Looking back on 30 years of Linux history with Red Hat's Stephen Smoogen

The Linux kernel and the second version of the GNU General Public License (GPLv2) turned 30 this year. As part of that major milestone we asked Red Hatters who had been using or contributing to Linux since the early days about their experiences. What was it like contributing to Linux, what was it like using it? Could you imagine that Linux would have the impact it's had on the world up until now?

Q: What was your first encounter with Linux? When did you start contributing to the project?

A: My first encounter with Linux was in early 1992, when I was working on a digitized astronomy project. We had been struggling with getting moved from a 1972 miniframe to newer hardware. At first we tried an RTOS, but a professor said: ‘Hey, try this other OS and see if you can get it to do what you want’. Where I had been stuck with getting basic email compiled, with Linux I had a fully operational ‘Unix-like’ system working after 20 floppies. 

Q: Do you remember what your first contribution was? What was the experience like?

A: I think some of my first Linux related contributions were getting Softlanding Linux System (SLS) and a kernel working with a GPIO board in late 1992. Most of my work was in the form of trying some code, break the box, ask some questions, try another code change etc. 

Q: Back in the day it was possible to “let the magic smoke out” of various hardware components when installing Linux. Do you have any stories about blowing up monitors or otherwise bricking hardware while you were learning how to do all of this stuff?

A: All of my magic smoke stories are with other people. I had one 386 and no backups and a set of hardware which was failing. All of my changes were tested in triplicate somewhere before I implemented them. The biggest magic smoke I saw was in a college lab where there were 30 identical monitors and a new X was compiled for them. 

The person who put the new X on all the boxes had put in the monitor values from the line above in the X README file and then they rebooted all the boxes at once via a breaker box. The boxes started up and then the awful whine of tubes being driven past their values. And then bang bang bang. The breaker then went off, saving most of the monitors. 

Q: After you started contributing, why did you stick with it? What were you hoping Linux would become?

A: When I started with Linux, I had been using Unix for about 4 years and the cost for a workstation was around USD $16,000.00. I had tried DOS and Windows, but found the power of shell scripts, pipes and gcc tools to be too much to go back it. And Linux on my i386 allowed me to have all the tools and commands I was using on the Unix box for ~$1600.00. That was my primary driver starting out. Since then, I have found Linux to be a useful tool for most of the problems I have run into - which is probably why I have stuck with it so long.

Q: How long did you contribute as a volunteer? How did you come to work on open source full time?

A: Most of my contributions for Linux were minor until I started work at Red Hat in 1997. Before then, I had worked at proprietary companies which did not allow me to share much. Even talking to the kernel developers for early kernel requests took ‘lawyer time’ to make sure I wasn’t ‘tainting’ our products. In 1997, I got the opportunity via a mass layoff at my former company to work as a Technical Support technician for Red Hat. 

I remember asking permission to post to various public mailing lists and getting these stares of horror from my new coworkers. ‘You don’t have to get permission to post solutions or fixes. Heck, please do so I can spend more time on other things.’

Q: Over the years we’ve seen a huge variety of people contribute to Linux and open source in various ways, from high school dropouts through PhD candidates and beyond, across a range of disciplines. What was your background in that sense, and what led you to open source in the first place?

A: My original work in Linux started when I was in college working on an Astrophysics major. I had been an administrator for a 1972 Prime 300 which had been used to operate a remote telescope looking for supernovae , but that hardware was failing fast. My tasks had devolved into trying to get the 1960’s-era Fortran code and images off the system onto anything I could before it was all gone. 

At first I was using spare IBM PC 8088 systems, but we ended up getting an i386 which we started using an RTOS on, with the idea that it would replace the Prime. The software for that system was incredibly hard [to use] and expensive and even getting SLIP or an ethernet driver working was a chore. 

When the option of a free OS which worked out of the box happened it was like a miracle. In 2 weeks I had gotten the SLS system with a 0.99 kernel doing things that the other OS had not done after 6 months. I was able to get fixes and use newer open software and give feedback to people who listened. (I had tried a BSD before Linux as I was more familiar with that, but had run into the community at a bad time. So I ended up switching to SysV style administration.)

Q: Thinking back to those early days, did you ever imagine where Linux would be now? Did you ever imagine you’d still be contributing so many years later?

A: In all honesty, during the early years I thought Linux was going to be replaced by BSD as soon as the licensing and court cases got cleared up. However, each time I went back I found that the BSD community and myself didn’t talk the same language anymore. So I kept going back to Linux and finding myself able to contribute over and over again. 

Q: During those years, what has surprised you the most about the evolution of Linux?

A: That it kept going. Most of the other communities I have seen tend to reach some sort of plateau and then say "that's good enough." And while various Linux distributions have done so, and various subparts have tried to, there is always room for someone to say "you know, I think I can do this differently and/or better" and then do so. It might take a lot of chutzpah, luck and charisma, but the fact that we have thrown out multiple "[it's] stable so why fix it" things over the years has always surprised me when it got done. 

Q: Have you ever encountered any industry-specific challenges for Linux during your career -- for example, financial services or telcos -- or has adoption been similar across verticals?

A: In the 1990’s and early 2000’s, most of the challenges I had were license related. There was a general fear that the GPL infected everything it touched, so any system running it must all be GPL. That was pretty much consistent  from universities to government to software companies. That was more of the challenge than Linux running on ‘slower/smaller’ systems than the big Unix boxes. Usually once the lawyers who understood copyright and licensing issues had talked it through, it would get adoption for the simple reason that paying for 8 systems for the price of 1 large Unix box made it an affordable fit. 

Q: Now, looking forward -- how do you think Linux will continue to evolve? Where do you think it might be in the next 10 years? 30 years?

A: The evolution of Linux will depend on who comes in next. Linux adoption had the luck of a couple of things tied together.

• BSD for PC was in limbo.

• UNIX had made major headwind in universities in the late 1980’s, replacing much more expensive mainframe and miniframe computers. 

• Windows and MacOS were expensive for programmers.

• GNU and similar compiler software was free in freedom and price.

This allowed a large number of people to get actual computer time when they had no access before. I knew several graduate students in 1991 who had Computer Science degrees but had never seen a computer before. They had to use punch-cards and mail them to a central university and wait for the results to come back. So they were really good at self-checking code but only in small amounts.

The ability to actually have a computer in their house and do all that programming was immeasurable to them. (The size and complexity of programs that could be written have probably gone up for good or ill because of this also.)

We’ll have to see if future generations find that freedom useful and have the ability to make the changes they find useful to them.