overgard

37 ms·

781.
by overgard 1y ago
I remember in the early 2000's when compositing window managers first came about people went wild with the effects in the most tacky way possible (me too -- it was fun at the time!) Everything transparent, rotating cubes for different
782.
by overgard 1y ago
The worst part of AI is the way it's aggressively pushed. Sometimes I have to turn off AI completions in the IDE just because it becomes extremely aggressive in showing me very wrong snippets of code in an incredibly distracting way. I
783.
by overgard 1y ago
If you want to get stuff shipped though, you need a good engineering culture. If you want to move fast you need experienced people. This feels like the argument against automated tests. "It'll slow us down!" So will debugging
784.
by overgard 1y ago
The article is talking about timescales around 2 years out though. If you're hitting that point and you haven't fixed your engineering culture..
785.
by overgard 1y ago
I don't know about his conclusion of "spend 2 weeks on architecture". If these people don't realize they need to index a database table, 2 weeks of talking about what they're going to do isn't going to help. Th
786.
by overgard 1y ago
I definitely feel the "bad at asking questions" part, a lot of times I'll walk away for a second while it's working, and then I come back and it's gone down some intricate path I really didn't want and if it ha
787.
by overgard 1y ago
I've heard the colloquial phrase "type 3 diabetes" appear a lot in talk about Alzheimers, so I'd be shocked if there isn't a lot of research going into that. I just read a book about how metabolism is related to a l
788.
by overgard 1y ago
No way dude, this does a disservice to the insanity that is C++'s syntax. Wake me up when you have 6 different initialization syntaxes or fun things like 4[array]
789.
by overgard 1y ago
I don't think types are expensive for MVP code unless they're highly complicated (but why would you do that?) Primitives and interfaces are super easy to type and worth the extra couple seconds.
790.
by overgard 1y ago
Weird, seems like a lot of ugly publicity to create more work for themselves. I wonder why Shopify chose to do this.
791.
by overgard 1y ago
I'm not sure I'd say that React's DX is actually that great. I've watched C++ developers come over from using QT and the learning curve is massive. Plus almost all the bugs come from dealing with React's renders. I
792.
by overgard 1y ago
AI right now seems more like a religious movement than a business one. It doesn't matter how much it costs (to the true believers), its about getting to AGI first.
793.
by overgard 1y ago
Assuming users accept those ads. Like, would they make it clear with a "sponsored section", or would they just try to worm it into the output? I could see a lot of potential ways that users reject the ad service, especially if it&
794.
by overgard 1y ago
If they're profitable, why on earth are they seeking crazy amounts of investment month after month? It seems like they'll raise 10 billion one month, and then immediately turn around and raise another 10 billion a month or two aft
795.
by overgard 1y ago
In terms of sources, I would trust Zitron a lot more than Altman or Amodei. To be charitable, those CEOs are known for their hyperbole and for saying whatever is convenient in the moment, but they certainly aren't that careful about be
796.
by overgard 1y ago
Because to make GPT-5 or Claude better than previous models, you need to do more reasoning which burns a lot more tokens. So, your per-token costs may drop, but you may also need a lot more tokens.
797.
by overgard 1y ago
I thought the thing that made DeepSeek interesting (besides competition from China) was that its inference costs were something like 1/10th. So unless that gap has been bridged (has it?) I don't think a calculation based on DeepSe
798.
by overgard 1y ago
From a privacy and security standpoint, hell no!
799.
by overgard 1y ago
I'd love for this kind of thing without the incredibly obnoxious commentary. I felt like I was reading propaganda rather than a how-to.
800.
by overgard 1y ago
I don’t know much about audit logs, but the more concerning thing to me is it sounds like it’s up to the program reading the file to register an access? Shouldn’t that be something at the file system level? I’m a bit baffled why this is a c
801.
by overgard 1y ago
Nitpick, but having side projects you don’t plan to scale isn’t anything unique to the AI era
802.
by overgard 1y ago
The studies I've seen for AI actually improving productivity are a lot more modest than what the hype would have you believe. For example: https://www.youtube.com/watch?v=tbDDYKRFjhk Skepticism isn't the same thin
803.
by overgard 1y ago
I'm not a fan of the argument that LLMs have gotten X times better in the past few years, so thusly they will continue to get X times better in the next few years. From what I can see, all the growth has mostly come from optimizing a f
804.
by overgard 1y ago
I don't think you can apply the incompleteness theorem like that, LLMs aren't constrained to formal systems
805.
by overgard 1y ago
One thing that seems to always get lost in the AI hype cycle is actually screening articles by the source. Maybe it's just because in this age of social media we don't have much in the way of journalism. I have zero interest in wh
806.
by overgard 1y ago
My issue with LLMs isnt really the tech itself. They're occassionally useful, although HOW useful seems to depend on a persons skillset. My main beef with the AI hype is that it's allowing a lot of idiots to significantly devalue
807.
by overgard 1y ago
Meta question, but why do these AI apps always have pink-to-blue gradients on buttons? I see it all over the place!
808.
by overgard 1y ago
Well, I admire the nerd logic, but it seems like it would just unnecessarily cripple sales if people searching for it under the common name can't find it.
809.
by overgard 1y ago
I think it's an unlikely future. What I think is more likely is people will realize that every line of code written is, to an extent, a liability, and generating massive amounts of sloppy insecure poorly performing code is a massive li
810.
by overgard 1y ago
Well, I was hedging a bit because I try to not overstate the case, but I'm just as happy to say: LLM's can't reason. Because it's not what they're built to do. They predict what text is likely to appear next. But ev

More ›