Do you work in tech?

One canwork in tech and not use LLMs. Your comment sounds like it should be the norm, or already is.

Also, the discussion is about using LLMs in open source projects, which put a lot more focus on understandable and maintainable code than the software industry typically happens to do.

Even more gain! Do you know what prozac does cost? Its family will need to pay the pharmacy bills if it doesn’t want their loved one ending up homeless.

Indoctrinate the spider that it will catch many more bugs by using AI tools, and then let it pay.

Here the pull request and discussion:

https://github.com/void-linux/void-packages/pull/62351#issuecomment-5641804457

Changes not tested.

Generally, what we have in the EU is strict liability for commercial products, but not for things like gifts or FOSS software developed in a non-commercial manner. Except when damages are caused on purpose or because of gross negligence.

Some examples (German law, which is quite strict about personal liability):

• You cause an accident -> you are liable.

• Your small kid causes an accident -> you are liable.

• Your unleashed dog runs over the street, cause a truck to steer into a shop -> you are liable.

• Your car does not work because of improper maintenance, and causes an accident -> you are liable.

• Your dishwasher floods the flat below you with water while you are away -> you are liable.

• Your car’s assisted braking system goes out of control because of defective software delivered by the manufacturer -> the manufacturer is liable.

• You gift somebody a car, he has an accident -> no liability for you.

• You knowingly gift somebody a car with defective brakes, without telling him, he has an accident -> you are liable.

• Your physician uses some X-Ray device software which has a catastrophic bug and causes that you have radioactive burns all over your body -> the manufacturer of the software is liable.

(and here is where EU law probably differs form US law: In the US, software vendors can and do disclaim all liability. THIS IS NOT POSSIBLE IN THE EU, and thus such a contractual clause is probably invalid in its entirety, while leaving the contract valid. (And that is why you should NOT write your own licenses.))

• That X-Ray software uses a non-commercial contribution for solving matrices in Numerical Python which a volunteer wrote in 2015 -> the FOSS contributor is not liable, the responsibility is for the manufacturer to test for correctness of the free stuff.

It is generally not a good idea to write an own license. Because it is complex legal stuff.

Also, when using licenses, one should check whether the interactions with other licenses are as intended. Some licenses are compatible in a two-way manner, some only in one way, some not at all. The GNU project has a page for that.

To clarify any confusion: I posted this, but this is not my blog article; it is from a different person.

Well, I already wrote that copyleft licenses get hate from corporations…

So, your argument is merely that the point he blogs about does not matter? Or do you somehow not want people to discuss it? Why?

Be ready to supply your fingerprints using Microsoft Authenticator app and your free source access subscription for only 49 USD per month! /s

Everybody can build whatever they want.

I do not see how that adresses his core argument - that permissive licenses enabled things we do not want and which subvert individual and collective freedoms.

More to the point, the GPL and its sisters are designed to grant rights to people.

While the “everybody” in your sentence above includes corporations like Microsoft, Facebook and Anthropic which do very unfree things with open source tools.

This is not the same intent.

Personally, I also think that much of the hate which one would find e.g. on reddit or the orange site against copyleft licenses in general, and the GPL specifically was probably astroturfing from corporations for which it was just an obstacle to appropiate more unpaid volunteer work for the commons for free.

For the Linux kernel - and for GNU/Linux as an OS - the GPL was never a problem. On the contrary: Without it, the kernel would not be what it is (the historic Torvalds - Tabenbaum debate shows the reasons quite clearly).

The same is true, by the way for the GNU Guix project: Very few arguments against it are logical or really hold water. It is, for example, perfectly possile to publish a Guix channel for a commercial game or proprietary software on ones own web site - just as you can do with a docker image. And the way Guix takes care of dependencies and updates makes it probably the easiest way to do such. Because you can use Guix as a package manager on most distros - I use it on Debian - I find it more reasonable than using snaps, docker, or flatpaks.

The reason for the hate is likely something else: That you can’t commercialise and embrace the whole, source-based ecosystem.

For example, you can’t use the Guix infrastructure to promote or advertise commercial, non+free stuff. But expecting or demanding to be able to do that is just ridiculous: Nobody expects from Microsoft or Apple that they feel obliged to promote or market Adobe or Oracle products. Alone the idea that companies could demand that shows how entitled the corporations behave in respect to free software.

Personally, I also think that much of the hate which one would find e.g. on reddit or the orange site against copyleft licenses in general, and the GPL specifically was probably astroturfing from corporations for which it was just an obstacle to appropiate more work for the commons for free.

The same is true, by the way for the GNU Guix project: Very few arguments against it are logical or really hold water. It is, for example, perfectly possile to publish a Guix channel for a commercial game or proprietary software on ones own web site - just as you can do with a docker image. And the way Guix takes care of dependencies and updates makes it probably the easiest way to do such.

The reason for the hate is likely something else: That you can’t commercialise the whole ecosystem. For example, you can’t use the Guix infrastructure to promote or advertise commercial, non+free stuff.

But expecting or demanding to do that is just ridiculous: Nobody expects from Microsoft or Apple that they feel obliged to promote or market Adobe or Oracle products. Alone the idea that companies could demand that shows how entitled the corporations behave in respect to free software.

As far as I know I’m the only person who uses the term “vibe bobsled” and, well, I doubt it’s a term that’s particularly likely to catch on, but I find it personally useful.

Bobsledding, if you are unaware, is a particularly strange and interesting sport. It’s a lot of fun, but you don’t have a lot of agency in it. You sit in a bobsled, you go down an icy track, and really, there is only one way to go.

[…]

The vehicle is the LLM, you are the passenger. And I think the amount of agency people have over their journey is greatly reduced from what they feel like it is. More than just a slippery slope, it is a pre-crafted journey.

At the top of the chute, people tell themselves they’re going to use these tools as a kind of fancy autocomplete. As they descend, they say they’ll spin up some agents to explore ideas, but they’ll write the code themselves. Next their agents are generating the code for them, but don’t worry, but they’ll review all the output. Soon they’re plummeting downward and well, they don’t actually review the code being spat out much anymore, but they trust the agents, heck maybe the agents are actually better coders than they are they say. And where does it go from there? From “I don’t even code anymore” to “I don’t even prompt anymore”?

At every stage of the process, the coder in question removes themselves from the process of producing code, and gives in towards a faith-based initiative of code production, that the LLM knows and does a good job of what it’s doing. But what is the source of gravity pulling the sled along this icy chute?

It’s simple. Generation is not the slow part of coding. Theory-building and review are. And plausible-enough things are extremely hard to debug and understand. But the machines are so fast at producing things. If you are going to review their work, you aren’t really taking advantage of their most powerful property, which is speed. But theory-building and review are also the programmer’s most important role.

Setting Caps Lock as the Hyper modifier key.

Setting a keyboard layout that is comfortable for both typing programming symbols and the own language (I use UK international). Consistency is king here.

And configuring the compose key.

Otherwise, I don’t customize much, I think of it as a time sink.

Linux has a strict separation of user data and system.

Anything that you can configure and adust without entering the root password is stored as user data in your home account.

If you back it up and restore it on another computer with the same distro, it will work.

With standard PC hardware, that would work.

Some things like firmware or drivers can be auto-detected and installed on demand. Symptom would be that e.g. your WiFi or video card does not work.

The safe way is to back up user data (including config) and install newly. This will work almost always as the kernel provides hardware abstraction - that is its main purpose.

Installers like Debian are also quite quick… quicker than restoring a backup.

Unfortunately, one of the things that Debian is, is almost unable to manage packaging modern dependency trees. While more recent distributions like Guix can recursively import dependencies from a dozen programming languages’ package repositories, with a result that is generally acceptable to add to the distribution, Debian’s policies don’t make that very possible for a progam to accomplish. Perhaps some will use LLMs to do that. If they succeeed, Debian will become dependent on proprietary software for development, while still needing people in the loop, doing even less appealing scut-work.

It is very interesting what Guix achieves.

I run Debian on all my home systems, increasingly with some Guix on top.

Yesterday, I was setting up an old X220 Thinkpad with Debian stable for a purpose of quick data entry in my living room. (My work Thinkpad can’t do that because due to some hardware problem, it crashes when one tries to suspend it and closes the lid). I wanted to have Taskwarrior on it, which I want to sync with the Taskwarrior instance on my Gemini PDA, for things like shopping lists. The PDA runs Sailfish OS, which contains Taskwarrior.Turns out it is Taskwarrior 3, which has a new sync protocol. Debian Trixie still has Taskwarrior 2 which is not compatible.

The easiest solution? Installing Guix as package manager and “guix install taskwarrior”.

And Guix also has a current Rust compiler and cargo build tool, which is nice for building and running personal tools,and things like jujutsu.

Still, Debian is the base, and the contributions of Joey Hess and the other maintainers are invaluable.

AI will generate code that is mostly correct, the human time cost of (properly) reviewig this code and correcting it is generally less than also writing it.

I do not see it play out like this in practice, for example at my workplace. The work of debugging, reviewing, verifiction, and maintenance is simply offloaded to other people. Which, as a result, have more work, not less.

BTW, I think LaTeX with git is more customizable and extensible and should be taught instead of teaching MS Office stuff.

Right. I am using it over 30 years now, for uni reports, phd rhesis, talks, letters, CVs, and never had any regret.