Toward a "modern" Emacs [LWN.net]

Toward a "modern" Emacs

Posted Sep 26, 2020 4:03 UTC (Sat) by rsidd (guest, #2582)
In reply to: Toward a "modern" Emacs by pizza
Parent article: Toward a "modern" Emacs

Erm, open source *is* free software. Almost every OSI-approved licence is a FSF-approved licence and vice versa. The very few exceptions are arcane and unimportant.

to post comments

Toward a "modern" Emacs

Posted Sep 26, 2020 6:08 UTC (Sat) by IanKelling (subscriber, #89418) [Link] (14 responses)

There's a lot of open source that is not free software, see the practical differences section of https://www.gnu.org/philosophy/open-source-misses-the-poi....

Also, the real world definition is that almost all OSes that run on hardware and call themselves "open source" include nonfree code. OSes that call themselves free software are much more likely to mean all the code is free software. The free software definition explicitly addresses which code to consider, and its put in practice in https://www.gnu.org/distros/common-distros.en.html, https://ryf.fsf.org/about/criteria, and https://www.gnu.org/philosophy/javascript-trap.en.html.

An exception to this rule is 8 bit arduino's which are often called open source and are fully free software (but the arduino development environment includes optional nonfree code).

Toward a "modern" Emacs

Posted Sep 26, 2020 12:17 UTC (Sat) by pizza (subscriber, #46) [Link] (13 responses)

In practice, it's "open source" for the vendor/developer, but "closed source" for the downstream end-user, because the former isn't required to pass along the same rights (ie the source code) to the latter.

In the real world, I've been forced to use a lot of proprietary software that is built on top of "open source" codebases. This includes multiple android devices and several variations of LLVM. After all, _I_ don't have the source code to what is actually running.

copyleft software, not "free" software

Posted Sep 26, 2020 18:36 UTC (Sat) by marcH (subscriber, #57642) [Link] (12 responses)

Thanks, I think this Nth million discussion about what "free software" means made me finally realize how poorly named hence poorly advertised it is. This is not just the confusion with "gratis" which does not help but is really not the main issue.

Of course I'm a bit slow: if something needs to be rephrased and discussed a million times then it's likely not just because everyone's stupid. Discussions that tend to get heated are also much more often triggered by misunderstandings than disagreements. Finally, who expects good names from software engineers? :-)

So the key concept/idea behind "free software" or more specifically "copyleft software" is transitivity: making sure not just the users but also _the users of the users_ are free to do what they want. This of course implies the first users are _less_ free to do what they want which means "free" is self-contradicting. Boom.

Even before someone wrote that "my freedom stops where yours begin" we knew freedom was a messy concept best avoided when possible. For some fun search for "weaponized first amendment".

This also seems like a classic case of trying to avoid a technical and unusual but very smart and extremely accurate term "copyleft" for the fear of confusing readers, dumbing it down and confusing everyone much more. And then not wondering what went wrong.

I will stop referring to GPL software as "free software" from now on, it will help save a fair amount of digital ink. "Copyleft software" it is.

PS: not sure yet how to best name BSD-like licenses. "permissive" is very vague.

copyleft software, not "free" software

Posted Sep 26, 2020 18:39 UTC (Sat) by mpr22 (subscriber, #60784) [Link] (2 responses)

Permissive is the best word I can think of to describe it.

(The other way I can think of is to describe the GPL as "socialist", the BSD and MIT licences as "liberal", and proprietary licences as "reactionary", but that doesn't work very well.)

copyleft software, not "free" software

Posted Sep 27, 2020 8:35 UTC (Sun) by Wol (subscriber, #4433) [Link]

Developer-Free, and User-Free?

GPL frees the *user*, BSD frees the *developer*.

Cheers,
Wol

copyleft software, not "free" software

Posted Sep 27, 2020 16:50 UTC (Sun) by IanKelling (subscriber, #89418) [Link]

https://www.gnu.org/licenses/license-recommendations.en.html uses "last pushover licenses". Aka pushover license.

copyleft software, not "free" software

Posted Sep 26, 2020 20:09 UTC (Sat) by smcv (subscriber, #53363) [Link]

"Permissive" is a term I've seen used frequently in e.g. Debian to refer to non-copyleft licenses (BSD, MIT/X11, Zlib, CC-BY, Unlicense, WTFPL and many similar licenses) as a group. Sometimes people draw a spectrum from strong copyleft (e.g. GPL) to permissive, with weak copyleft licenses like the LGPL in the middle as a compromise between the two.

Using "free software" to refer only to copyleft licenses and "open source" to refer to permissive licenses seems really confusing, since it's in conflict with the way the originators of those terms use them: the FSF considers both copyleft and permissive licenses to be free software, and the OSI considers both copyleft and permissive licences to be open source.

The free software movement and the open source movement (inasmuch as either of those is monolithic) have different goals and different focuses, and perhaps free software is more closely aligned with copyleft while open source is more closely aligned with permissive licensing, but I think a lot of people who consider themselves to be part of one community and not the other would agree that both styles of licensing have their place (although probably not on the specifics of which one is more suitable for a particular project).

copyleft software, not "free" software

Posted Sep 27, 2020 7:46 UTC (Sun) by rsidd (guest, #2582) [Link] (6 responses)

Apart from what you say, the FSF lists BSD/MIT/similar licences as free software. Copyleft has no role in RMS's or the FSF's definition of free software.

copyleft software, not "free" software

Posted Sep 27, 2020 17:05 UTC (Sun) by marcH (subscriber, #57642) [Link] (5 responses)

> Copyleft has no role in RMS's or the FSF's definition of free software.

... which is amazing considering he literally invented the entire "less-free-for-developers and more-free-for-users" software category, and probably the super smart "copyleft" word for it too. How come _he_ did not drop the super confusing "free" terminology? All his followers would have... followed, as they do when they use "free software" now, which helps feeding the never ending stream of debates where the basics seem to be re-explained over and over again.

copyleft software, not "free" software

Posted Sep 27, 2020 17:15 UTC (Sun) by rsidd (guest, #2582) [Link] (4 responses)

RMS's definition of "free software" is clear enough, and non-copyleft licences like BSD/MIT/Apache entirely qualify.

“Free software” means software that respects users' freedom and community. Roughly, it means that the users have the freedom to run, copy, distribute, study, change and improve the software. Thus, “free software” is a matter of liberty, not price. To understand the concept, you should think of “free” as in “free speech,” not as in “free beer”. We sometimes call it “libre software,” borrowing the French or Spanish word for “free” as in freedom, to show we do not mean the software is gratis.

Various comments on copyleft and non-copyleft licences are here. Note that objections to individual non-copyleft licences are not for their permissiveness but to unclear names (what does "BSD licence" or "MIT licence" mean?) or GPL incompatibility (eg the BSD licence with advertising clause; CDDL; APSL). The Apache 2.0 licence is strongly recommended as a permissive free software licence.

In short, this entire thread is a red herring, based on a misunderstanding of "free software".

copyleft software, not "free" software

Posted Sep 27, 2020 17:26 UTC (Sun) by marcH (subscriber, #57642) [Link]

> In short, this entire thread is a red herring, based on a misunderstanding of "free software".

Misunderstanding as in the gazillions other threads before it and in all the future ones too. I'm impressed by how quickly you found and posted the references, you must be really used to that :-)

copyleft software, not "free" software

Posted Sep 28, 2020 2:38 UTC (Mon) by jwarnica (subscriber, #27492) [Link]

> community

The problems is that RMS defines community as a) "the keybindings I personally enjoyed in 1975 long before industry and sanity discarded the concept of 'cording keyboard' " b) a particular non-technical and highly idealistic political philosophic and c) if yoiur a women, falling over yourself to get to me.

Which is, I grant, a *particular* community, but not one that is welcoming to pretty much anyone who has first used a computer since either the IBM PC or Apple MAC was introduced.

copyleft software, not "free" software

Posted Sep 29, 2020 10:19 UTC (Tue) by smcv (subscriber, #53363) [Link] (1 responses)

With hindsight, it's unfortunate that the originators of the free software term didn't talk about "software freedom" rather than "free software", which would have meant the same thing while avoiding the whole free beer/free speech ambiguity.

Regarding the difference between free software and copyleft: free software (or software freedom if you prefer) is a goal, but copyleft is one of several possible strategies for achieving that goal (permissive licensing is another). Copyleft is closely associated with the term "free software" because the FSF has chosen to use copyleft as their main strategy, but it isn't actually their objective.

copyleft software, not "free" software

Posted Oct 5, 2020 21:05 UTC (Mon) by marcH (subscriber, #57642) [Link]

> With hindsight, it's unfortunate that the originators of the free software term didn't talk about "software freedom" rather than "free software", which would have meant the same thing while avoiding the whole free beer/free speech ambiguity.

... and also avoided any copyleft/permissive ambiguity - thanks!

BTW "Software Freedom Conservancy" is an exceptionally clear name.

While their top goal seems to be copyleft enforcement, funny enough they don't seem to request copyright assignment before sending a patch. Their website also seems less focused on philosophy and more on getting copyleft things done but maybe that's just my impression.

copyleft software, not "free" software

Posted Oct 1, 2020 19:36 UTC (Thu) by ecree (guest, #95790) [Link]

> not sure yet how to best name BSD-like licenses.

I believe the canonical term is 'copycenter'. See http://www.catb.org/jargon/html/C/copycenter.html