traverseda

4,101 karmajoined vor 11 Jahren

traverse. da not this line @ gmail .com

comments

vorgesterndiscuss

The token is stored on a LUKS encrypted drive on kwallet. You'd have just as much luck getting my password from my firefox installs sqlite DB. This is also how email clients in general work.

Also this isn't really an agent. At least not a long lived one. Each email or transaction gets it's own session, the llm can make a few tool calls but must emit json as the final result. Very very short context lengths, very predictable results.

The bot does not have access to my passwords, there are pre-defined scripts that fetch the data that the run an LLM call for each transaction discovered.

vorgesterndiscuss

Credit union atlantic. The possibility of you being able to profitably support small regional credit unions is pretty much 0. Maybe with a general browser use AI and an SMS portal. Also you'd need to bypass anti-bot protections, maybe solve captchas.

In my opinion that is the thing holding back almost all of these accounting products that make consumers lives better. I've solved it because all my cards and bills happen to support email, so I can use that as my source of truth for all my credit cards.

vorgesterndiscuss

This doesn't surprise me at all. You can really constrain this problem, give very narrow context, and get pretty reliable and reproducible results.

I've gotten very good results with some vibe-coded deepseek book keeping. https://github.com/traverseda/beansync

Parses emails or other sources, extracts numbers, correlates different transactions, web search, asks questions, stores notes (regex based, very simple).

The hard part is getting good data, I'm sure that lexus nexus or whoever can get API access to my bank account and all my credit cards, but I can't. Email turned out to be the best way for most of my providers. Managed to avoid 2factor auth so far, but it will suck when I need it.

vor 5 Tagendiscuss

There is a big difference between capturing value and creating value. 20 developers can build and maintain a facebook or reddit clone, so why aren't they kicking the asses of the big companies with the much higher costs?

vor 6 Tagendiscuss

Well the goal is to stop killing games, to make it possible for people to run their own games indefinitely. That's a lot harder than baseline consumer protections though.

vor 6 Tagendiscuss

With a video game it's not clear that you're purchasing a revocable license. That's why it's called "buying" a video game. If online stores were clear that you were actually leasing a game, I don't think this would be a problem.

Publishers and storefronts need to be clear that what you're doing isn't buying, or start selling tickets (season passes) that have a clear end date, or use some other mechanism that isn't "buying".

The fundamental problem is that it's unclear what you're buying, and the contract can change at any time. Are you buying an item, a ticket, leasing, a subscription like an MMO, etc. These are all different things, and it misleads consumers when they're conflated.

The terms are also very one-sided, and your "purchase" can be ended by one party at will with very limited notice. Even basic consumer protection like requiring six months notice before ending your software lease would help.

vor 10 Tagendiscuss

So the play here is obvious, use the teleoperation as training data for a more general purpose AI controller. You need that data to make a model in the first place.

What doesn't make sense to me is the cost. Yes, $8000 is probably low for this robot but it's a reasonable price range for something like this. The AI credits though? I know vision LLMs are not cheap, they're not going to run something like Llama3.2vision on every frame. Very curious about the embodied AI architecture that this is going to use, and how it can get cheap enough that it's not going to use $500/month in electricity every month.

vor 11 Tagendiscuss

From a few days ago, uses ultrasound.

https://news.ycombinator.com/item?id=48685558

vor 15 Tagendiscuss

Regen braking,

vor 17 Tagendiscuss

If memory serves the cachyos-nix flakes sets a cache for pre-built kernels, but that cache isn't available until you rebuild. So if you want to use the cache you need to do two steps, add cachyos to your inputs, rebuild, switch the kernel, rebuild.

Then reboot to use the new kernel.

vor 20 Tagendiscuss

Yes, it requires a daemon running inside the VM and can be finicky though.

vor 25 Tagendiscuss

Last I checked it was about $3000 per life saved by the most efficient methods. Looking at effective altruism communities. Mostly malaria nets.

vor 29 Tagendiscuss

If you are using a real computer or can sideload apps.

vor 30 Tagendiscuss

It's a WOPI app. Designed for embedding microsoft 365 office app in other web apps, but other people implement it. There's callabria, and this.

You'd use it to connect nextcloud to one of these providers. Probably some other enterprise apps I don't know about can use it to edit documents.

letzten Monatdiscuss

Also problems Ukrainian defense needs to solve, and that the Canadian military is trying to solve. This is everyone's problem. It's also biased towards defense use.

letzten Monatdiscuss

>that has to deal with the other constraints of the web platform.

Well there's your problem right there

vor 2 Monatendiscuss

>You can be protected by safe harbour provisions, or you can editorialise your content. *I don't think* you should have both.

Personal opinion, not legal opinion.

vor 2 Monatendiscuss

I'm also not a lawyer, I was making that as a more vague moral distinction on the topic of free speech and accountability.

For practical reasons I think those algorithms are absolutely necessary. We need spam filters. A good line to draw would be "bring your own algorithm". A technical challenge to be sure, bit breaking up social media backend providers and content filtering seems like one of the only safe ways to allow these massive platforms to exist.

The algorithm can be just "Dan filters out spam".

vor 2 Monatendiscuss

You can be protected by safe harbour provisions, or you can editorialise your content. I don't think you should have both.

Free speech does not cover scams and fraud, something that happens on their platform. Society doesn't take any action against them for publishing illegal content, scams, libel, fraud, because they aren't a newspaper. They're more like a newspaper printing house.

In my opinion they should probably be losing those protections and should suffer legal consequences for the content their users post. The moderation has reached a point where they ate defacto editorialising content.

An alternative to that could be opting in to some kind of third party moderation arbitration process.

vor 2 Monatendiscuss

You can run your own github actions compatible-ish server. https://github.com/nektos/act

Personally my favorite is probably drone-ci.

I'd suggest not buying in too hard on any one of these CI systems and just writing shell scripts. Shell scripts are portable, and you can use whatever to trigger them.