mmonaghan

I see posts pop up with this sentiment often and I'm of two minds:

- llms let us build exactly what we want fairly quickly nowadays, or at least a prototype and

- being able to build these features (or separate products) quickly is exactly the trap

> It’s the same test I keep applying when picking the right problems to solve — does it make the boat go faster?

Who knows this though? It's rarely engineers. Product often has a good sense within their lane. Leadership usually (at good companies) is mostly aligned, with some nuance depending on who's speaking.

It's a good sentiment but rarely a single person's decision.

As I said, I see these pop up and I'm usually annoyed because what new take could someone bring at this point, but I read them every single time :p

Build it, post about it, and see. This seems like something that would take an hour with a decent llm. Perhaps another hour or two to get the ui in a decent enough state. Probably shouldn't plan on making money on this though

People love hating on x and nikita but both the app & web experience have improved dramatically over the last year+. He's done a great job in a year and there seems to be even more internal stuff he's improved that we haven't seen.

Platitudes sound nice but in the real world you need incentives and there are plenty. Other than real madrid and barcelona and a few italian clubs, very few euro clubs are run as non-profits. Easy to argue that maximizing financial return is the reason football has grown in prominence over the last 20-30 years.

I say that but I also agree i dont think fifa or any of the regional orgs need more or any financial incentives. Keep profits for the club, keep the governing body to governing, and the regional orgs to organizing.

I actually think this is an area where the us could do a lot of good - sweep in, clean out fifa and as many of the regional orgs as you're able (they're connected), and keep the pressure on. Only the US is really able to do this and we hate cheating in sports.

I think distillation is inevitable and pretty hard to prevent but I also dunno why anyone would spend billions to train frontier models if there's enough distillation going on to get near it with much lower spend.

This is true for any company following an open source strategy. You'll have the weights but you'll need to run inference, figure out your system prompt, sampling, quantization, etc etc. Loads of tuning.

elasticsearch the first example that comes to mind. you can run it yourself but elastic gives you so many lessons learned and tunes ootb that it sings with relatively little effort, though still reqiures some.

about a million dbs i could make the same comparison for

Agree but I think they'll spend whatever they can to subsidize and break dependence on US models. Whatever isn't spend on US models will be spent on chinese models. Europe will remain a small single digit % unless they get their act together.

Who's got a monopoly though lol?

how profitable though

> the idea that anyone could own that is obscene

What if china ends up owning it? Do you think that's better or worse for the world? You're commenting your open opinions on a site run by a company that could not exist in china and cannot today. You can ask ant, xai, and chatgpt models questions and get answers that do their best to reflect the world. There's a set of questions you cannot expect trustworthy answers for from chinese models and you think they're stopping at those few questions?

get real bro

They're dumping, dude. They subsidize important industries so they can give it way so cheaply that free markets can't compete, and when those companies fail, they swoop in and make a profit. uber did the same thing with taxis! This isn't a free resource - it costs billions to build out the infra and maintain the inference to provide this, and you think they're doing it out of the goodness of their hearts lol.

Come on bro at least admit you're happier giving china your data than the us lol. You're definitely not using european models I can tell.

People say this but windows and mac won the os wars and consumer linux distros are dead. No one I know uses open weights models to code day-to-day and very few use open weights models in prod for most llm tasks. This is less true for image/video stuff, where I think the competition is more vibrant.

I think the best gauge is company spend. Open weight models are a very small share compared to frontier models, and I'd bet that many companies mostly using ow models would switch to frontier if they could afford it. I also think most companies who are picking ow over frontier probably have deeper financial issues they should focus on.

Flailing. I hope its because he fears for his job and not because he senses yet another price-fixing scandal on the horizon.

ehhhh I'm not a corgi stan but I'd bet they just took the design & copy, which is totally fine imo. Often better to just take another's design instead of spend a bunch of cycles figuring out flow. This is doubly true for secondary pages/layouts/features.

If I were them, I'd have changed copy and probably done some internal testing to smooth rough edges/improve where needed but sounds like they're moving as quickly as possible.

If they did just copy paste code, straight to jail...

The country has changed massively through voting in the last 25 years. I would rather not change to a system where political groups attack others with guns and bombs to change things, because that's the alternative you're proposing.

Touch grass, find a significant other, and build a life.

I think there's evolution at play here - if you dislike AI enough to opt out of using any ai-generated code, you will likely suffer. I think there's definitely a conversation to be had about whether to disclose AI use or not but that's a separate issue if you assume that everyone is using it in some respect.

Tried everything and love 1pass. Dont want to have to think about it too much.

I think this is tentatively good for bitwarden - making money means you can more easily invest in the team and product. Counter to the prevailing notion in comments here, I much prefer a vc/paid product for security-critical tools.

Hope they didn't wait too long before deciding to kill the free tier.

Good for them. Much easier to build a great product if you're making money from it. I tried bitwarden a while ago but ended up going with dashlane for a few years. I'm on 1password now and really like it but more competition is always better.

This is clickbait meant to make you angry - the move has been planned for more than ten years, well before datacenter power usage was an issue. NV Energy has made it clear that they're ending their wholesale agreement and for some reason, Liberty and Cali/Nevada have made no effort to fill the gap.

If you're pissed or worried about this, write your state reps and demand more power gen. Nuclear, solar, geothermal, whatever floats your boat - we need to build more of it all around.

Kimi nowhere close to opus on extended use but definitely highly competitive with sonnet. I will probably end up using kimi for personal stuff when I find some time to get it running or get a non-anthropic/openai harness set up on my personal machine.

I think its two systems masquerading as one - employed-and-insured and everyone else.

If you're the former, it works great. If you're the latter, it can be mediocre to BRUTAL. Medical debt is our #1 or 2 cause of bankruptcy iirc.

Regardless of which class you are, if you can access the care, our outcomes are the best in the world for most things.

Yes

If you can successfully integrate with quest/labcorp and one or two of the big ehrs, i will happily pay double what you're charging.

after reading it, it's super positive and really great. I wouldn't consider myself the target audience for this, but ill probably work it into my morning practice a little for a couple weeks.

trendslop is a great way to describe it

Yes, it's been insane. Hitting limits by noon since late last week. Not even using opus 4.6 with 1mm context, just normal opus and a bit of sonnet/haiku.

I hate these types of study because yea, in general people are really moral and want the right things but when it comes down to specifics, very few people are voting for a new nuclear plant near their house (I would love to vote yes on this).

Finally someone's taking another swing at this. Can't even remember what MSFTs last pmr tool was called but I used it and it was decent enough.

I've been wanting something like this for ages so let's hope this one works.

I did not expect there to be any bright spots but I hope this turns into one of those things (like the 2002 AUMF) that just lasts so long we see benefits from it and eventually kill the underlying blockers.

Hey just a heads up that your resume link is broken! Looks like it was truncated somewhere.