LawrenceKerr

Depends what you mean with "agents"... LLMs in coding harnesses, or fully autonomous agents?

If you mean the former, I think you'll find probably 90% of devs are "crazy", running Claude Code or Codex on their bare metal using the auto-classifier.

I used to always run claude with `--dangerously-skip-permissions` in a dev container, but I had to rebuild my container 10 times a day to update the firewall so it can access documentation.

Now I find the included auto-classifier is very reliable. If you keep tasks contained instead of open-ended long-horizon work, it doesn't seem much goes wrong with frontier models. The worst unintended action claude did for me in the past months is that it once discarded an unstaged file which wasn't committed yet, without asking.

But I realize it's still playing with fire, no matter how good the auto-classifier is.

Yes, I think a lot of us feel this nostalgic yearning for a revival of more science&techno-optimistic times.

We are tired of the constant cynicism, negativity, and small-minded thinking. It's too easy to be a cynic. To put everyone & everything down, and to feel intellectually superior. You see it all over on the Internet and in the real world too, it's a true mind virus. It'll never die, but one can only hope more people just get tired of it.

I will teach my kids that a cynic is a man who knows the price of everything, and the value of nothing.

> Ideally for who?

I'm just speaking for myself.

> What you're suggesting is to turn this "tool for a human" into a slop producer, so it's able to produce as much content as fast as possible, with as little human input as possible.

I'm starting to really hate the word "slop". It's such a lazy way of discussing AI content generation. I think we can all agree the "slop" produced by AI right now is (if prompted well, curated & edited), of a much higher quality than the average designer was able to produce in reasonable time before AI.

The whole point of technology, since the dawn of mankind, is to reduce the amount of human inputs, so we can focus more on what truly matters. In that sense, AI is fantastic. That doesn't mean you shouldn't just let it rip and turn off all of your own thinking and effort in editing the "slop" it creates. But I'm just getting really tired having to point that out in literally every conversation about AI. Can we just move on already, please.

> Couldn't you fork this and implement this "slop factory feature" yourself instead of asking this tool to be re-built from the ground up to cater to this?

Of course I could? What's your point? "Shut up and vibe code it yourself"?

Great but imho Canva alternatives that stand a chance should be made API and MCP-first, not just an afterthought.

That's the thing I miss with Canva right now. Yes there are API & MCP options (4 different ways even I think, it's pretty messy), but it's just not sufficient. Ideally you can fully vibe code a poster or a video from the context of your repo, which has all the context and assets, then edit the result online in their point-and-click editor.

Pretty standard at this point, especially for non-critical open source projects like this...

Perhaps it was in poor taste. Nazi jokes are no laughing matter.

France and Spain are most definitely not socialist under the classical definition of socialism.

Socialism is an economic and political system in which the means of production, such as businesses, factories, land, and capital, are owned or controlled by the state, rather than primarily by private individuals and companies.

But it's true that there's been an ongoing political effort to reframe socialism from its original definition to simply meaning a capitalist welfare state with worker protections

The problem is most people (who may say they want censorship-free social media) do not really want it to be censorship-free. They want it to be censorship-free for the opinions and views they hold, not for those they oppose.

I think X arguably comes closest, and at least it costs bots money to post with a checkmark (unlike Reddit, which is a bot free-for-all - since its very origins actually). Plus, the feed & community notes algorithms are actually open source (whether or not that's also the deployed version, I can only speculate).

Does this require a Meta / WhatsApp business account?

If not, is this a solution using WhatsApp Web in the backend (like Baileys does)? If this is the case, how does your solution avoid getting blocked or banned? I used Baileys once in an MVP with only 7 users, and my agent got blocked really quickly after just a few days of using it. So I guess the official path is the only way to go?

Doesn't seem like the best of weather now in Iceland, hopefully it clears up. Otherwise it would suck for the people who have planned the trip months if not years in advance only to get a clouded sky, but that's always the risk with astrotravel.

Yes, it's a bit of an odd h1, website also looks very Claude-generated, but regardless, probably most people understand the intent.

Many such open source projects already (which is fantastic). I lose track of them.

Today it happened I needed a simple way to embed & query 1TB+ of documents, and I was looking at open source options. Can anyone tell me what their go-to solution is now? Could this be the one?

And what are the key differences vs. other open source RAG tools like kotaemon?

It's incredible when it finally clicks.

Arguably the best thing you can do is simply keep a dream journal (aside from deliberately waking up in the middle of the night and falling asleep with a clear intention - but that is more invasive in one's schedule and can also trigger nightmares).