Hacker News Reader

I agree with nearly everything you've said, except I would suggest that the maintainer himself started the framing of him as The Bad Guy.

The thread has changed a lot since I first saw it, but right out of the gate the issuer opener was matter of fact and mostly respectful, and the maintainer called him a spammer and a troll and deleted his and other's stuff.

I can't imagine you could do much better if you wanted people to think you're The Bad Guy.

It's been my observation over the years that there are people who troll for this sort of thing and they are good at saying things nicely but then they don't actually meet the criteria of being anyone who had their "rights" violated or are willing to do anything other than "go away" when the license is finally reapplied after harassment over a minutia with no actual harm being proved to an individual. They may be technically correct about the license concerns, but the absoluteness and blaming they do isn't worth the outcomes.

I just closed the repo I'm working on to the public and will not be making it Open Source moving forward as I originally intended. I've been through this before myself, and it's just best to leave my code unlicensed and copyrighted by me. I'm grateful this ugly beast reared its head and reminded me how much vitriol non-contributors can deliver.

I just closed the repo I'm working on to the public and will not be making it Open Source moving forward as I originally intended.

I'm really sorry to hear that, and while it is obviously 100% up to you, I hope you to reconsider.

Every contribution to The Commons furthers the public good. Unlike the busybodies who care more about lecturing people about following the "rules" of "Open Source" to the letter, than about the spirit of fostering contributions to The Commons to further the public good, you clearly understand what actually matters.

Don't let the annoying minority keep you from doing something you know will make a positive difference in the world. If you set clear boundaries and stay matter-of-fact, the public overall will back you.

In the same spirit as the 4 "non-Free-Culture" Creative Commons licenses, I encourage you to check out non-open-source PolyForm software licenses, including a noncommercial license, a free trial license, etc: https://github.com/polyformproject/polyform-licenses#license...

Every contribution to The Commons furthers the public good, regardless of whether it's technically "Open Source" according to the Debian mailing list.

The maintainer seems reluctant to come to senses, pity. https://github.com/ovity/octotree/issues/1041

Not only my issue got closed as a "spam", my account now has been blocked by "their organization". http://ix.io/2CCK/text

I really hope Github's learning and (job) recommendation algorithms do not take this into account.

The pathway is a DMCA takedown notice. It's rarely appropriate, but this is the time for it.

A DMCA takedown can only be sent by the owners or someone acting on their behalf. (Though, microsoft is more than fine on making exceptions for the RIAA in the case of youtube-dl I doubt they will make an exception of individual devs)

Legal twist: Since the project is under AGPL, this technically makes anybody who downloads/uses the project a copyright owner (you have the right to make copies), thus anybody could send a valid DMCA request to the repo.

Obviously we're a bit in uncharted territory and there's some nuances and things vary by jurisdiction, it should be receivable one way or another, and it would take 10 years in court to really determine it.

That is absolutely not how the AGPL nor (US) copyright law works. If I license someone else to make copies, they do not become a "copyright owner". You are allowed to make copies of the versions of the work released under AGPL, but not newer versions not released under those terms.

>>> ... but not newer versions not released under those terms.

That's exactly where shit gets really complicated ^^

1) The software couldn't be relicensed in the first place because there are parts for which they don't own the copyright. Can't relicense something you don't worn. The attempt to change the license of the software is void at best.

2) The AGPL is contagious, forcing all future contributions to be AGPL as well. Their claim that contributions started being under another license at some point is void, because these contributions were necessarily under AGPL.

Notice there's an interesting conundrum in how the two elements interact and create a deadlock. It's on purpose, the GPL/AGPL is meant to prevent relicensing.

It's copyright violations all over the place. There's gotta be a thousand and one ways to make them accountable. You could personally start redistributing/reselling the software assuming it's AGPL so you get some grounds and go after them for saying it's not.

... that's not at all how copyright law or the AGPL works.

I think you accidentally read some old Microsoft or new Google propaganda somewhere.

> I really hope Github's learning and (job) recommendation algorithms do not take this into account.

Github is becoming too powerful.

They seem to have cooled off now. They apologized for being dismissive, thanked people, and discussion has become much more civil.

https://github.com/christianbundy/octotree/issues/1#issuecom...

I have never heard of octotree before this, I haven't used it before, I don't plan to use it, and I don't have any stake in the outcome.

Thus, that said, as a completely disinterested party, the Octotree maintainer really does appear to be doing his best to come off as The Bad Guy. All of his behavior in the relevant Github issues looks like someone who is having a tantrum and wanting to take his ball and go home.

So, while I pretty much totally agree with everything you've written, especially the part about "I must really commend the noncombative, factual tone taken by the issue opener" it's also hard for me to fault others who are letting their annoyance show when simple, factual requests are met with tirades of assholery.

I share your view in part, but I think you've ignored the basis for which the maintainer is reasonably understood as a malicious actor: it's not the overreaction, it's the deceit.

They said that they had conducted a complete rewrite. There is no reason to do this other than to mislead and create a self-serving false narrative.

And what is the malicious act? To get rich off of the highly profitable code in those pull requests?

The "deceit" is saying whatever to get people to go away, not to hurt anyone, or to profit off of anyone else's hard work. The right thing to do is just to ask people to assign copyright (see eg https://news.ycombinator.com/item?id=24954772 ), but lacking the maturity do what a responsible adult needs to do is just immaturity, not malice.

You're correct that the motive is important in assessing the act, but so is the willingness to correct the issue when notified of it, and the manner in which one responds. "Immaturity" rapidly loses any cachet of innocence when it persists in the face of experience.

More important than motive is the real impact caused by the act. And the realistic harm caused by the refusal to ask people nicely for copyright assignment or to rewrite those contributions, other than to this person's own reputation, is pretty insignificant.

This person just wants everyone to go away. It's disrespectful to the community, ill-mannered, and utterly graceless, but I find it hard to call it malicious.

The author IMHO can relicense their code to whatever their like.

Based on package.json only one author mentioned, Buu Nguyen <buunguyen@gmail.com>

[1]https://github.com/ovity/octotree/blob/master/package.json

There are (at least) 66 distinct contributors, according to the commit history.

Yep, lots of them are still the most recent author according to git-blame too: https://github.com/christianbundy/octotree/issues/1#issuecom...

I didn't flag your comment below, but just to be clear, at least in the US the law considers any individual contributor to be the copyright holder of their own code. They can assign or license it to someone else, but they're the original copyright owners of that code.

Copyright ownership is not a matter of your opinion. It is settled law, mutually recognized by the 179 countries (out of the 195 in the world) who have signed the Berne Convention [1]. Copyright ownership is established as soon as a work is "fixed in a tangible medium of expresssion" [2]—in other words, as soon as the contributor wrote it on their laptop. The "author" entry in the package.json file is irrelevant.

[1]: https://en.wikipedia.org/wiki/Berne_Convention

[2]: https://www.plagiarismtoday.com/2017/10/03/with-copyright-wh...

1) There's definitely not 179 countries recognizing US copyright.

2) The Berne convention is a farce, it doesn't define what's supposed to be enforced or how. It's up to each European country to do their own thing.

You really can't say that copyright is anything settled or mutually recognized across the world. It's quite a mess and it's really complicated.

Nothing you wrote is a meaningful retort to the comment you replied to. Opinion at best.

> The author IMHO can relicense their code to whatever their like.

Assume it's true and consider what it can lead to. Person A writes code under GPL. Person B extends that code significantly, also under GPL. Person A relicenses his code to a closed source.

Now person B, of course, is the rightful owner of his own extensions, but they don't make sense without person A's code. So suddenly person B's work greatly decreases in value.

I'd assume it's not as simple as just relicensing one's own code; other contributions, depending on that, should be taken into account. Or you can't just relicense.

While broken, I suppose this is what would happen:

1. Person A writes project X under GPL

2. Person B extends project X under GPL

3. Person A removes all traces of contributions from project X and releases under non-GPL

4. Person B rightfully continues to use project X under GPL, possibly even hosting a fork and continuing collaboration.

(I'm not sure if step 3 is feasible - while person A is free to use their own copyrighted work however they please, it might not be possibly to legally erase all trace of foreign contributions.)

---

What normally should happen is that Person A needs to contact all contributors, and have them agree to the relicense their contributions. If all authors agree, then there is no issue. However, step 4 still applies - everyone receiving the code under GPL is free to keep using (and distributing) it under GPL.

Only contributions done to the relicensed work will be uniquely protected by the new license.

My understanding is that such relicensing rapidly becomes impossible once there are more than a very few significant contributors. Most proponents of Open Source licenses would consider this a feature, not a bug. In any event, this is one of the reasons that maintainers of major projects (including GNU, IIRC) frequently require copyright assignment as a requirement for copyrightable contributions.

> Most proponents of Open Source licenses would consider this a feature, ...

Indeed ...

> ... maintainers of major projects (including GNU, IIRC) frequently require copyright assignment as a requirement for copyrightable contributions.

... and many folks, including myself, refuse to agree to any such agreements.

Personally, I'd prefer BSD, MIT, or similar anyways but... if a project is GPL, one of the best ways to ensure that it remains GPL is for individual contributors to retain their copyright on any contributions.

Take those paragons of open source, VMware [0] and the "open-vm-tools" [1] project, for example. For a variety of reasons, various components are under different licenses:

> The code is being released under GPL v2 and GPL v2 compatible licenses. To be more specific, the Linux kernel modules are being released under the GPL v2, while almost all of the user level components are being released under the LGPL v2.1. The SVGA and mouse drivers have been available under the X11 license for quite some time. There are certain third party components released under BSD style licenses, to which VMware has in some cases contributed, and will continue to distribute with open-vm-tools.

Okay, so it sounds like it's all open source and should remain that way so I shouldn't have any concerns about contributing bugfixes, enhancements, patches, etc., right? Wait for it...

> Do I need to sign something before making a contribution?

> Yes. We have a standard contribution agreement that covers all contributions made to the project. It gives VMware and you joint copyright interests in the code you are contributing. The agreement also gives VMware flexibility with licensing and also helps avoid any copyright/licensing related issues that may arise in the future.

There it is. Nice choice of wording, by the way: it gives VMware "flexibility" with licensing and "helps avoid any copyright/licensing related issues".

VMware certainly is fond of "flexibility with licensing"' as history has shown.

--

Besides all of the other arguments, I think it's completely absurd that an individual should be required to enter into a one-sided legal agreement with a large corporation just to contribute some code -- even just two or three lines -- that they've already written and just want to give away.

--

[0]: That's sarcasm, just to be clear!

[1]: https://github.com/vmware/open-vm-tools

Person A still have the IP of the software IMHO, so person A can do whatever with the software. Person B still have the latest-before-relicensed software, person B can fork or create new software based on latest-before-relicenced software.

Similarity like Qt software company doing the relicenced thing and Illumos project keeping latest-before-relicenced software.

This is not true: person B owns the copyright to his/her contributions. As soon as person B added code, the copyright of the project no longer belonged solely to person A. And that's not a matter of opinion but legal fact.

It would however be possible for person A to remove all contributions of person B (and possibly rewrite the same functionality).

If you have released a work under GPL, that work stays under GPL. The term "relicensing" here means that going forward, new new works will be released under new licenses. So version 1 can be GPL, but version 2 can have other terms. We often don't think of new software releases to be independent works, but they kind of are.

Nothing can change the fact that version 1 was released under the terms of the GPL, but that does not mean you can use version 2 released under different terms.

Edit: this is assuming the maintainer is the primary author of the code. The current guidance from the FSF and other organizations that you need support from the authors of 95% of the work.

If you have received any contributions, no matter how small, then you are no longer the sole author.

All copyright holders must agree in order to relicense unless all such holders signed a contributor license agreement which signs over all rights to the original author, and as multiple contributions have been made.

Every single contributor must be contacted, and they must all agree.

(Also, of course note that relicensing does not invalidate previous licenses - if I have a copy from when it was under AGPL, then nothing will ever change that.)