[jakelazaroff]

Most people have no problem with non-open source software. The gnashing of teeth comes in when projects like Terraform become successful specifically because they're open source, and then the maintainer changes to a closed source license that would have prevented the project from being successful in the first place.

Doubly so when they relicense outside contributors' work with a closed source license because those contributors signed a CLA.

[arp242]

Quite a few people commenting here are having problems with it in the case of Bearblog though, including some pretty wild accusations.

And lets be real here: https://github.com/HermanMartinus/bearblog/graphs/contributo...

Looking at the details of that, the only two (small) substantial code changes from other people are "User can delete their own account" from 2020, and "Use cloudflare online dns api to perform domain check" from 2021.

[jinzo]

If we are real, it's also quite clear that contributions are not accepted at least from 2023. And the Readme talked about the project not meant to be self-hosted in the past.

I have no horse in the race, just think that maybe this project is not a good measure of contributions.

[jakelazaroff]

> Quite a few people commenting here are having problems with it in the case of Bearblog though, including some pretty wild accusations.

That's why I said "most people" (of which I think HN commenters are not a representative sample) rather than "nobody" :)

[8organicbits]

The trick is not to get attached to a project name. `Terraform` is a trademark of IBM (previously Hashicorp). Terraform used to refer to an open-source IaC project, but now it doesn't. OpenTofu, https://opentofu.org, is probably the most accurate name for the continuation of that open-source project.

[account42]

This is of course the correct way to proceed as an affected user but does not mean the bait and switch cannot be criticized.

[_puk]

Sounds like we need a "forever open source" license.

A commitment that any significant derivative retain the original (or some later version) of the original license.

"Free to do whatever as long as it retains this license. A commitment that this license will not change, even by the original author".

No special cases, just a blanket license for all derivatives.

If it exists, what are the barriers to adoption? Why don't we all use it?

[Aurornis]

> A commitment that this license will not change, even by the original author

Unless you’re entering into a contract with the project maintainer (which you’re not, if you’re just downloading or using it) then such a commitment means nothing.

Applying an open source license to your work means you’ve licensed other people to use it under those terms.

You can make all the commitments you want in the license, but it doesn’t actually commit you to keeping all future work open source as well under the law.

So you could write this license and make the commitment, but if you changed your mind later and decided not to open source future commits to the project that you made then nobody could stop you. Not unless you had entered into a contractual agreement with them and exchanged some consideration (money).

[yjftsjthsd-h]

I'm really not a lawyer, but I'm skeptical that such a thing is even possible; is it legally possible to say that you as the copyright owner will never relicense something?

(What I'm given to understand does work is using a copyleft license and taking code from multiple parties without a CLA, because then relicensing requires all the copyright owners to agree, which for a large enough project is impractical.)

[dragonwriter]

> I'm really not a lawyer, but I'm skeptical that such a thing is even possible; is it legally possible to say that you as the copyright owner will never relicense something?

It’s possible to say anything. Without something like a contract with reciprocal commitments to make it binding, the legal effect of saying it is limited (though not necessarily zero, because legal concepts like promissory estoppel exist.)

[_puk]

How about a standard entity "OSI perhaps?!", that commits a file to an early stage of the repository (could this be automated), who then cannot / will not give approval for a relicense?

[j1elo]

Relicensing still can be done, just keeping that file out. (and reimplementing it the same but with new code, if it was really needed for something important)

[miggol]

Love the idea but the thing is, if it's just one file then it's probably easy enough to work that contribution out of the repository.

It is quite fun to try and think of ways that this could work though. Perhaps a bot that code-paraphrases (paracodes?) every accepted PR. Or maybe there's some crypto magic you could do to make the only option a clean room rewrite.

[sakjur]

I’d suggest the GPL family without a CLA as an approximation of that intent.

> If it exists, what are the barriers to adoption? Why don't we all use it?

My theory is that people in general don’t care that much, or (particularly in the case of corporations) consider permissive licenses to be ”freer” than copyleft.

[account42]

Maybe this makes more sense as a change to trademark law than a license thing. If a name OpenSourceProject has gotten popular due to being open source then I would say it's in the interest of the public to ensure that OpenSourceProject keeps referring to an open source project an isn't misappropriated for something else. It's a kind of bait and switch fraud that I think should be generally illegal. On the other hand I don't see why we should care about the creator not being allowed to re-license his own work as long as no one has been mislead about what they are getting.