ComputerGuru

6 ms·

151.
by ComputerGuru 5mo ago
Seems to be llm written article and the tooling around the model is undeniably influenced by knowledge of the tests. In all cases, GPT 3.5 isn’t a good benchmark for most serious uses and was considered to be pretty stupid, though I underst
152.
by ComputerGuru 5mo ago
SecureStore is an open spec/protocol for managing secrets in a secure and portable manner, while it defines the decryption key formats (currently: key-based, password-based, or a mix of both interchangeably) it doesn't get into th
153.
by ComputerGuru 5mo ago
We actually just ported SecureStore to go, it’s sort of like this but with cross platform clis and intended to also allow sharing secrets across services and languages, in a secure and embedded fashion! It’s available in rust, php, .net, JS
154.
by ComputerGuru 5mo ago
For comparison, my code signing cert via Azure (no Microsoft store account required, can be used to self-publish binaries/installers the old fashion way) is $10/month, or about a third of the price Sectigo is charging you. I figur
155.
by ComputerGuru 5mo ago
Thanks for sharing your experience. I have been code signing releases for over a decade as an indie publisher myself, until I found myself effectively iced out by the HSM requirement, the increased cost, and the shortened cert lifetimes, wh
156.
by ComputerGuru 5mo ago
So does this confirm the end of llama?
157.
by ComputerGuru 5mo ago
It’s become neigh impossible to get your own code signing cert these days. The 2025 update from the CA forum required code signing certs to be short lived (no more three or five year certs) and stored exclusively on an HSM. As a result, mos
158.
by ComputerGuru 5mo ago
That’s not how any of this works. There are separate teams within (each division of) Microsoft that could easily pull the plug on your account (or if not the entire account then your account’s access to the specific service or family of ser
159.
by ComputerGuru 5mo ago
Thanks for the clarification!
160.
by ComputerGuru 5mo ago
I have mixed feelings about this. It's absolutely phenomenal that such a treasure trove was unlocked thanks to AI, but presenting the AI results are "definitive" (even with an "edit" or "report" feature th
161.
by ComputerGuru 5mo ago
There's actually no guarantee that if the "experiment" were allowed to continue that the results would have been as great. If the biomass accumulated faster than it could be broken down, we might not have seen the same result
162.
by ComputerGuru 5mo ago
Nothing on @bushido (props for humbly admitting what happened) but AI has completely destroyed people’s ability to appreciate the effort and domain knowledge that goes into the design of just about any properly successful undertaking.
163.
Neovim TreeSitter plugin has been archived on GitHub(github.com)
21 points by ComputerGuru 6mo ago|1 comments
164.
by ComputerGuru 6mo ago
I’m not sure what this means for the future of the project; I can’t find any information about its deprecation anywhere.
165.
by ComputerGuru 6mo ago
You’re both saying the same thing.
166.
by ComputerGuru 6mo ago
A very different problem then the one I proposed an answer to, no?
167.
by ComputerGuru 6mo ago
Every other month I get an email from a legacy pre-GH bug tracker that's either a "me too" or "bug fixed in latest release" a decade after I filed these one-offs you would be so quick to throw away. Bugs with no a
168.
by ComputerGuru 6mo ago
You can carry exactly (or roughly) as much energy in the form of antimatter as you would energy in the form of fuel.
169.
by ComputerGuru 6mo ago
I think "written by genAI" should be a bigger turnoff than "written in Rust".
170.
by ComputerGuru 6mo ago
Plenty of (cattle or pet) tooling essentially devolves to SSH under those layers of abstraction.
171.
by ComputerGuru 6mo ago
SRV is essentially a simple layer of abstraction that provides (via one approach) the required end result (reachability + UX) that is easy to add to any $PROTO client without. Supporting ESNI would complicate the actual lib/protocol, i
172.
by ComputerGuru 6mo ago
So fwiw, browsing history shouldn’t be anywhere near that big making it unlikely there what it was. It compresses well, if they were to do it I’m sure they’d do it at regular intervals instead of a years’ worth at a time, etc. And, of cours
173.
by ComputerGuru 6mo ago
> Mimalloc made the claim that they were the fastest/best when they released and that didn't hold up to real world testing That’s… ahistorical, at least so far as I remember. It wasn’t marketed as either of those; it was market
174.
by ComputerGuru 6mo ago
That’s a considerable regression for mimalloc between 2.1 and 2.2 – did you track it down or report it upstream? Edit: I see mimalloc v3 is out – I missed that! That probably moots this discussion altogether.
175.
by ComputerGuru 6mo ago
Isn't that no longer necessary? Doesn't linuxulator + podman suffice?
176.
by ComputerGuru 6mo ago
This obligates me to share this absolute gem of date/time history folklore: https://neosmart.net/forums/threads/an-extended-history-of-t...
177.
by ComputerGuru 6mo ago
Open source is the only way to go. Out-of-the-box, most Linux distributions automatically report the battery status on my (admittedly ancient) Performance MX and I get a desktop notification when the battery is running low so I can run and
178.
by ComputerGuru 6mo ago
Possibly of interest to readers (but not directly competing) is securestore, our open-source, open-protocol, cross-language framework (cli tool + libraries) for securely storing secrets and programmatically retrieving them in production: h
179.
by ComputerGuru 6mo ago
Slightly related: I have a tool that writes random (incompressible) data to a disk and lets you verify it back without storing a copy (by using a csprng seed), initially developed for benchmarking SSDs that used to cheat to get better perfo
180.
by ComputerGuru 7mo ago
Reviews of the tool on twitter indicate that it completely nerfs the models in the process. It won't refuse, but it generates absolutely stupid responses instead.

More ›