[aurelian15]

As weird as it may seem, you should not forget that free software licenses are built upon the fabric of copyright. Without copyright, free software could not exist in its current form. For GPL-like "copyleft" licenses, there would be no way to enforce that binary distributions of derived works are accompanied by their source code. Similarly, in the context of permissive BSD/MIT-style licenses, there would be no way to enforce attribution.

So, given that FOSS---which a large portion of the HN crowd depends on---cannot work without copyright (at least not in its current form), the recent discussions may be less of a surprise.

[cortesoft]

Maybe... although I personally think that the GPL and other 'copy left' licenses aren't the reason open source has prospered, nor do I think enforcing attribution really helps the FOSS world that much.

People write and share code because it is useful to do that, not because licenses require them to.

I think FOSS would do fine with no copyright, and in fact more software might end up open source if we had ZERO copyright... why not make your code open source and get back contributions when your code would end up being shared anyway?

[mcguire]

You may think that, but history argues otherwise. See the Unix Wars (https://en.wikipedia.org/wiki/Unix_wars).

[cortesoft]

I am not sure I follow. Are you saying licenses are what stopped the UNIX wars?

[ripdog]

IMO Linux won because its license forced everyone who used and extended it to open their changes - this changes the calculus for firms building products on Linux to make it more worthwhile to upstream their changes to reduce their local maintenance burden.

Compare this to something like the playstation 4, which used freeBSD as the base of their OS, and contributed nothing back to the project at all.

[Jasper_]

Linux "won" because SGI very loudly shit the bed and betting on Itanium, taking IRIX with it, DEC was in the process of making the future of Tru64 very confusing with the Alpha, a computer that didn't support their flagship OS, and HP was EOLing HP/UX because they bet everything on IBM's OS/2 Warp.

The only big UNIX vendor left was Sun Microsystems, and Solaris indeed dominated the 90s dotcom era. Everybody was running SPARC and SunOS servers.

It wouldn't be until the mid 2000s when Linux started picking up the pieces left behind after Red Hat started their server product and certification program.

For a long time, Linux was strictly a hobbyist OS. It later dominated by simply being the last one standing after everyone else fell.

[mcguire]

The true competitive threat to the Unix vendors wasn't each other, it was Microsoft.

And "for a long time" was actually a fairly short time. Linux began to approach feature-comparability fast, and ran on PCs, not $10,000 workstations (that were getting beat power-wise by PCs).

[cortesoft]

This seems a bit of a chicken and egg problem... why would the early adopters have chosen Linux, before others had been forced to contribute back their changes? The first company to adopt it wouldn't have received any benefits, only an obligation. Why pick it over BSD?

I think there were likely other factors that made it win out.

[ripdog]

Oh, for sure. There are always other factors in the real world.

But I think that's the main one that allowed Linux to snowball ahead of the other Unix(-likes).

[ric2b]

They would also get the "promise" that the system they were betting on would also get contributions from other companies, making a safer long term bet.

[mcguire]

And because the GPL makes it difficult to fork and make the branch a proprietary product.

[mcguire]

No. AT&T created Unix but was unable to market it due to a previous antitrust action. So they gave it away. (They required a license signature (I've signed that! :-)), but did not charge and were very lenient.)

The UCBerkeley Computer Science Research Group (if I've got my acronyms right) was one of the recipients and went on to add on many, many features and releasing the result under the BSD license.

Many people built companies like Sun around selling BSD Unix, including many alumni of UCB.

Then AT&T got out from under the consent agreement and began selling its own Unix, System V. By this time, Unix was a major player in the workstation market (a market that has largely disappeared as PCs got more powerful).

By, say, the mid 1980s, there were many, many companies selling many, many varieties of Unix, all descended from the original Unix via BSD or via BSD+System V. Most of them had some unique, valuable features (Irix's graphics, AIX's LVM and journaling file system, etc.) and all of them had modifications to lock customers into their version. This is where the POSIX standards come from (second only to ecmascript market manipulation goofiness), and why things like Autoconf/automake and the much-loved imake (not really) exist. There was much in-fighting; Sun vs everybody else, everybody else vs. IBM, etc.

Then two things happened: PCs got more powerful and began eating into the bottom of the workstation market (PCs ran DOS+Windows, which was and arguably still is, technically inferior to multi-user-by-design systems like Unix[1].) And PCs got more powerful and began to be able to run more advanced OSs (think "memory management").

At this point, the Unix world began to conflict with the Windows world. Unix was technically superior, Windows had more public and developer mind-share. But the Unix world was still more interested in fighting each other and stapled all of their arms and legs to that particular tree.

The end result was that Windows became and remains the most-used operating system[2]. All (almost) of the commercial Unixs died (almost; there's still some animated corpses around)[3]. The two counter-examples are MacOS, which is completely locked to Apple hardware, and Linux.

Linux is the interesting case. Windows and commercial Unix all had a 15- to 20-year head start. But Linux achieved (mostly) feature parity quickly and did not break down into multiple, competing streams. Both of those are due to the GPL; you can fork GPL software all you want, but you cannot add a feature to a fork and expect it not to be back-ported into the original if it's useful. You also have a very hard time locking users into your fork.

The bottom line is that Microsoft won the Unix wars, because the Unix licenses allowed companies to take Unix proprietary.

[1] Modern Windows is kinda-sorta based on VMS, another workstation OS, but not really and then they walked that back, and so on....

[2] I don't really consider Android or iOS to be general-purpose OSs. And they're both rather their own little islands, no matter how much the underlying tech shares with the rest of the universe.

[3] The Free/Net/Open/DragonFly BSDs are, I'm sorry to say, noise. And did you notice that I had to mention four of them?

[em-bee]

people do, but companies don't.

projects like openwrt would not exist if it weren't for copyright enforcement.

[dublin]

There were other open source licenses at the party before the GPL dropped its controversial "viral" turd in the punchbowl - and many of them still exist nearly unmodified. (e.g. BSD with attribution removed, etc.)

[cortesoft]

I know, but I am just questioning whether any license is needed for open source to prosper.

I am positing that if licenses didn't exist, and anyone could do anything they want with any bit of code they see, open source would still prosper.

[okamiueru]

That just isn't how things work. I believe you are making the wrong assumption that the same amount of open source work would exist. If that were the case, then yes, it wouldn't matter that much. But, a lot of contributions to open source wouldn't have existed if it wasn't for the licenses. That's the examples you've been given.

[cortesoft]

There might be some contributions that wouldn't have happened, but there also might have been others that did happen. My hypothetical was in a world with no copyright or license at all.... so all proprietary code would be copyable, too

[okamiueru]

Proprietary code would by definition not be copyable, because it would be... proprietary. It's the opposite to open source. The thing that incentivises open sourcing proprietary source code is exactly things like licenses... Your imagined scenario is just "no license for open source, and free reign for closed source". It makes no sense to me.