Home

Updated my actual-budget container today, and met with authentication failure - I am locked out of my budgets.

I turned to the github repo for troubleshoot, and.. Noticed that their commits are pretty much dominated by AI now.

At least they seem to be being up-front about their AI usage by tagging commits with the tag, but it still feels icky to me.

(As a ‘bonus’, they also only have their support channel in discord - but I guess that’s the norm for many open-source projects.)

Is this fact of life now? Are most projects being written and managed by LLM nowadays? Not sure I can trust the projects nowadays.

Is there alternative budget app which does not involve LLM?

I would assume any project that has more than a handful of contributors to have AI-assisted code in it.

I’m probably living in my little start-up bubble, so my view is probably skewed. The majority of commits I see have not had any code written by humans. Planned, specified and reviewed by senior developers with fancy degrees and a decade of experience in average, though.

Things move fast, but I’m sure a lot of older and bigger organisations are taking it slower because of the legal unknown.

Aside from fundamentalists the usage of LLMs and coding agents will increase. It’s a tool in the toolbox now and many devs do or will play around with it. Some will have to learn to not overdo it; but that’s nothing new and a lot of fancy technologies or frameworks along the way caused some disruptions because people jumped on the hype train without applying some caution or critical thinking; but that evens out after a while.

Might be we see a big drop in usage when costs increase, but it’s also very very possible that the many technological advances we currently make (hardware to run models becoming more streamlined and the models themselves being tuned more and more) will mean, that we indeed reach a point where this can be done comparatively cheap and maybe even local (to some degree) without having to take out a loan.

I wouldn’t say “managed by LLM” though, just because you spot (partially) agent written commits. It’s hard to judge from the outside how much knowledge the maintainer puts into it. There a big band between vibe coding and fully manual coding. And if we are honest, even “fully manual” is a flexible term (does code completion count? does looking at stack overflow count? does looking at other implementations count? using a search engine?).

The world is changing, for better or worse. But cut devs some slack and let them get used to the tools. (And to re-iterate that: bad quality and bugs were a thing before agents as well. It just took longer.)

It’s not a tool, it’s a chaos generator.

Don’t let people who build houses do it with cardboard boxes instead of bricks, even if it looks like a house in the end.

Someone posted this in another thread and I was surprised to see how many well-known projects were listed among them

https://codeberg.org/small-hack/open-slopware

You might not like it, but the whole software development is starting to include AI. So you can expect that every project out there will have some kind of LLM written code in it. And maybe not now, but give it a few months or years and it will be in every major project. And even if you can’t see it, there will always be the PR going in from someone who used LLMs.

Found the fundamentalist

It lists no alternative to actual(-budget) :( I don’t have time to create my own alternative, maybe I should go back to spreadsheets?

So LLMs are going to achieve what Microsoft has been unable to, destroy open source and upend the world of coding. Nice. We really are living in the dumbest timeline. Can’t wait for Nintendo’s lawyers to decide they found a fragment of Nintendo code in the output of an LLM and start the lawfare to destroy the pesky breeding ground of emulator writers.

Said it in another thread, I have yet to meet a strong advocate for LLMs that isn’t a cunt.

There’s no reason for thinking that LLMs will destroy open source. People complain about slopware, but might also give us a lot of new cool projects.

Said it in another thread, I have yet to meet a strong advocate for LLMs that isn’t a cunt.

I know someone who is very convinced of LLMs, and while I have strong disagreements with her in that topic, she definitely isn’t a cunt.

Llm is just a tool and it usage will only increase over time.

There is nothing intrinsically bad in that, as with any tool, it’s not bad per se, but how we use it.

So, push for ethic and proper usage of AI, rather.

Projects will use AI more and more, nothing bad in that. Provided it’s used properly, vetted, tested, verified and such.

Just like humans. Bullshit code and bad developers existed before agents helped make them.

You haven’t been keeping up to date on the damage they are doing to existing projects have you? Ask the team behind curl how useful the slop submissions are. They are reaching breaking point just keeping up the flood of crap. It’s not that LLMs will out compete, they will overwhelm and infect.

There is a significant difference between AI assisted and AI produced from what I’ve seen and experienced so far.

Assisted takes generated code and uses it to inform the code actually written, letting it fulfill a boiler plate function or the place of a junior coder at worst.

Then there are those project committing the AI produced code largely unreviewed and unchanged.

Former is mostly fine but needs an experienced coder who trained writing code unassisted (where are new coders of that caliber going to come from now?), the latter is a morasse of slop.

For one, what’s her name? Perhaps I just forgot meeting her, otherwise my point still stands.

For two, I’m Australian, it would be extremely insensitive for you to make negative statements about my dialect.

I will admit I am being inflammatory with that language, but honestly the only person I have met who I would call an honestly good person who was super positive about LLMs, giant math nerd who talked over my head for 10 minutes about the implementation of convolutional networks. He might have just been super excited I was taking an interest and doing my best to ask questions that didn’t make me sound to dumb.

Yes, experience matters a lot. I think the comparison of an coding agent being like a trainee is somewhat appropriate. Leave them to their own devices, and you likely don’t get something you should be shipping to production. But guide them appropriately, and they are helpful. The difference obviously is, that a trainee learns, an agent not so much. At least not on an abstract level. Of course the more code you have, the more patterns they can then copy. But if your baseline rots, so will the code the agent derives from that baseline.

I have been following the curl issue for quite a while. I think that it’s different from what I wanted to tell you here. Curl is getting too many PRs that are AI generated and the human maintainers can’t deal with them all.

What we are here seeing is people using coding agents and LLMs to help with their own work. So it’s a different kind of thing. You can also imagine a situation where projects are getting hammered by AI bot traffic, but are getting also helped by using LLMs with their project itself.

The ActualBudget project has a support community on Discord: https://discord.gg/8JfAXSgfRf. Voice your grievances there.