When did you first learn of this "Unix thing"?

[atax1a]

we read The Cuckoo's Egg in 6th grade

 

[mogabi]

In my peripheral vision, I heard of it a long time ago, but it was always one of those computer things that seemed outside my level of comprehension. Being brand new to FreeBSD, it still does. I feel like I need a book to help me understand what this "Unix thing" is.

 

[drhowarddrfine]

I feel like I need a book to help me understand what this "Unix thing" is.

Some people like "Absolute FreeBSD" but, if you find something to read, do it without a goal in mind. Don't read to learn how to install a desktop. Read it to learn everything. And read it slowly--without purpose beyond understanding what you read. No rush. No pressure. Try any examples and see what happens--or doesn't--and then find out why it didn't work the way you expected. Not because you need it. Because you want to learn. Then, one day, something will happen where you'll say, "Oh! I remember using that!".

 

[vassia]

Heard of FreeBSD as a Linux alternative.

 

[Tomolok]

As a relatively young member here,my journey with Unix started in around 2019 when I got my first PC for learning programming.Because of all the frustration with Windows(you know,all the advertisement,no package management,no convenient editors like vi/vim,hard-to-manipulate registration table(It was on Windows7)),I went to Ubuntu,and then very much enjoyed that you can basically freely do anything with your system.I was fascinated by how you just configure your system with plain text file,how it came with an out-of-box C compiler.But then I switched to Debian because of all the canonical nonsense.Months ago I started to give FreeBSD a try.So far it is satisfying,especially with the package management and high-quality documentation(I believe it is even better than the Debian one).I still like Debian and some other "old-schooled" distros like Slackware.But so far FreeBSD provide the best experiences if you disregard some compatibility with hardware.

 

[saken658]

1978 in college our computer center had a PDP-11/70 running RSTS/E (from DEC), but on weekends they ran UNIX, strangely enough for a project involving computer generated music.

My next encounter was 1992 downloading 386BSD, then the "patchkit" followed by FreeBSD 1.0.

 

[astyle]

.I was fascinated by how you just configure your system with plain text file,how it came with an out-of-box C compiler

Yeah, that was one of my initial Windows frustrations - compilers were hard to even come by. The very first compiler I have experience with is Borland C, which was insanely expensive at the time. Then I heard of Java. It was free, but it was not native-metal performance. Then Visual Studio was next on my radar screen. Insanely expensive and complex. Then Python came around - the Windows variant was free, but there was no good way to run Python programs on Windows, unless you had the special Python IDE (IDLE, it was called back in the day). But after I discovered Linux, that's when it was an explosion in the number of languages freely available! And then, with BSDs, it was the same thing.

 

[Tomolok]

Yeah, that was one of my initial Windows frustrations - compilers were hard to even come by. The very first compiler I have experience with is Borland C, which was insanely expensive at the time. Then I heard of Java. It was free, but it was not native-metal performance. Then Visual Studio was next on my radar screen. Insanely expensive and complex. Then Python came around - the Windows variant was free, but there was no good way to run Python programs on Windows, unless you had the special Python IDE (IDLE, it was called back in the day). But after I discovered Linux, that's when it was an explosion in the number of languages freely available! And then, with BSDs, it was the same thing.

I have a say in that python thing:
For some reason,Python installation on my Windows machine at that time was quite strange and frustration.I often found that somehow I cannot run python in cmd,or the .py files are not taken as Python source file at all.

 

[T-Aoki]

The very first compiler I have experience with is Borland C, which was insanely expensive at the time.

Borland C was quite cheap at the era, comparing with competitors.

Major proprietary compilers like Lattice C, Whitesmith C and Microsoft C were sold (if I recall correctly) JPY 200,000 to 300,000 per installation AND per user, while Borland C was around JPY 30,000 (former version, Turbo C sold via MSA was around JPY 20,000) per user that use it "at exactly the same time" (Philippe Kahn said the license "Just like a book") in Japan. Around 1/10. And the generated code was NOT 10 times slower. Almost same or a bit slower compared with MS-C (the champion at the era, as far as I know). Code size generated was not 10 times larger, smaller or a bit larger (almost the same).

At the era, single compiler license just allowed to install it into single specific computer, and if the computer is shared between multiple users, i.e., 5 users, 5 licenses were needed. And even if the exactly specific single user alone use the compiler, installing the compiler into multiple computers (rare at the era, though, as computers in Japan was expensive) i.e., 3 computers, 3 licenses were needed.

I've left Borland C/C++ just because they didn't release 32bit Japanese OS/2 version of it. So I've purchased a license of Watcom C for OS/2 (if I recall correctly, around JPY 40,000 or 50,000). But finally gave up as I couldn't adapt myself with Presentation Manager programmings that I actually want to do.

 

[Al Seitsinger]

I grew up with loose copies of the Anarchist Cookbook, the Redhat Linux bible, and AOL. I started on Redhat Linux (I convinced my Mom to buy me the CD's on eBay instead of downloading), along with some network equipment, and books back around 2000. Then, I learned how to write firewalls and host things. It wasn't until after that when I started with Slackware and FreeBSD.

 

[astyle]

Major proprietary compilers like Lattice C, Whitesmith C and Microsoft C were sold (if I recall correctly) JPY 200,000 to 300,000 per installation AND per user, while Borland C was around JPY 30,000 (former version, Turbo C sold via MSA was around JPY 20,000) per user that use it "at exactly the same time" (Philippe Kahn said the license "Just like a book") in Japan. Around 1/10. And the generated code was NOT 10 times slower. Almost same or a bit slower compared with MS-C (the champion at the era, as far as I know). Code size generated was not 10 times larger, smaller or a bit larger (almost the same).

If you read my comment a bit more carefully, you'd notice that I said that Java (not Python) was not exactly fast.

 

[MG]

Trying to remember but not sure what was the first time. Some guys at school were spreading an early Slackware on 10 floppy disks, combined with Crazybytes and Twilight cdroms . I think it initially didn't became really succesful because the disk partitioning was very complicated and Microsoft killed everything at a Windows reinstall. I never heard of System V-like UNIX back then. The local library had a VMS system with dumb terminals as book search consoles for visitors. That was absolute old and obscure to me too.

 

[Bucky]

In 2000 I was wanting to run an email server at home. Spent months trying to find a reliable way to do it with Windows 98 - failed 100% of the time; Windows 98 simply wouldn't run continuously. A friend who became my mentor said to use FreeBSD and toss Windows and I've never looked back though I didn't get a working email server (qmail) up/running until 2002.

 

[Crivens]

I first stumbled over Minix, could not get it working and was angry. Later, in university, I had access to some VMS, then the CS department started to roll out guest accounts. I immediately got one (ID was below 400!), access to some RS6000 shared with a lot of other students. With 3-5MB of quota, totally unthinkable today. I first ran NetBSD from a mixed mode CD, which held install images for several architectures at once. Being a total dunce at that, I found the source packages and started to do the job of "make makefiles" by hand, fixing up all the package makefiles untill I was able to play xpilot. Dumb on that part, made me learn a lot about the details. Today I'm running 15.1-Stable, the kids leave not enough time to do more. Once they have left, maybe, I can finally build my complete own system. Maybe before that.

 

[ralphbsz]

Borland C was quite cheap at the era, comparing with competitors.

Major proprietary compilers like Lattice C, Whitesmith C and Microsoft C were sold (if I recall correctly) JPY 200,000 to 300,000 per installation AND per user, while Borland C was around JPY 30,000 ...

That was a major culture difference back then. Compilers for commercial operating systems (Windows, Digital VMS, IBM MVS, ...) all cost money. On Windows, they were affordable. On big computers (where the computer itself cost 100K$ to M$) they were expensive. But they came with excellent support ... I had several phone calls with DEC's Fortran compiler group when I was a user, so much that they offered me a job when I was interviewing after getting my degree.

In contrast, the Unix culture was that at least the C compiler was always free. I think even when Unix had to be licensed for money (AT&T) or was only sold with the hardware (most others), that compiler was always included. One of the reasons was that the C compiler was absolutely needed to do install and customization of the system itself. Remember the outcry when Sun tried to change this, and the C compiler became an extra cost option; to some extent, that was the genesis of the Gnu compiler: before it was a hobby project by some crazy political activists (RMS and his minions), and then it suddenly became a very useful tool for lots of Sun machines. That forced Gnu to become better (and attracted more contributors and maintainers), which then directly led to the FOSS movement and Linux.

But even on Unix some compilers were very expensive. I remember my wife spending a lot of money on the Portland Group compilers for her cluster of Silly Graphics machines. It was a good investment, since it made the code run much faster, meaning fewer computers had to be bought, operated and administered.

At one point, Digital executives considered the idea to even make the run-time libraries for specific languages be not free. That would have meant that a program compiled on machine A would not necessarily run on machine B, unless B's owner pays extra for the specific language support. The developers hated that so much that they decided that the base OS would contain at least one utility written in every language. Even obscure and not very useful languages (such as RPG-II).