Goodbye GitHub ... ?

[astyle]

Ahhh... Did you guys know that most of the Ports Collection's source tarballs are actually hosted on Github?

A lot of projects moved to Github precisely because because it's unsustainable for the smaller guys to maintain the metal it takes to host all that code.

Also - it looks like people are beginning to forget what it's like to cooperate with each other. I'm seeing a proliferation of one-dev shows. Used to be one-dev-show Linux distros (which get abandoned after a year or two), now it's one-dev-show for almost anything.

An AI won't give you a hard time for forgetting to run your code through CI/CD (a.k.a. Jenkins).
An AI won't give you a hard time for planting a backdoor in the code (like ban you from the project entirely).
An AI won't assume that you even understand the code that it generates. Heck, an AI won't even ask for your permission for anything.

As for any kind licensing for code - it's entirely an honor system, and a pretty toothless one at that, out there on the Internet. Code on any publicly accessible repo (be it Github, GitLab, codeberg or whatever) will be stolen from everyone left and right, because there's very little chance the author has the money (or other resources) to pick an actual fight in court over the license violation. Or is the author gonna pick a digital fight (like actually go learn some techniques beyond fork bombs) ?

 

[JohnK]

I'm guilty of going 'one-dev' lately. Mainly because I'm sick of the slopers and I'm finding it harder and harder to ignore the fact that a slop machine will just digest my code (my projects may be dumb or my code not genius but each line was hard won).

Besides what's the point? It's not like 10/15 years ago when people would create open source just to sell their support; if you were smart enough you could have some pretty cool code/things for free. Now-a-days people can just get their code from the nearest slop machine.

For me, programming is a hobby and I'd prefer to collaborate with smart(er) people but they're all burned out.

 

[Cthulhux]

As for any kind licensing for code - it's entirely an honor system, and a pretty toothless one at that, out there on the Internet.

There have been court decisions about that, and you’re wrong.

 

[Cthulhux]

For me, programming is a hobby and I'd prefer to collaborate with smart(er) people but they're all burned out.

The secret to not burning out with one-dev projects is to just let them rest for a year or two every now and then. It’s what keeps most of mine going.

 

[doul]

Okay, I think in the end I won’t use either GitHub nor Codeberg. Your messages have pushed me to take the plunge and I’m going to leverage the full power of FreeBSD (ZFS, jails, etc.) to move to self-hosting. Currently I have a Gitea server running locally, but there’s still a lot of configuration left to do. Plus, it gives me an excellent excuse to tinker even more with computer stuff using FreeBSD, which is definitely the most important thing :-D

 

[hruodr]

Currently I have a Gitea server running locally, but there’s still a lot of configuration left to do. Plus, it gives me an excellent excuse to tinker even more with computer stuff using FreeBSD, which is definitely the most important thing :-D

To know git is sure very good, it is ubiquitous, but something like git makes sense only for big projects with a lot
of developers.

Try fossil, it contains SCM and tracking system integrated. Runs almost out of the box, with few configuration. You can
have many repos, one for each project, you can easily move them to other computers or servers, backup them. For personal
work has a lot of advantages over git.

 

[doul]

Try fossil,

I’m definitely going to give it a try, I had never heard of Fossil before, but after taking a quick look I find it very interesting. It’s in the BSD spirit, with natural integration of SQLite and Markdown, and, above all, it’s a complete project management system.

I fully understand that while the way things work may be identical in certain points, it is fundamentally different at its core : fossil / git

In any case, to find out, I have to try it.

 

[astyle]

There have been court decisions about that, and you’re wrong.

yeah, and take a look at who was involved in those court fights - it was definitely not the one-dev shows (that we see so many of these days).

These days, there's just way too many one-dev shows for a court fight to do much good. Even Microsoft gave up trying to fight the piracy of its Office suite. Even Sony backed off the widespread DMCA harassment. Apparently Neflix still issues DMCA notices here and there, but do note, Netflix is a big player, they have the resources to pick a fight over a copyright license violation, unlike a random joe who has a few KB of code published on Github.

 

[JohnK]

I fully understand that while the way things work may be identical in certain points, it is fundamentally different at its core : fossil / git

Yes fsl and git are different at their core but you do not need to concern yourself with any of that. if you go down the fsl route, do as I suggested and set up a dedicated directory to host your bare repositories; the directory could be on the computer you use for dev or on another machine (but not both) as long as you have direct or ssh access you should be good.

If you host that directory behind ssh access I also recommend a dedicated user on that system.

 

[Cthulhux]

Even Microsoft gave up trying to fight the piracy of its Office suite.

I am not interested in giving up the rights to my software.

 

[hruodr]

It’s in the BSD spirit

Well, fossil is not really 100% unix, but in this special case one closes his eyes.

Fossil is a Swiss knife, it has one binary for everything, it acts as server and as client, contains everything, the vcs, forum,
wiki, ticket software, etc, all in one binary. As server it can act alone, through inetd, as cgi. Really not unix.

One could say, It belongs, as sqlite, to the tcl / tk environment. The repositories of tcl / tk are fossil repositories.

An again, the repository, is one file containing everything, all data including the wiki, forum, tickets. It is a sqlite3 database.
Fossil is a proof of the power of sqlite3.

If you host that directory behind ssh access I also recommend a dedicated user on that system.

fossil serves through https, either alone or through a web server, no ssh necessary.

 

[Cthulhux]

The repositories of tcl / tk are fossil repositories.

And a variant of Tcl is Fossil’s scripting language.

 

[JohnK]

fossil serves through https, either alone or through a web server, no ssh necessary.

sounds like you may have a few new prompts for the slop machine (I wonder what could possibly be my reason(s) for recommending ssh(1)).

 

[astyle]

I am not interested in giving up the rights to my software.

It's like you have something that EVERYBODY actually wants, AND you have the money/resources to pick fights with people who don't follow your rules.

 

[Cthulhux]

It's like you have something that EVERYBODY actually wants

If nobody wants it, why do companies add it to their billion-dollar copyright violation machines?

 

[hollowone]

Hello,

Goodbye GitHub: that’s what I would have liked to say...

Codeberg is my original repository; I added GitHub to use Codex Cloud, which I no longer use, relying instead solely on the Codex CLI. I don’t like the GitHub interface: it’s slow to load, the Markdown rendering is unattractive, and it creates the extra burden of maintaining two separate repositories. While GitHub offers high visibility, it’s clear that, after two years, it hasn’t provided me with any real additional benefit. Furthermore, my increased confidence in Codex means that maintaining a second repository to isolate coding agent is no longer justified. For remote backups, I use Proton Drive to store my code and Git files.

Things have evolved: I’ve become more proficient with Git, I have greater trust in Codex, and I have an additional backup on my work computer. I plan to keep my triple-branch structure -> `codex/<feature>` -> `test` -> `main` on Codeberg, while putting GitHub on hold for the time being. The only potential drawback with Codeberg is their stance on AI; they discourage large volumes of AI-generated code. While my own work doesn’t fall into that category, their criteria aren’t strictly quantitative, and they could potentially reject any AI-generated code outright.

What do you think of? How do you handle it on your end? Thanks.

everybody has a preference. I personally use my own self-hosted gitea out of any public git hosting for my own projects. I use github for something I like to abandon yet give to people. at work I use self-hosted in the corporate cloud we use. I use AI in my coding here and there so Codeberg as any other strict anti AI or hating AI policy is a bit to philosophical than practical. All I care is data security vs. compliance required and there is never a single answer to that. it always depends on customer and project.

 

[astyle]

If nobody wants it, why do companies add it to their billion-dollar copyright violation machines?

so that you give up the rights to your software. Just because you made it, that doesn't mean you have it in you to fight in court for the right to do what you want with it. You want the big guys to fight for you over software you wrote - you give up the rights to your software. Your only real choice is to write software or not. Once it's published, all rights are gone, and the big guys are the only ones actually capable of picking a fight and inflicting any kind of real pain.

 

[Cthulhux]

You want the big guys to fight for you over software you wrote

I do not.

Once it's published, all rights are gone

This is - from a legal perspective - entirely wrong, regardless of whether I can enforce my rights or not.
Frankly, anyone using “AI” to get any code written probably violates my rights at some point, so “AI” users basically commit a crime against me.

And I make sure to let them know. They have voluntarily decided to violate my rights.

Curious: how do you communicate with people who commit a crime against you? I’m always open for more ideas.

 

[astyle]

Curious: how do you communicate with people who commit a crime against you?

See? you can't even track down who violated the license terms of your software, let alone pick a fight.

This is exactly why I'm saying that software licenses are franky an honor system that is toothless.

What rights are you even talking about if you can't enforce them anyway? What deterrent do you have against the violators? Practically nothing.

 

[atax1a]

gout hub

 

[Cthulhux]

See? you can't even track down who violated the license terms of your software, let alone pick a fight.

I see: Theft is only a crime if you find the thief.

What rights are you even talking about if you can't enforce them anyway?

So theft is obviously legal, as you can't enforce your stuff not being stolen, ever.

I'll end this here, before I lose my temper.

 

[JohnK]

"ethics" and "abilities" aside, I do feel as though the slop robots will be the death knell for a lot of things/methods we're used to because I've noticed a lot more people who refuse/unwilling to dive into the weeds. Most people just seem more complacent lately.

Even if it's for a brief period can we bring back some good old-fashioned editor fights instead (FWITW, I use vim; come at me)?

EDIT (editor): I should note though, I wasn't around for the 'editor wars' and I learned about emacs after I learned vim so my fight may not be terribly "strong" (but i promise to try and be a fun opponent).

 

[Crivens]

FWITW, I use vim; come at me

Did you say Atari?

 

[6502]

So theft is obviously legal, as you can't enforce your stuff not being stolen, ever.

I'll end this here, before I lose my temper.

If you want to keep copyright, release executable without open source.

 

[JohnK]

Did you say Atari?

wait!? ugh! I assume that's an "insult" I'm supposed to react to, but I don't know that one so I cannot gauge a proper response (that's not fair--defeated in the first round! you need to start less obscure). ...may I just say: "u r a dumb head"?