sulam

5 ms·

31.
by sulam 3mo ago
I think the stored procedure equivalent would be a "on delete, cascade these tombstones" -- both safer and cleaner.
32.
by sulam 3mo ago
That’s misunderstanding why these models are behind. A large part of why they’re behind is they aren’t able to do the reinforcement learning post-training steps that takes a pre-trained model and turns it into a frontier model like GPT 5 or
33.
by sulam 3mo ago
To HN moderators: title needs to note that this talk is 13 years old.
34.
by sulam 3mo ago
The argument that really hits home for me, after 30+ years in this industry, is stored procedures. The “Stored Procedures are Evil” argument to me is an artifact of an industry that promotes treating engineers and infrastructure as entirely
35.
by sulam 3mo ago
I mean, ignoring the leakage issue, which requires a specific behavior from creators that may or may not play out the way described — isn’t this just a huge creator trust issue (noted on the last line of the blog post)? Can’t I just prompt
36.
by sulam 3mo ago
I thought I read somewhere that many of these little red dots are turning out to be nothing more than bog-standard brown dwarfs in our own galaxy that are confusing the signal. These days we have some pretty powerful agents who can read the
37.
by sulam 3mo ago
Those already exist.
38.
by sulam 3mo ago
How do you copy code from Office? Is the source code public?
39.
by sulam 3mo ago
Unfortunately as a resident of the SF Bay Area, calling my elected representative is next to useless. :/
40.
by sulam 4mo ago
Have you tried optimizing this prompt so that it’s shorter but gets the same results? I see these super verbose prompts all the time from people who learned prompt engineering in the ‘24-early ‘25 timeframe and they seem unnecessary to me (
41.
by sulam 4mo ago
I’ve had Opus randomly insert (correct) Russian words into responses. It’s like their training data includes some bilingual forums where idiomatic Russian speakers congregate.
42.
by sulam 4mo ago
Or you could just play Factorio.
43.
by sulam 4mo ago
The debugging was interesting. I'm just going to have to learn to live with this I feel like, but the very LLM-ish language in the blog post was kind of annoying.
44.
by sulam 4mo ago
Django has strong honey badger energy!
45.
by sulam 4mo ago
Personally I prefer the API pricing because I feel like I'm not going to get rug pulled on my work. When it comes to personal stuff, I use the shit out of my sub, but it's not making me money.
46.
by sulam 4mo ago
A Falcony Heavy probably generates 1 kiloton of CO2 per launch. Data centers on the planet are highly variable depending on their energy mix. It's true that a large a datacenter running on natural gas or coal power is significantly mor
47.
by sulam 4mo ago
The intelligence is an emergent property of their ability to predict how a statement will proceed, therefore it is inevitably a reiteration or transformation at best. Lots of intelligent things can be produced from that, but nothing truly n
48.
by sulam 4mo ago
Tl;dr — but it’s too heavy You’re making some decent points here, but you’re either forgetting or ignoring the major thing that people usually neglect to mention when they want to make a case for this (crazy) idea — weight. Unless SpaceX is
49.
by sulam 4mo ago
Sentence constructions like this definitely scream AI: "That's a reasonable bias for an exploratory tool. It's a ruinous one for a triage queue..." I will upgrade the "why it matters" to "and now AI output
50.
by sulam 5mo ago
I won’t nitpick too hard because the percentage doesn’t move much, but Netscape went public in ‘95 so it’s got to be before that. The minimum would be ‘91 (first website) but that’s too early. Super geeky but I will peg it at April 1 1995,
51.
by sulam 5mo ago
It’s slightly more nuanced than layoffs = capex. You’re right, they don’t. That said, they do create free cash flow, which the market uses as one important input into the value of a given stock. Moving FCF positively when capex spending is
52.
by sulam 5mo ago
Stadia was a minor footnote compared to Android, Pixel, and the other large organizations at Google. But there was plenty of hiring there during the pandemic, so your broader point is not wrong.
53.
by sulam 6mo ago
Yeah those idiot farmers with all their machinery and services are really missing out on your trenchant observations.
54.
by sulam 6mo ago
I think their assumption is that there will not be enough people with money to pay the prices, monopoly-generated or not.
55.
by sulam 6mo ago
I wouldn’t blame Google for Oracle being a lawnmower.
56.
by sulam 6mo ago
The fact that this is optional means it will still happen, simply because of the signaling doing it quarterly will provide.
57.
by sulam 6mo ago
So much hate for these, but they do one thing really really well, which is handle a full 14 hour flight with no charge and great noise cancellation. That the noise cancellation on the new model is even better will probably make them a buy f
58.
by sulam 7mo ago
Try to play a simple over the board style game with 5.4 with whatever notation you chose to use (or just descriptions, literally anything). Prediction: it will start out fine, but the mid game will be very hard to keep it on track, and the
59.
by sulam 7mo ago
I use the chess example because it’s especially instructive. It would NOT be trivial to train an LLM to play chess, next token prediction breaks down when you have so many positions to remember and you can’t adequately assign value to inter
60.
by sulam 7mo ago
Because there are some really fundamental things they cannot do with next token prediction. For instance, their memory is akin to someone who reads the phone book and memorizes the entire thing, but can't tell you what a phone number

More ›