Another important anniversary. I wonder how well this thread will
compare with "25 years of Google" [0]
As a 50 year old, GNU has been there in the background for most of my life
as a computing professional. It has not grown much as an organisation or
ideas base, but clung firmly to timeless principles.
That continuity and longevity is testimony to the power of an ideal,
that computing belongs in the hands of the people, as a tool and not
as a weapon in the hands of those who seek domination.
Except that GNU anniversary page had been around for months, and posted a few times already (and ignored mostly on here), while the Google 25 content is fresh today, if comparing news response.
I can't find it, but I remember a quote something like this
> The initial GNU announcement was crazy. Rewriting all of Unix? It was like Stallman had said, "I need help building a free rocket ship to Mars, and I'm getting started right after I finish this Thanksgiving dinner."
I hope to one day be known for 1 really crazy idea that somehow paid off, and not the 999 obviously bad ideas that it took to get there.
GNU has much to be proud of. It pioneered the use of "copyleft" and has produced a stream of influential software that is critical even today.
For me the greatest achievement of GNU is proving that software does not need to be proprietary. There was a time when Windows and other proprietary Unixes ruled the roost. Due to the easy availability of fundamental GNU software at the right time we saw the cambrian explosion of open computing software during the internet era.
However, GNU seems to have not changed (enough) with the times. A new breed of open source software is steadily chipping away at GNU's monopoly (e.g. LLVM, musl etc). GNU's website, tooling and codebase increasingly looks outdated, monolithic and crusty. Sure there are gems still to be found in GNU's software stable. I also like that they hold fast to certain noble software principles.
TL;DR GNU needs to reinvent and refresh itself while still preserving its core principles.
> However, GNU seems to have not changed (enough) with the times. A new breed of open source software is steadily chipping away at GNU's monopoly (e.g. LLVM, musl etc).
I don't think "losing their monopoly on free software" is something that the FSF or GNU are all that worried about. Their main concern is ideological/social/political - that free software is created, exists, and that the licensing rules around it are followed. If every piece of software they ever wrote got superseded by another piece of free software that was better I think they would be overjoyed.
I see suckless[0] tools as the spiritual successor to GNU. Free, open source (and I think copyleft?), minimal, and encapsulates a comprehensive desktop experience.
Call me crazy but my idea of a comprehensive desktop experience doesn’t involve recompiling tools like my terminal emulator or WM to change configuration.
‘No-bloat’ is kind of antithetical to GNU, where the various applications give you everything and the kitchen sink with as many compilation flags as you like to rip out features if you really don’t want them. But one person’s bloat is another person’s fully featured system; I don’t think projects should be opinionated about ‘bloat’ if they don’t want to alienate some section of potential users.
I can't help but think that the idea of copyleft was already pioneered and applied during the same time in the Soviet block of countries, where virtually copyright did not exist. Of course without all the fanfares...
So it was all copyleft, right? Definitely can remember us look with "awe and wonder" at Lawrence Lessig when he first visited one university in one ex-soviet country to lecture about the value of sharing software to a room full of people that have never ever in their life bought any.
The key thing about "copyleft" is that it uses copyright mechanisms to ensure the code stays free (as in freedom). This is different than disowning/ignoring copyright, and more of an inversion of how proprietary software works.
Pirating software doesn't help too much when you want to study/modify it.
The core is rotting probably, but there's a lot of other software than the compiler and OS core facilities under GNU which is pretty much thriving, so it's hard to judge and do something about getting back to the drawing board. Given also that Linux is not GNU and that there's a lot of alternatives to GCC and GNU *utils programs.
We can't take GNU without FSF: we can't think of GNU as a corporate brand to embrace as an identity that has to "win" an arms race.
As a 50 year old, GNU has been there in the background for most of my life as a computing professional. It has not grown much as an organisation or ideas base, but clung firmly to timeless principles.
That continuity and longevity is testimony to the power of an ideal, that computing belongs in the hands of the people, as a tool and not as a weapon in the hands of those who seek domination.
[0] https://news.ycombinator.com/item?id=37669341