Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
TZubiri
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
9 ms
·
151.
▲
by
TZubiri
2mo ago
What are the main workloads of Debian? It's worth noting that Debian has no code production, so the LLM debate applies differently than an open source project. I'd assume the main workload is testing, reviewing code for maliciousn
152.
▲
by
TZubiri
2mo ago
Why would you assume that we have to move from techX to techX+1? If I release http4 would you assume that you have to migrate to http4 at some point?
153.
▲
by
TZubiri
2mo ago
requests is great and all, but you can just use sockets and send raw http strings. Don't treat it like some impenetrable huge OS project. Unless OpenAI deprecates the HTTP endpoint, you can pretty much ignore the whole SDK packages.
154.
▲
by
TZubiri
2mo ago
May depend on whether their profits come from less than or more than 25% of 'children-harming'.
155.
▲
by
TZubiri
2mo ago
It's an official tailscale product no? Fwiw, I don't think vendor lock-in is a bad thing, but being open source doesn't mean there's no vendor-lock in. If a client builds their system on top of an open source technology,
156.
▲
by
TZubiri
2mo ago
> if port forwarding works for you > port forwarding is actually a half-decent idea You make it sound like both options are equally valid, and as if port-forwarding were some esoteric technology. It's not, one is good, the other
157.
▲
by
TZubiri
2mo ago
It's a pattern that occurs in commercial software yes, but Open Source has its own version of building mirror ecosystems. Collaborating to bleeding edge issues on robust infrastructure is very hard, so junior engineers that want to col
158.
▲
by
TZubiri
2mo ago
>"like netcat, but over Tailscale's data plane" Half of the Software offering nowadays seems to be selling vendor lock-in at no added value and then making a profit. Sure there's always some negligible added value, an
159.
▲
by
TZubiri
2mo ago
Probably not benchmarked on games. Doing so might sacrifice quality on other more important benchmarks, and when it's used for legal and medical purposes, it's the right choice.
160.
▲
by
TZubiri
2mo ago
> LLM's are bad at designing prompts. You are going down a maddening rabbit hole. Prompts are the things humans write, you are building gas town but unironically
161.
▲
by
TZubiri
2mo ago
Set temperature to 0
162.
▲
by
TZubiri
2mo ago
Suggesting business names for businesses, I mean they are great, but they already exist, multiple times even.
163.
▲
by
TZubiri
2mo ago
I can see how this would be a vulnerability in the context of a security researcher that wants to exaggerate their findings in order to get paid a bounty.
164.
▲
by
TZubiri
2mo ago
Funny to think of W95 like some kind of luxury that a shopkeeper has to keep teenagers out, as if you were trying to test drive a Mercedes, or trying out guitars like that scene from Wayne's World (1992)
165.
▲
by
TZubiri
2mo ago
OpenAI implemented measures to reduce reverse engineering, following the Anthropic lead. They disabled the temperature and seed parameters. There's still logprobs, so they aren't as closed up as Anthropic yet. I might write about
166.
▲
by
TZubiri
2mo ago
Those are all law-abiding organizations, which AA is not. INB4: "Here is one time one of those organizations broke the law". Don't go there, absolute lowest level of conversation.
167.
▲
by
TZubiri
2mo ago
Nowadays, if you really want to sound modern and avant garde, try pitching a non-AI startup.
168.
▲
by
TZubiri
2mo ago
I left this comment 2 weeks ago on a youtube short on the subject: "I have an opinion on the legal matter. But I think it's worth noting that destroying the data was a categoric strategic blunder by the defendant. If you don'
169.
▲
by
TZubiri
2mo ago
Of course speed is good, but if you don't add an artificial latency (or better, use the extra time for some QA, guardrails, etc...), the model will come off as creepy at best, and the conversation will feel awkward for the user. Humans
170.
▲
by
TZubiri
2mo ago
The Claude Security program looks similar to OpenAI Cyber Trusted Access. The approval process seems quite intimidating, has anyone managed to implement these compliance standards?
171.
▲
by
TZubiri
2mo ago
I thought it was a simple 8 pointing to a perpendicular direction. Apparently the model is a bit more complex. I don't think complexity in the predictive model is (necessarily) attributable to a more complex bee language. One possible
172.
▲
by
TZubiri
2mo ago
You are just shifting where the eyes are, the amount of eyeballs (developer time) is the same. Unless the proposal is to increase the amount of (other) developers time. The proposal ends up amounting to either shifting stuff around, or aski
173.
▲
by
TZubiri
2mo ago
Yes, in agentic coding, "code" is an artifact, in the sense of "release artifacts" the word used by VCS systems like GitHub. And the agentic step is not (or less) deterministic, so the burden on VCS is augmented by the n
174.
▲
by
TZubiri
2mo ago
If those transitive dependencies are baked into the language runtime, how is anything materially different? You just shift the vector around.
175.
▲
by
TZubiri
2mo ago
> https://docs.rs/arrayref/latest/arrayref/ >This package contains just four macros, which enable the creation of array references to portions of arrays or slices (or things that can be sliced). The pack
176.
▲
by
TZubiri
2mo ago
I think that even for core modules like an http server or client, but that's not the case and we don't need to get into that debate. let's look at the actual package from OP > https://docs.rs/arrayref/l
177.
▲
by
TZubiri
2mo ago
I agree that building your own is hard and something that can be done by exceptional engineers, but software is famously a winner-take-all industry, even if the optimal strategy for lower percentile programmers and median programmers to imp
178.
▲
by
TZubiri
2mo ago
>I think we should be taking a more “batteries included” approach https://www.youtube.com/watch?v=GZOuz-SG7-g Funny how you skipped "I should build my own batteries" and when straight to "increasing and ce
179.
▲
by
TZubiri
2mo ago
So customers use OpenRouter to avoid lock-in to specific model providers, and in doing so they lock-in into openrouter's aggregation API? Lock-in to what exactly? A string(string) function without any further restrictions? Note that ag
180.
▲
by
TZubiri
2mo ago
I did it once when I tried vibecoding, to a friend, nothing professional, and I watched myself stop at the explanation of how I did something. I think it comes from a place of desire, self delusion and shame (of not being able to do it alon
More ›