Hacker News Reader

Comment by bruce511 | Hacker News Reader

You can do that. There are lots of models. Just don't call it Open Source. That term has a meaning, and you may not like the meaning, but that doesn't mean you change the term. It means you publish under a different license.

View on HN

Working on a OSS product myself, there's a much darker side I've seen played out many times by many companies including well known ones like Deutsche Bank and Royal Bank of Canada where they abuse your time and knowledge by making you think they will purchase support from you but end up never doing it.

The problem is, the assumption under which the "old" OSS models operate no longer hold true - primarily, that the actors (which used to be universities) play fair and contribute back to the ecosystem, and there is no reckless exploitation by single actors that kill off other actors who do contribute back.

I do love open source software, I contribute to open source software in all kinds of ways from filing bug tickets over develop workarounding hacks to legitimate enhancements. I have no problem with small and medium businesses can profit off of that.

But I - and many others - have a serious problems with (especially) the large hyperscalers doing just about nothing to contribute back. We need licensing models that prevent hyperscalers squeezing out the rest of the ecosystem like fucking lemons and keeping all the revenue.

> But I - and many others - have a serious problems with (especially) the large hyperscalers doing just about nothing to contribute back. We need licensing models that prevent hyperscalers squeezing out the rest of the ecosystem like fucking lemons and keeping all the revenue.

If you have a problem with this, license the product, for example, under GNU Affero General Public License (AGPL).

>> The problem is, the assumption under which the "old" OSS models operate no longer hold true - primarily, that the actors (which used to be universities) play fair and contribute back to the ecosystem, and there is no reckless exploitation by single actors that kill off other actors who do contribute back.

I think it's more likely your assumption that "fair" has got anything at all to do with licenses. It's never been about "fair" and it's never been about "contributing back". (I'm not aware of either of those terms appearing in any license I regularly encounter.)

Open Source is not about "fairness". And it's certainly not about "contributing back". It's a license under which software can (and in some specific cases) must, be distributed.

To me, the problem is, that people don't understand what is actually written in the licenses, and so in that absence make up all kinds of alternative ideas about what it means just from the title.

>> But I have a serious problems with (especially) the large hyperscalers doing just about nothing to contribute back. We need licensing models that prevent hyperscalers squeezing out the rest of the ecosystem and keeping all the revenue.

Fair enough. Then stop developing under a license that let's them do exactly that. Clearly (some / all?) OSS and Free licenses do not fit your goals. So stop developing under them.

What you do with your time is up to you. What license you choose for your code is up to you. What projects you contribute to is up to you. But don't complain about others when they simply exercise their rights under those licenses.

> It's never been about "fair" and it's never been about "contributing back".

I've agreed with basically everything you said everywhere in this thread, but this is, imho, historically inaccurate.

Both Free Software and Open Source exist because of socio-political reasons, that are then implemented via legal means.

Free Software was born because Stallman thought that he should be able, as a user, to use, inspect, modify and redistribute modifications to software that ran on his own machine, because that's the moral thing to do. This can be enforced via licenses that make sure the user have this freedom.

Open Source is a software development model, inspired by the development of the Linux kernel, born because its proponents were convinced that the more people worked on a piece of software the higher the quality of that software would be. This can be enforced via licenses that make sure developers have the freedom to inspect, modify and redistribute their modifications (plus a bunch of other things listed in the OSD[0])

In other words, the licenses are just the means through which you realize, on one side, the "fairness" (or the ideological/political objective, in the case of Free Software) and on the other the "contributing [back]" (or the technical and social objective in the case of Open Source Software, even thought I agree this one's a bit of a stretch).

[0] Which is, in a way, more restrictive than the Free Software definition. This is, IMHO, ironic considering that many people view the FSF as the zealots and the OSI as the pragmatists.

I hear you, and I agree with your summary of the origins. From Stallman's point of view the "fairness" (if we view it in that light) is that the _user_ has all the rights, and the _developer_ has none. He wanted to fix a printer driver (as a user of that printer) and so he believed he should have access to that code.

He approached it very much from the user point of view - and certainly that is born out in the Free licenses, and in many ways (sans virality) in Open Source ones. If "fairness" is implied, it's very much in the direction of the user (which makes sense given the context of the time.) I would argue that the FSF licenses skew so heavily towards the user that they are anything but fair.

What I have a problem with is that "current" companies (the object of this thread being a case in point) want to water down these user rights. They want to give more rights to developers. They want developers to have some control - to limit who can run the software, who can profit from it, and so on. All in the name of "fairness" so the little guy can stand up to the big guy.

I'm all for fairness. I'm all for companies having the right to determine how their work can be used, and by whom. I have nothing against licenses that promote that. All I note is that those are not Open Source licenses. Most of my own projects are under a source-supplied license. It's not Open Source. And I don't claim it is.

The equivalent argument exists about giving back. Stallman wanted to edit the driver and give it to his friends. He wasn't driven by the desire to push those changes back to the original authors, and he wasn't driven by any moral impetus to require people to feed things back to him (in terms of code, or money.) Given how much effort he put into the license I can't assume he just "forgot to put that in."

Does some section of the community contribute back? yes, of course. Is it a requirement of the license though? Specifically not. OSS and Free licenses go out of their way to confer freedoms on the -users-. Some (very, very, tiny) percentage of users feel a moral obligation to donate finances, or contribute code. And there's nothing wrong with that (obviously.) But the license does not implicitly, or explicitly, require that.

One can of course complain that "big company x" is using work without "giving anything back". But when they are using it within the terms of the license, a license you as the author explicitly chose, it rings a bit hollow to me.

Of course lots of companies abuse the term because they want the benefits (marketing, technical etc) of declaring themselves Open Source. But then to release under a non-open-source license is slimy (at best). Or they release under the OSS license, and complain when the product is adopted by "the wrong people." Which (at best) just means they didn't bother to understand the license they were releasing under.

Reply on news.ycombinator.com