johntash

5 ms·

31.
by johntash 4mo ago
fwiw when I opened the site, my fans spun up and firefox slowed a lot. No idea what it's doing, but it's definitely using a ton of cpu.
32.
by johntash 4mo ago
have you tried using proton//steam on linux lately? I find almost every game I want to play actually works decently well now. The ones that don't are multiplayer/online ones that require some sort of anti-cheat.
33.
by johntash 4mo ago
If you already use jira, maybe try putting the specs directly in the jira ticket? And/or in a confluence page? I'm not sold on SDD yet, but part of that is because we already have decently detailed tickets and designs.
34.
by johntash 4mo ago
Gotcha, I guess I should have asked if you were using 'claude -p' or something else to re-use the subscription. It looks pretty interesting though, I'll give it a shot
35.
by johntash 4mo ago
Don't forget to watch SGU too, and I think there were a few other movies too?
36.
by johntash 4mo ago
Any idea if claude code will continue working with paseo after the billing/usage changes in a couple weeks?
37.
by johntash 5mo ago
Did you set up solar nodes in random spots or only areas you own/control? I've wondered if it'd be legal to just chuck a couple into some random trees.
38.
by johntash 5mo ago
Sometimes I feel bad that I lost years of chat history from then too, but you made me realize that might be a good thing.
39.
by johntash 5mo ago
It might be too late, but have you thought about using something like ledger/beancount for the storage? Having a nice tui/plaid sync for plaintextaccounting would be nice. https://plaintextaccounting.org/
40.
by johntash 5mo ago
Why not actually use smtp+imap/pop3 if that is the intention? Also I'm going to complain a little bit about the name: SMTP is for sending (not reading) mail, so asmtp to me at first glance looked like 'async-smtp' which
41.
by johntash 5mo ago
My understanding is that zed wrote a wrapper around 'claude -p' for ACP support so I imagine there isn't too much of a memory savings there since it's still actually running claude code.
42.
by johntash 5mo ago
I really liked Subnautica, but haven't been following the drama of subnautica 2 since the old creators were fired. _Should_ I give my money to unknown worlds?
43.
by johntash 5mo ago
frontier models are still way better than local models from what I've seen. To get close to them with large context windows and decent performance, you need more than a reasonable machine imo. I'm hoping local llms start rapidly
44.
by johntash 5mo ago
I assume this will affect things like using claude code inside of zed or vscode too?
45.
by johntash 5mo ago
How are you expecting search to work without JS? I'm pretty sure you would need either some sort of backend OR JS for any sort of search to work. Or maybe another option would be to outsource the search part to "site:yourdomain&
46.
by johntash 5mo ago
I'm assuming they're using the enterprise plan where you have to pay per token and there's no included usage?
47.
by johntash 5mo ago
> You need all those speakers? One for each room? Not op, but one reason I have a speaker in _almost_ every room is for notifications. Simple stuff like "the garage door is still open" or weather alerts, etc. I rarely actuall
48.
by johntash 5mo ago
> what is it with C level executives and AI? My assumption is that they are extremely excited about AI, because they are also extremely excited about being able to reduce their workforce while expecting more output from smaller teams.
49.
by johntash 5mo ago
If you haven't used it before, give it a shot. Worst case you waste a few years of your life. Doom emacs and Spacemacs are both good starter kits to give you an idea of what you could do.
50.
by johntash 5mo ago
Matter more to who though? Probably not the people making most of the profits :)
51.
by johntash 6mo ago
I sorta get that reasoning, but is a 24 hour cooldown really going to stop scammers? They're already used to multi-day scams, so wouldn't they just say they'll call back in a day to finish the process?
52.
by johntash 6mo ago
I like that this is an irc->matrix bridge instead of a matrix->bridge. I honestly don't use either much anymore, but I'll probably give it a try. I didn't know irssi was still maintained in 2026. There hasn't bee
53.
by johntash 6mo ago
> apply my database migrations and populate the database with static datasets before I deploy my application You could a) have the app acquire a lock in the db and do its own migrations, or b) create a k8s job that runs the migration too
54.
by johntash 6mo ago
> https://depsguard.com if anyone is interested. I really appreciate that you didn't include a "curl | bash" command to paste for installing, but at the same time it's what I was expecting when I clicked.
55.
by johntash 6mo ago
How do you keep llms from writing _too much_? I've built a few similar tools and systems, and they're all way too easy for the llm to just keep documenting things to the point the whole system is a mess and becomes less useful t
56.
by johntash 6mo ago
Do you use local models for these, or are you okay with giving private details to anthropic/openai? (that's one of my biggest hurdles for really adopting any useful assistant type of agent)
57.
by johntash 6mo ago
Keep in mind this isn't just about personal data on work hardware. It also leads to things like "we noticed you didn't move your mouse or type anything for 45 minutes, what were you doing?" type of micromanagement.
58.
by johntash 6mo ago
Out of the listed options, Garage is actually really simple for something that runs on multiple servers. It was also the only easy option I found that can replicate over WAN and isn't super latency sensitive.
59.
by johntash 6mo ago
One thought I've had a few times is "well.. this is good enough, maybe a future model will make it better." so I won't completely disagree. But my counter argument is that the generated code can easily balloon in size a
60.
by johntash 6mo ago
Do you have any recommendations of something like konsist/harmonize that work for multiple languages by any chance? I've been looking for a solution to org-wide architectural linting that can has custom rules or something so I co

More ›