Interview with Patrick Volkerding
Linux Journal: First I would like to find out a little about you. How old are you?
Pat: I'm 27.
Linux Journal: What you do for work, school, etc?
Pat: For school, well.... I finally got my BS in computer science from Moorhead State University last Spring. On the 8 year plan, actually. I started out in computer engineering at Boston University in `85, did that for 2 years, and then took a year off before transferring into the CS program at MSU.
I just recently got a job for a San Francisco-based company making medical archiving systems controlled by Linux boxes. It's been pretty interesting work.
Linux Journal: I am looking for articles on the commercial use of Linux. I would be real interested in an article on what you are doing there.
What you do for fun?
Pat: I do all kinds of things for fun. Linux is my big fun project right now - gets pretty crazy sometimes trying to keep up with all of the development going on; for instance, last week the new C libraries, GCC, and kernel were all released within a couple days of each other. Luckily, I like keeping my machine current. Judging from the mail I get when things fall a bit behind, so does everyone else.
Hmmmmm.... other things? Well, I like to brew my own beer. That was my big hobby until Linux began to demand more of my time. It's a fun process. My favorite part is firing up my 140K BTU burner and boiling it up for an hour or so. The aroma it releases when you throw the hops into the rolling boil is magnificent! The Homebrew Digest and rec.crafts.brewing were my big Internet hangouts before I came across Linux.
I also love music, especially the Grateful Dead. I've gone to, oh... 75 or so of their shows. The summers of `87 and `88 I followed the band all over the US in my `67 Firebird convertible. I've also got lots of portable recording equipment and tape the shows whenever I can. The Dead let their fans do this, which is pretty unique in the music world.
I play guitar, too. I'm still waiting for Jerry [Garcia] to invite me up on stage. :^)
Linux Journal: Another DeadHead? I've only gone to 30 or 40 shows. My first show was in 1974 and I have around 100 tapes from shows.
Now, on to the real stuff.
When did you first start working with computers?
Pat: Way back in 1973 when I was just a kid, I went on a field trip with my class to the computer department at North Dakota State University. The room where they kept the machines totally amazed me - lots of big whirring machines with flashing lights all over the place and rows of those big drives with disk platters. One of the sysops showed me how to play Star Trek on a DecWriter teletype-style terminal. It was an instantaneous addiction.
At the time, though, there was no way I could get a home computer. I don't even think such things existed, so I started getting interested in electronics, which was more accessible. I'd build logic gates out of relays and things like that. When the first personal computers like the TRS-80, Apple ][, and Atari 400/800 came out, I became a fixture in many of the stores that sold them. I couldn't afford one, but the store owners would let me hang out and use their machines. I taught myself BASIC and would write little store demo programs to ensure I'd stay welcome there.
I got an Apple ][ Plus with a 300 baud AppleCat modem right around 1980, when it was the hot machine. I used Apple as my only computer right up until 1990. I even had a C compiler and Unix-like operating system for it. It was nothing even resembling Linux, though.
Linux Journal: When did you first start working with Linux? Pat: I first heard about Linux in late 1992 from a friend named Wes at a party in Fargo, North Dakota. I didn't download it right away, but when I needed to find a LISP interpreter for a project at school, I remembered seeing people mention clisp ran on Linux. So, I ended up downloading one of the versions of Peter MacDonald's SLS distribution. Linux Journal: Describe what Slackware is?
Pat: Well, I guess I can assume we all know what Linux is. :^) Slackware consists of a basic Linux system (the kernel, shared libraries, and basic utilities), and a number of optional software packages such as the GNU C and C++ compilers, networking and mail handling software, and the X window system.
Linux Journal: Why did you decide to do a distribution?
Pat: That's a good one. I never really did decide to do a distribution. What happened was that my AI professor wanted me to show him how to install Linux so that he could use it on his machine at home, and share it with some graduate students who were also doing a lot of work in LISP. So, we went into the PC lab and installed the SLS version of Linux.
Having dealt with Linux for a few weeks, I'd put together a pile of notes describing all the little things that needed to be fixed after the main installation was complete. After spending nearly as much time going through the list and reconfiguring whatever needed it as we had putting the software on the machine in the first place, my professor looked at me and said, “Is there some way we can fix the install disks so that new machines will have these fixes right away?”. That was the start of the project. I changed parts of the original SLS installation scripts, fixing some bugs and adding a feature that installed important packages like the shared libraries and the kernel image automatically.
I also edited the description files on the installation disks to make them more informative. Most importantly, I went through the software packages, fixing any problems I found. Most of the packages worked perfectly well, but some needed help. The mail, networking, and uucp software had a number of incorrect file permissions that prevented it from functioning out of the box. Some applications would coredump without any explanation—for those I'd go out looking for source code on the net. SLS only came with source code for a small amount of the distribution, but often there would be new versions out anyway, so I'd grab the source for those and port them over. When I started on the task, I think the Linux kernel was at around 0.98pl4 (someone else may remember that better than I do...), and I put together improved SLS releases for my professor through version 0.99pl9. By this time I'd gotten ahead of SLS on maybe half of the packages in the distribution, and had done some reconfiguration on most of the remaining half. I'd done some coding myself to fix long-standing problems like a finger bug that would say users had `Never logged in' whenever they weren't online. The difference between SLS and Slackware was starting to be more than just cosmetic.
In May, or maybe as late as June of `93, I'd brought my own distribution up to the 4.4.1 C libraries and Linux kernel 0.99pl11A. This brought significant improvements to the networking and really seemed to stabilize the system. My friends at MSU thought it was great and urged me to put it up for FTP. I thought for sure SLS would be putting out a new version that included these things soon enough, so I held off for a few weeks. During this time I saw a lot of people asking on the net when there would be a release that included some of these new things, so I made a post entitled “Anyone want an SLS-like 0.99pl11A system?” I got a tremendous response to the post.
After talking with the local sysadmin at MSU, I got permission to open an anonymous FTP server on one of the machines - an old 3b2. I made an announcement and watched with horror as multitudes of FTP connections crashed the 3b2 over, and over, and over. Those who did get copies of the 1.00 Slackware release did say some nice things about it on the net. My archive space problems didn't last long, either. Some people associated with Walnut Creek CDROM (and ironically enough, members of the 386BSD core group) offered me the current archive space on ftp.cdrom.com.
Linux Journal: Why did you call it Slackware?
Pat: My friend J.R. “Bob” Dobbs suggested it. ;^) Although I've seen people say that it carries negative connotations, I've grown to like the name. It's what I started calling it back when it was really just a hacked version of SLS and I had no intention of putting it up for public retrieval. When I finally did put it up for FTP, I kept the name. I think I named it “Slackware” because I didn't want people to take it all that seriously at first.
It's a big responsibility setting up software for possibly thousands of people to use (and find bugs in). Besides, I think it sounds better than “Microsoft”, don't you?
Linux Journal: Some of the people out here in Seattle call them MicroSquish. :-) I admit that I initially avoided going from SLS to Slackware because I didn't take the name seriously. But the feedback I heard on the Internet pointed out why I should take it seriously. What did you expect to happen with the distribution?
Pat: I never planned for it to last as long as it has. I thought Peter MacDonald (of SLS) would take a look at what I was doing and would fix the problems with SLS. Instead, he claimed distribution rights on the Slackware install scripts since they were derived from ones included in SLS. I was allowed to keep what I had up for FTP, but told Peter I wouldn't make other changes to Slackware until I'd written new installation scripts to replace the ones that came from SLS. I wrote the new scripts, and after putting that much work into things I wasn't going to give up. I did everything I could to make Slackware the distribution of choice, integrating new software and upgrades into the release as fast as they came out. It's a lot of work, and sometimes I wonder how long I can go on for.
Linux Journal: What sort of help have you received?
Pat: Most recently, Savio Lam wrote the dialog program used to create the color installation menus on the latest release of Slackware (1.1.2). Ian Kluft put the smail package together for me. Vince Shakan's newspak collection of configuration scripts were very useful for compiling applications like elm and Taylor UUCP. Louis LaBash just contributed a kit to compile a version of perl with working IPC. Early on, Allen Gwinn sent me the lpd package. All the users that send me bug reports helped a lot, too. I especially like when they don't flame me too severely in the process.
Other than a few key packages from development teams that I trust, like GNU GCC from H.J. Lu, or the XFree86 2.0 package, I compile nearly all of the software myself. There are still a few remaining bits of SLS in the package - you can spot them by looking for files with 1992 time stamps.
Linux Journal: Do you have any idea of the number of copies of Slackware that might be in use today?
Pat: Thousands, but it would be hard to estimate an exact count. I have no idea how many CDs have been pressed with Slackware on them, but I can think of four companies that have produced them. Harald T. Alvestrand, who is attempting to count Linux users, posted this estimate: I don't claim that these numbers are in any way unbiased. They are just the 240 machine descriptions that have arrived at the counter. Obviously, they give a slight bias towards net-oriented releases.
Of the seven “unknown” ones, two were misspellings of slackware. There are biases in the summing too; one who listed “slackware, sls” got only the “slackware” part counted. At least, it is a datum.
(Now for some wild speculation: If this is average, and 8000 LGX CDs have been sold, and 9 of them turn up here, that would mean that I have a return rate of 0.11 percent on LGX users. A similar return rate on the others would mean 218,000 machines in the Linux installed base. My statistics professor would flunk me for extrapolation.....)
Linux Journal: Does Slackware have a future?
Pat: I would like to think so. I really enjoy working with Linux, and have had a blast making a complete package like Slackware available and easy enough for beginners to install. Ian Murdock (of the Debian distribution) and I have tossed around the idea of a merger since last fall. It's possible that this could eventually happen.
At this point, I've got some nice scripts that create packages, including the installation scripts that create the symbolic links. Other than answering my mail, it's not that hard to keep Slackware up to date. If I'm going to bother to keep my own machine current, I might as well be updating the packages on the FTP site at the same time.
Linux Journal: Do you want/expect any commercial benefit from Slackware?
Pat: I haven't accepted any so far. It would be nice to make money as a result of it, but not from selling the actual package. I'm not interested in going into the mail-order CD-ROM business or anything like that, but my experience with Linux has taught me a lot of valuable skills. It looks like the project has saved me from a life of COBOL. What more could I ask for than that?
Linux Journal: Thanks for doing this. I think the readers are going to be real interested in this monthly interview column and I really want to get people from all walks of the Linux world interviewed.
|Designing Electronics with Linux||May 22, 2013|
|Dynamic DNS—an Object Lesson in Problem Solving||May 21, 2013|
|Using Salt Stack and Vagrant for Drupal Development||May 20, 2013|
|Making Linux and Android Get Along (It's Not as Hard as It Sounds)||May 16, 2013|
|Drupal Is a Framework: Why Everyone Needs to Understand This||May 15, 2013|
|Home, My Backup Data Center||May 13, 2013|
- New Products
- Linux Systems Administrator
- Senior Perl Developer
- Technical Support Rep
- UX Designer
- Designing Electronics with Linux
- Dynamic DNS—an Object Lesson in Problem Solving
- Making Linux and Android Get Along (It's Not as Hard as It Sounds)
- Using Salt Stack and Vagrant for Drupal Development
- Reply to comment | Linux Journal
2 hours 56 min ago
- Reply to comment | Linux Journal
3 hours 13 min ago
- Favorite (and easily brute-forced) pw's
5 hours 4 min ago
- Have you tried Boxen? It's a
10 hours 56 min ago
- seo services in india
15 hours 27 min ago
- For KDE install kio-mtp
15 hours 28 min ago
- Evernote is much more...
17 hours 28 min ago
- Reply to comment | Linux Journal
1 day 2 hours ago
- Dynamic DNS
1 day 2 hours ago
- Reply to comment | Linux Journal
1 day 3 hours ago
Enter to Win an Adafruit Pi Cobbler Breakout Kit for Raspberry Pi
It's Raspberry Pi month at Linux Journal. Each week in May, Adafruit will be giving away a Pi-related prize to a lucky, randomly drawn LJ reader. Winners will be announced weekly.
Fill out the fields below to enter to win this week's prize-- a Pi Cobbler Breakout Kit for Raspberry Pi.
Congratulations to our winners so far:
- 5-8-13, Pi Starter Pack: Jack Davis
- 5-15-13, Pi Model B 512MB RAM: Patrick Dunn
- 5-21-13, Prototyping Pi Plate Kit: Philip Kirby
- Next winner announced on 5-27-13!
Free Webinar: Hadoop
How to Build an Optimal Hadoop Cluster to Store and Maintain Unlimited Amounts of Data Using Microservers
Realizing the promise of Apache® Hadoop® requires the effective deployment of compute, memory, storage and networking to achieve optimal results. With its flexibility and multitude of options, it is easy to over or under provision the server infrastructure, resulting in poor performance and high TCO. Join us for an in depth, technical discussion with industry experts from leading Hadoop and server companies who will provide insights into the key considerations for designing and deploying an optimal Hadoop cluster.
Some of key questions to be discussed are:
- What is the “typical” Hadoop cluster and what should be installed on the different machine types?
- Why should you consider the typical workload patterns when making your hardware decisions?
- Are all microservers created equal for Hadoop deployments?
- How do I plan for expansion if I require more compute, memory, storage or networking?