It's generally considered that industry/government was aware of the issues relating to asbestos in the '30s (and _started_ doing things about it then) but it wasn't until the 70's/80's (depending on the country) that its use was mostly stopped (and in places like Australia it wasn't outright banned until 2003).

He might not have been able to explain why _but_ I'd bet anything he would have wanted to if he could.

Isn't crypto used in these places simply a more convenient form on pinning to the USD than due to any inherent properties of the crypto itself?

> Please, (a) for your own sake and (b) for the sake of open source innovation, use the tool that you admit is better.

This is...such a strange take. To follow your logic to an extreme, everyone should use a very small handful of languages that are the "best" in their domain with ne'er a care for their personal comfort or preference.

> for your own sake

They're sticking with Pandas exactly for their own sake since they like being able to use Copilot.

> for the sake of open source innovation

Ohh by all means let's all be constantly relearning and rehashing just to follow the latest and greatest in open source innovation this week.

Tools are designed to be _used_ and if you like using a tool _and_ it does the job you require of it that seems just fine to me; especially if you're also taking the time to evaluate what else is out there occasionally.

100%, I had to pop into the HN comments after opening it because I wasn't sure which it was!

Though perhaps there is something we're missing that the target audience would more quickly identify it as real.

I mean, maybe deep fakes are _better_ at sowing distrust of information than other forms of propaganda but it seems disingenuous to treat it completly differently.

Don't be embarrassed, its what computers are for! I've done the same thing recently too. It honestly feels like a better use of a high-core desktop CPU than have it sit idle 99% of the time.

The essence of this is true of basically every PhD in a quantitative field. I think its one of the things that makes people walk away after starting.

Lots of chatter here about alternatives that all seem to miss the point that the alternatives suffer from the same filesystem performance issues that this release claims to improve. Might not be an issue for you but if you've ever had to work with a volume mount with a ton of small files (like php or node dependencies) then this could be a real life saver.