Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
ulrikrasmussen
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
1.
▲
by
ulrikrasmussen
4d ago
Almost exclusively Fable at xhigh. I sometimes run out of tokens, but most weeks it's not a problem. The limiting factor is by far how fast I can push understanding of what was made into my brain, and I find that I waste time using Opu
2.
▲
by
ulrikrasmussen
5d ago
The geometric nightmares! I had those as a kid whenever I was very tired or had a fever, and kept having them until my late teens. Never seen them described and didn't know it was a thing.
3.
▲
by
ulrikrasmussen
6d ago
That will require a second pass though, because you have to free all your strings again.
4.
▲
by
ulrikrasmussen
6d ago
Yes, you could allocate it on the stack. I think back then (still?) a filename could be at most 260 characters, each encoded with 16 bits, so about 52k of stack allocation.
5.
▲
by
ulrikrasmussen
6d ago
But the naive way of doing this also wouldn't really require two passes, right? It would just require more memory because you would first save all file names in an array (stopping at 100), then pick a random one in constant time.
6.
▲
by
ulrikrasmussen
10d ago
Yes, but there is no alternative other than giving up. Starting a new OS from scratch with zero apps is a much worse starting point compared to a platform where 99.9% of apps work (minus those relying on Play Integrity with strong integri
7.
▲
by
ulrikrasmussen
11d ago
I remember that early image models couldn't generate a cyclops no matter how you prompted it, it would at best put a third eye in the forehead.
8.
▲
by
ulrikrasmussen
11d ago
I think LLMs have some of the same risks and benefits of stimulant drugs. They can make you more productive if used effectively as a tool, but they can also delude you into thinking you are better than you are and create a dependence such t
9.
▲
by
ulrikrasmussen
12d ago
Well, that's at least a positive. I hope this means they are also forced to accept the weakest credential type, and that they can't require you to present the identity proof that the wallet also provides when verifying your age. B
10.
▲
by
ulrikrasmussen
12d ago
We also have a Danish wallet now, AltID, which implements an anonymized (assuming no collusion between issuer and eavesdropper or service provider) age verification protocol based on batches of single-use tokens which contain no personal in
11.
▲
by
ulrikrasmussen
15d ago
Ban targeted advertising
12.
▲
by
ulrikrasmussen
24d ago
What law gives a developer the right to be published on app stores? And why would it make a difference if the US administration sanctions the EU?
13.
▲
by
ulrikrasmussen
24d ago
No it isn't, this is also the security model that prevents those same agents from cloning your phone before giving it back to get access to all services you are logged in to, including any SSH keys or government identity apps you may h
14.
▲
by
ulrikrasmussen
24d ago
It's not a trivial task, but yes, they could and should. They could also just officially start supporting existing android alternatives such as GrapheneOS by requiring banks and government ID apps to be independent of Play Integrity, w
15.
▲
by
ulrikrasmussen
25d ago
Yes, but Google is making that a lot more difficult soon, at least on stock Android
16.
▲
by
ulrikrasmussen
25d ago
But this can also just be removed from the app stores and the developer certificates can be revoked. Due to the centralized control and distribution of software on smartphones, no smartphone based payment system is really sovereign - it onl
17.
▲
by
ulrikrasmussen
25d ago
I think that is hard to do securely - wiping the phone implies wiping the keys from secure storage, and those cannot be restored by design.
18.
▲
by
ulrikrasmussen
25d ago
We did a prototype to integrate an agent into our application and basically just gave it a tool to discover the OpenAPI spec and call endpoints. It worked surprisingly well! One caveat was that some responses were too big and would poison t
19.
▲
by
ulrikrasmussen
27d ago
I have already commented about this is probably the worst advice I have heard in a long time, and I would definitely not let this guy anywhere near a codebase I had to maintain. But this is also essentially blogspam. The author intentionall
20.
▲
by
ulrikrasmussen
27d ago
Then please help me understand, it is entirely possible that I misunderstood you, but you are not clarifying your point.
21.
▲
by
ulrikrasmussen
27d ago
Then it sounds like you can only have a true microservice architecture if your application doesn't really need transactional guarantees, consistency or fault tolerance.
22.
▲
by
ulrikrasmussen
27d ago
Is this satire? Changing your easy to maintain monolith architecture to a microservice architecture to enable parallel agent development, really? That's got to be some of the worst advice I'd heard in a long time. You will have co
23.
▲
by
ulrikrasmussen
28d ago
Yes, and this is the reason that many of us have not, and will not, ever buy an Apple product. They are the ultimate technodictators, and their products are only for people who like to trade their freedom for moderate convenience.
24.
▲
by
ulrikrasmussen
29d ago
Yes, theres a similar one for Microsoft Authenticator which obviously tries to con you into installing it by also advertising that it allows you to "sign in to your Microsoft accounts".
25.
▲
by
ulrikrasmussen
29d ago
I see the exact same thing on Play Store, and I think Google is outright complicit in fraud. The top result when you search for an app, even if by exact name, is always, literally 100% of the time, not the app you are looking for, but som
26.
▲
by
ulrikrasmussen
1mo ago
There is no single conforming "corporate world" that applies to all companies. If AI is treated religion in your company that sounds like a local problem, even if it seems like a trend. Besides, GP was not talking about corporate,
27.
▲
by
ulrikrasmussen
1mo ago
What is this defeatist attitude? I don't know where you work, but it is not my impression that AI has instantly turned all developers into mindless AI-pilled sheep. If you feel like a simple suggestion like that will cost you your care
28.
▲
by
ulrikrasmussen
1mo ago
This is a common problem, and I don't get why LLMs have not been tuned to stop this nonsense. It is writing comments as if the audience is you, the user in the session, while obviously code comments are meant for future readers.
29.
▲
by
ulrikrasmussen
1mo ago
The vast majority of people who lose weight by calorie restriction gain it back, so I wouldn't call it highly effective. Short term yes, but long term, definitely not.
30.
▲
by
ulrikrasmussen
1mo ago
If you generate PRs too big to review for others, then they are too big to review for yourself. This means you are delegating the task of understanding the code to an LLM, and the end result is inevitably that noone in the organization unde
More ›