zmmmmm

21 ms·

451.
by zmmmmm 1y ago
The real question for me is how much they target GPU support towards accelerating language and vision models in the Macbook Pro lineup. I don't know if Apple actually cares but they've got a huge opportunity to steal the spotlight
452.
by zmmmmm 1y ago
agreed on that ... I'm not too happy with how CloudFlare tried to flex their influence on AI training, so I certainly wouldn't want them in charge of gatekeeping the whole internet. But the truth is, we are already there. CloudFl
453.
by zmmmmm 1y ago
There's an opportunity for a service like CloudFlare here give people a simple toggle that manages geoblocks on legal liability factors. It's way too much for every organisation to individually track every country's laws day
454.
by zmmmmm 1y ago
> Apparently they told it to build a Slack clone and left it unattended for 30 hours, and it built a Slack clone using 11,000 lines of code it's going to be an issue I think, now that lots of these agents support computer use, we a
455.
by zmmmmm 1y ago
I only see "ethical" under the innovation / consortium part. Don't see how that applies to people producing LLMs outside of the consortium?
456.
by zmmmmm 1y ago
it sounds like a nothing burger? Pretty much the only thing tech companies have to do in terms of transparency is create a static web page with some self flattering fluff on it? I was expecting something more like a mandatory BOM style list
457.
by zmmmmm 1y ago
I don't like the artificial distinction b/w thinking and coding. I think they are intimately interwoven. Which is actually one thing I really like about the LLM because it takes away the pain of iterating on several different appr
458.
by zmmmmm 1y ago
I routinely ask the LLM to summarise the high level points as guidance and add them to the AGENTS.md / CONVENTIONS.md etc. It is limited due to context bloat but it's quite effective at getting it to persist important things that
459.
by zmmmmm 1y ago
a lot of "local" models are still very large to download and slow to run on regular hardware. I think it's great to have a way to evaluate them cheaply in the cloud before deciding to pull down the model to run locally. At so
460.
by zmmmmm 1y ago
I generally detest vibe coding but this is really screaming out for someone to just let Claude loose for a day or two.
461.
by zmmmmm 1y ago
> A good question is why an app that worked fine over a decade ago apparently still costs $800k per year to support This is what I would like to know. Granted, Google makes it hard to be an app developer these days with constant requirem
462.
by zmmmmm 1y ago
It would actually be fine to me if they left the old app to rot and spun up a new one as a subscription model. Eventually the old one would break and we'd shrug our shoulders and move on. However instead they took the existing app and
463.
by zmmmmm 1y ago
> why does the author here feel so entitled Because they paid for it.
464.
by zmmmmm 1y ago
have just been in the process of rage uninstalling this app due to the ads To be clear, it's not just that they added ads, but they are obnoxiously in the main active screen while things are playing. Made me also disrespect Automattic
465.
by zmmmmm 1y ago
there's a worrying lack of structural integrity building up in this hype bubble and this adds more fuel to the fire. You essentially have Nvidia propping up its own valuation here by being its own customer. If they sold a bunch of H100
466.
by zmmmmm 1y ago
I've never yet accepted an AI written answer when responding to my emails all though I try it routinely. Mostly it just doesn't capture my style. But even when it does, there's some kind of essential spark missing. I think a
467.
by zmmmmm 1y ago
I would assume they keep it secure as far as the training data goes, I don't have too many doubts about that. But aside from that part, yes, I can't see how you could make a different assumption.
468.
by zmmmmm 1y ago
I spent 10 mins trying to find a clear statement of whether Google uses information submitted to Gemini for training and I couldn't find one. It is hard not to come to the conclusion they actively try to obfuscate it because there are
469.
by zmmmmm 1y ago
I prefer small units of work. It really surprises me how fast people have leapt from the 2x speedup (mind blowing level of productivity increase) to full agentic coding without really questioning if it's a good idea. When I have let Cl
470.
by zmmmmm 1y ago
ah, they are holding it wrong. I am always so skeptical of this style of response. Because if it takes hundreds of hours to learn to use something, how can it really be the silver bullet everyone was claiming earlier? Surely they were all i
471.
by zmmmmm 1y ago
> I've been a senior dev for a while now, a lot of my job _is_ reviewing other people's pull requests I kind of hate that I'm saying this, but I'm sort of similar and one thing I really like is having zero guilt about
472.
by zmmmmm 1y ago
> e.g. the recording indicator can be subtle or blocked if they are like the previous ones they have hardware level detection and decativation of the camera if the indicator light is blocked
473.
by zmmmmm 1y ago
> knowing that a derivative of Meta's Orion AR prototype will release to manufacturing in the next few years You actually know that? how? Just the leaked road map or something more concrete?
474.
by zmmmmm 1y ago
The brand on display here is Ray Ban. That's why they spent billions to lock in the partnership.
475.
by zmmmmm 1y ago
the thing is, if it loses the connection, why on earth would the correct behaviour be to just keep repeating the last response? It should just straight up say, "Sorry I'm having trouble connecting". Even the best case scenari
476.
by zmmmmm 1y ago
That's about gen 1, gen 2 is vastly more successful
477.
by zmmmmm 1y ago
No display but perhaps you can support these guys and hope they get there: https://mentra.glass/
478.
by zmmmmm 1y ago
there are millions in the wild ... so maybe not super popular but it's above penetration of anything you would call a niche product
479.
by zmmmmm 1y ago
yep. whatever else you say, Meta's willingness to throw some tech out there is thrilling from a geek / tech perspective.
480.
by zmmmmm 1y ago
I do think we're in for a bit of a reality check on how human attention works. I have a HUD in my car that shows me directions, speed etc and when I'm looking at that the rest of the view out the windscreen is hardly even there to

More ›