SoylentNews is people
SoylentNews
posted by
hubie
on Tuesday September 30, @02:39PM
from the for-decades-for-freedoms-for-all-users dept.
from the for-decades-for-freedoms-for-all-users dept.
The Free Software Foundation (FSF) turns forty on October 4, 2025. The Free Software Foundation will have then been defending the rights of all software users for the past 40 years. The long term goal is for all users have the freedom to run, edit, contribute to, and share software.
There will be an online event, with an in-person option for those that can get to Boston. In November there will also be a hackathon.
This discussion was created by hubie (1068) for logged-in users only, but now has been archived.
No new comments can be posted.
Forty Years of Commitment to Software Freedom
|
Log In/Create an Account
| Top
| 7 comments
| Search Discussion
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
(1)
Lady Luck brings added income today. Lady friend takes it away tonight.
(Score: 2, Interesting) by VLM on Tuesday September 30, @04:09PM (5 children)
Here's a couple (possibly good) ideas:
This would have been an interesting poll idea: Are you older or younger than the FSF?
Another idea: Whats the story of the first FSF software you used?
My story is late 80s maybe 1990 or so, I read quite a marketing rant about the miracle of postscript as a language (was that Don Lancaster in one of the electronics magazines or someone else?) and found ghostscript for msdos with a printer driver working on my dot matrix printer so I downloaded it and ... it worked, perfectly, but the overall system was so slow that turnaround from submit postscript code then see the printed result could be measured in hours, so my motivation rapidly waned. For the LOLs I tried to find a link from ghostscript's official websites and could not find a compiled copy of GS 1.something or 2.something for msdos. I thought it would be funny to run it in a dos emulator, see if GS runs under dosbox or an 86box VM running freedos or similar. Alas... and I have things to do today, but someday it would be fun.
I seem to recall obtaining a very early GCC for dos not long after it was ported and not being terribly happy with it as I was used to microware's much simpler OS-9 compiler and used to Turbo C under dos so having a port that was so complicated you could only run it if you were the porter himself was less than inspirational. I honestly don't remember if I ever got GCC working back in the late 1980s, not even for "hello world". IIRC the port was very memory hungry back when a meg cost like $100 and dos didn't do virtual memory. Linux did, so thats why I installed it. Better to compile slowly than not at all LOL.
I did enjoy one of the emacs ports or emulators or workalikes or maybe the official thing on dos shortly after. My memory of that is fuzzy because "back in the day" there were a LOT of unofficial not-really-emacs but identical keystrokes. Same for vi, you couldn't get vi but there was vi inspired editors. I distinctly remember using microemacs which somehow morphed into pico which had "issues" (don't recall) leading to ... yup ... nano. That nano. Uh huh. Microemacs goes back to at least the mid 80s.
And not long after that I heard about this "softlanding linux solutions" set of floppy disks to install linux... off to the races that was a lot of GNU software with a linux kernel, so calling it GNU-Linux was reasonable back in the day.
(Score: 3, Interesting) by VLM on Tuesday September 30, @04:19PM
Well the instant I posted that I recalled a distant memory, haven't thought of this since the late 80s. The electronics magazine EE who ranted about the miracle of postscript for all things EE had inspired me on this path so I obtained a post script smith chart.
I suppose you'd describe a smith chart to a non-EE as kind of a printed slide rule for microwave EE stuff. Technically its a printed nomogram or nomograph but the only people who know what that is, don't need to be told LOL. Tell normies something like "you calculate magic radio things by drawing lines with rulers and following curves on the graphs". You can, like, draw all the parts in a power amplifier and magically determine the matching network to make it work. Sometimes it even worked, rarely it even worked on the first try LOL.
Note that smith charts are copyrighted so I found it amusing to download a presumably free postscript script to generate a copyrighted image using free software, all downloaded from some rando 80s BBS that was mostly full of pirated games as they all were back in the day on my technically pirated copy of msdos that I surely never paid for (presumably some previous owner did?)
Anyway it was a memory hog and promptly crashed, ran out of memory. Rather annoyed. So no it did not work perfectly.
Recursion is elegant until it smashes the stack LOL and sometimes it takes a lot less than you'd think. Also you cant cheat the real world, if you want to render multi megapixel images like a full page graphic of a smith chart, you can't do that very well with 640K which should be enough for anyone.
(Score: 4, Interesting) by bzipitidoo on Tuesday September 30, @05:25PM (1 child)
Before Linux existed, a class I took as an undergrad had us working on Minix. I didn't like it. Thought it was super lame that we had to use this "split" utility to break a source code file into small enough pieces so that the text editor could handle it without running out of memory. In the early 1990s, I had several flavors of DOS, I tried OS/2 (3.0 Warp), and Linux when it was on version 1.3. What broke things for me permanently as far as commercial software was concerned was OS/2 not having networking included, and the terrible bugs in the Borland C++ compiler. Any program that tried to use more than 64K, Borland C++ would screw up by assigning the same 64K segment to what should have been different segments. Massive memory corruption that the OS couldn't see. I switched to gcc, on Linux, and never looked back.
(Score: 2) by turgid on Tuesday September 30, @08:59PM
I didn't like it. Thought it was super lame that we had to use this "split" utility to break a source code file into small enough pieces so that the text editor could handle it without running out of memory.
That was because of the 64k segments on the 16-bit PeeCee architecture? In those days, the 386 was very expensive and apparently coding for it was "very hard" mainly, I think, because people preferred to keep working code secret rather than publish and share it, and page tables were fancy mainframe things.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 3, Interesting) by crm114 on Tuesday September 30, @06:11PM
> This would have been an interesting poll idea: Are you older or younger than the FSF?
> Another idea: Whats the story of the first FSF software you used?
Great ideas!
For me, I'm older than FSF, and have been in software development longer than FSF
However
My intro to FSF was around 1993/94 - when I installed the SLS Linux distro [to build an SMTP mail server, to connect all our PMAIL clients on different Novell network spokes together. Remember those days?!].
SLS included a bunch of FSF utilities (including, as another poster mentioned, something that could generate PDFs, for the office LaserJet)
It was also my introduction to IPv4 and TCP/IP.
(Score: 4, Interesting) by turgid on Tuesday September 30, @08:55PM
I think for me the first significant piece of GNU code I used was DJGPP, the port of gcc to MS-DOS. It was truly awesome for the time. It was 32-bit and came with a DOS extender.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 0) by Anonymous Coward on Tuesday September 30, @06:48PM
I'm old enough to remember this
https://www.computerhistory.org/collections/catalog/102782482/ [computerhistory.org]