[gdwatson]

“They convinced a bunch of developers that their definition of Open Source that was specifically crafted to protect business interests is canon.”

They adopted the existing Debian Free Software Guidelines as the Open Source Definition. The DFSG are good, actually, and represent an important community consensus outside the FSF.

[sarchertech]

They looked around and found the guidelines that most closely matched their goals. Specifically DFSG already included a clause about not restricting commercial use.

Also if you read the original DFSG the clause about field of endeavor has been interpreted by OSI differently from the intent.

It was about saying your license can’t prevent an end user of your software from using it for a specific purpose. It really says nothing about restrictions on how you can sell the software.

The problem is OSI is now the sole interpreter of the definition.

[microtherion]

> “Free software” does not mean “noncommercial.” On the contrary, a free program must be available for commercial use, commercial development, and commercial distribution. This policy is of fundamental importance—without this, free software could not achieve its aims.

— https://www.gnu.org/philosophy/free-sw.en.html

[sarchertech]

Sure. That restriction came later.

[account42]

It's not a new restriction it's a clarification of something that already followed from the original definition.

[fragmede]

Why is that the problem? Trademarks are one of the three branches of intellectual property. The two words "open" and "source" look like generic terms, but "Open Source" has come to mean a relatively specific thing. So have Disney and Google and Coca-cola.

[jraph]

The DFSG and the OSD are the same text, but the OSI and the Debian project interpret it differently, and this difference is important.

Debian (and most other distributions, btw), for the most part (or entirely, I suppose), agrees with the FSF / the GNU project when deciding which license is free or non free. The OSI has a more permissive interpretation.

RMS speaks about that in a recent interview in French [1], English translation below:

> La FSF a financé Debian à son commencement. Mais rapidement, le projet, qui comptait plus de contributeurs, a voulu formuler une définition de la liberté différente, avec l’intention d’être équivalente.

> À l’époque, j’ai commis une erreur : j’aurais dû vérifier plus attentivement s’il pouvait y avoir des divergences d’interprétation entre le projet GNU et Debian. La définition me paraissait équivalente, même si elle était formulée autrement. J’ai dit : “C’est bon.” Mais en réalité, il y avait des problèmes potentiels.

> Plus tard, quand l’open source a émergé, ils ont repris la définition de Debian, je ne sais plus s'il ont changé quelques mots mais ils ont surtout changé l’interprétation. Dès lors, elle n’était plus équivalente à celle du logiciel libre. Il existe aujourd’hui des programmes considérés comme “open source” mais pas comme logiciels libres, et inversement.

> J’ai d’ailleurs expliqué ces différences dans mon essai Open Source Misses the Point.

English translation (not a native English speaker, I hope the translation is ok, especially considering that RMS is close to his words and it's probably easy to misrepresent him without noticing):

> The FSF funded Debian at its beginnings. But rapidly, the project, gaining more contributors, wanted a different phrasing for the definition of freedom, which the intent of being equivalent.

> Back then, I made a mistake: I should have checked more carefully if there could be different ways to interpret it between the GNU and the Debian projects. The definition seemed equivalent to me, even if it was phrased differently. I said: "OK". But in reality, there were potential issues.

> Later, when Open Source surfaced, they took Debian's definition, I don't know if they changed a few words but they above all changed the interpretation. Since then, it was not equivalent to the free software definition anymore. There exist open source software that's not free software, and conversely.

> By the way, I explain all this in my Open Source Misses the Point essay.

[1] https://linuxfr.org/news/40-ans-pour-l-informatique-libre-en...

[account42]

Do you have an example of a license where the OSI and Debian / FSF disagree on whether it's free?

[jraph]

Yep, the Sybase Open Watcom Public License. The OSI considers this license open source [1], the FSF and major distros don't [2], including Fedora [3] and Debian [4].

It is notably used by the Open Watcom C compiler, which is used to compile VirtualBox's BIOS. Which is the reason why you won't find VirtualBox in most distros, including Debian.

The reason the FSF and major distros don't consider it free is that there are cases where you can't use it privately without releasing your modifications. The license doesn't pass Debian's Desert Island test [5]:

> Imagine a castaway on a desert island with a solar-powered computer. This would make it impossible to fulfill any requirement to make changes publicly available or to send patches to some particular place. This holds even if such requirements are only upon request, as the castaway might be able to receive messages but be unable to send them. To be free, software must be modifiable by this unfortunate castaway, who must also be able to legally share modifications with friends on the island.

I don't have an example of a license that the FSF / GNU project considers free and that the OSI doesn't consider open source.

[1] https://directory.fsf.org/wiki/License:Watcom-1.0

[2] https://opensource.org/license/sybase-php

[3] https://en.wikipedia.org/wiki/Sybase_Open_Watcom_Public_Lice...

[4] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=376431

> Oops... it looks like OSI smoked something especially bad this time, I'm afraid. This license looks like someone took his time to collect every single problematic clause.

[5] https://people.debian.org/~bap/dfsg-faq.html#desert_island