Freely Redistributable Software is Alive and Well
I have just returned from spending four days at the Freely Redistributable Software Conference. It was held in Boston on February 2 thru 5 and sponsored by the Free Software Foundation. The first conference of its kind, it was a real treat to attend.
With less than 200 attendees, it wasn't a large conference but I feel it was very significant. For those of you who have attended Usenix in the early to mid-80s, this conference has much the same flavor. For those of you who didn't, Nerdcon might give you an idea of what I am talking about.
It is not that the attendees were all Nerds but they weren't marketing people or novices to computers.
Discussions tended to be serious and technical and the participants had the knowledge to back up their positions.
The conference began on Friday night with a reception. It was a good chance for attendees to meet each other. While I didn't see any important decisions being made, it was my first chance to get a feel for how significant the Linux influence would be.
Greg Wettstein of the Roger Maris Cancer Center came over to talk to me and said he hoped to meet Linus. I pointed out Linus who, along with his girlfriend, Tove, were at the next table. Greg went over, introduced himself and hung out with Linus for quite a while. Greg commented to me later on how impressed he was with Linus, for his work on the system of course, but also as a person. Greg was not the first or only person to make a very positive personal comment about Linus. Many, including myself, think that his personality, that is, his willingness to listen to other opinions—good or bad—is one of the major reasons Linux has become so popular. I think that his sense of humor is another important contribution.
Saturday offered a collection of tutorials including the two half-day sessions I presented on Linux—Linux:An Open System for Everyone and Installing and Running Linux. They were well attended (about 30 students) and there were definitely some new converts to Linux. While the tutorials were intended for Linux newcomers, there were a few serious Linux users in attendance. The general conclusion was that we had a good time and everyone learned something.
The auxiliary meeting location was the Cambridge Brewing Company (CBC), a brewpub with good food and good brew about two blocks away from the conference hotel. Linus and a large group of Linux-followers ended up there on Saturday night; ten of us including Dan Quinlan, Jon “Maddog” Hall, Erik Troan, Donnie Barnes, Eric Raymond ended up there for lunch on Sunday and another group including Greg Wettstein, Steve Imlach, Tom Sargent and Bryttan Bradley and myself ended up there on Sunday night.
While there were other important ideas discussed at CBC such as convincing 60 Minutes that they should do a segment on Linus and Linux, the quality of their stout was an important consideration. Personally I would call it a thumbs up.
This was the only day of open sessions. The first keynote (by Linus) and five of the ten sessions were on an aspect or use of Linux. The sidebar outlines all the sessions and tells you how you can get a copy of the proceedings.
Sunday started off with a keynote by Linus (after I had fun waking up the crowd and attempting to make Linus sound like an important mainstream executive).
Linus originally planned his keynote around being the Dr. Ruth Westheimer of Software. With a fake German accent he intended to ask questions like “Do you go blind if you program alone?” “Is it ok to date different operating systems?” “How do I know when I met the right OS?”
He then intended to continue by answering the questions. But, that didn't work out so he went on to his second idea, “ Software is almost but not quite totally like sex”. He decided it just didn't have the same zing to it so he went on to present a talk titled “Write Free Software, Travel the World and Meet People”.
He said “Rather than writing a great operating system I wrote a small, not so great operating and then made it free. That turned it into a great operating system.” Certainly the right attitude to take at the conference and, in practice, it seems like he is right.
I personally found great interest in Victor Yodaiken's talk about using Linux at New Mexico Institute of Mining and Technology. Briefly, what he has done is slipped a real-time kernel between Linux and the hardware. This had made it possible to handle hard real-time processing (this is where processes must run at a required time, not just “soon” and yet still have the utility and versatility of a general purpose operating system running on the same hardware. Victor has promised Linux Journal an article on his work.
There is an interesting comment in the conference paper on the Yugoslav Experience that helps support the idea that Linux is viable in an open market: “even though the price of pirated commercial Unix-like operating systems is comparable to the price of Linux, and with practically no legal limitations to using pirated copies, Linux is being more widely used for Internetting in both [the] academic community and companies.”
The next presentation was on Linux on the OSF Mach3 Microkernel. This fits in with Apple's announcement of Linux on the PowerMac. Let's just say this was a surprise. Michael addresses this subject in Stop the Presses.
Greg Wettstein presented a talk on his work with Linux at the Roger Maris Cancer Center. An article on his work appeared in Linux Journal issue 5. To condense a serious effort into a sentence, the cancer center is using over 30 Linux workstations running a custom-built patient care information system using Perl and Tcl/Tk.
In the following session, Erik Troan presented a talk on the package management system developed at Red Hat software. This package, RPM, makes in possible to easily update an existing Linux system. The bottom line is that Red Hat is encouraging vendors of Linux software to use their packaging method. This will make it easier to distribute and maintain software for Linux.
Monday I decided to play the role of student instead of presenter and attended Tom Christiansen's all-day Perl tutorial. At lunch we had a lively discussion of Linux, NetBSD, software licenses, is Emacs the answer to the world's problems and all those other important subjects.
During the afternoon I mentioned to Tom that I was glad I had proofed Arnold Robbin's (excellent) Awk book before attending the Perl session because of the confusion that ensues when I am trying to think in Awk and Perl at the same time. This led Arnold and Tom into a “which is better” discussion at afternoon break. No bloodshed and I think they are both right but it was interesting to listen to them. As I have been an Awk user for about ten years and am just starting to work with Perl, I intend to write an article or two for LJ on how I see the two languages fitting into the world.
While I found the conference extremely interesting and valuable I wanted to get feedback from others. Here is what a few attendees had to say.
Russell NelsonCrynwr Software, email@example.comCool stuff: Meeting people f2f (including Linux && girlfriend [Tove], Peter Deutsch, Tom Christiansen, Greg Wettstein and that fyl fellow. Badges with e-mail addresses on them, so you'd know who people are.
Uncool stuff: The price relative to the length. ... Badges printed with a too-small font, so I had to squint at people's chests to know who they were. ... I like to call it freed software. It's a little ungainly, but it carries more information than the confusing appellation “free”.
L. Peter DeutschAlladin Enterprises, firstname.lastname@example.orgCool stuff: the two keynote talks. I found Stallman's quite inspiring in terms of getting a more spiritual/emotional understanding of the “free” software concept. Meeting people face to face. Having the opportunity to present my own twist on “free” software to a lot of people who didn't entirely agree with it. The gamelan concert. The free Linux CD-ROMs.
Uncool stuff: I second Russell [Nelson]'s comment on price versus length. If I hadn't been presenting a paper, I probably wouldn't have come. The small attendance, presumably arising from the short notice.
(Russ and Peter's comments continued from each other. Russ commented that he thinks the FPL (which is what his paper was on) encourages free software even more than the GPL. He points out that products live and die by market share.)
Steve ImlachSynergy, email@example.comMy eyes have been opened to some great opportunities in our area using Linux on our hardware and some contributions that we may provide to the continuing Linux development.
First, I thought it was a good sign to see in the Boston Globe Business section about Apple and the Linux server on top of the Mach kernel. This is some good stuff. Synergy is already at work with the Mach 3 kernel and would be very interested in getting more information in this area.
Since we are manufacturers of high performance Single Board Computers, we are interested in first, porting Linux and second, helping with the real-time development. Synergy is also highly interested because it would not only help us in performance but also improve the bottom line to our customers.
The ideas keep coming. Linux can also provide a high performance testing environment for our boards because we would have the source to tweak in any areas we deem to be a snappy enhancement.
It's just a matter of getting started. I see a lot of full weekends ahead. Luckily, unlike our FSF friends, I have a job that supports this development in an ugly capitalistic evil manner. You know, it's the profit mongers like Synergy that donate to the FSF year after year because we haven't learned to steal what is free yet.
Ok, that was a little jab at RMS. His speech was a little tainted, I thought. His story about the “community” in college reminded me of how I felt when I played ball at school. I was sad it was ending and about the camaraderie I felt should never end. The thing is, you move on. That is not to say, however, that the same type of teamwork couldn't happen at the workplace. He shouldn't knock it. Has he tried it? I think he is exactly right when he says he doesn't like the competition. There is plenty of it out here in the real world. What could he do if he was pushed via that competition. Too bad. FSF stuff is used everywhere and has great admirers here at Synergy as reflected by the annual donations. I believe he should balance his remarks to reflect what is really allowing the FSF to survive. Cooperation with the business community. He seems to affect a lot of views in the FSF “community”.
Linus was the type of guy I would like to work with. He certainly shows it in the product and the way he handles himself while juggling everyone's ideas, good and bad. It's no wonder that he is such a popular fellow. I hope I may be factor in the coming days along with other engineers at Synergy that are more experienced than I in the advancement of real-time Linux. First, there will be the Linux learning curve as we delve through the source ...
Dr. Greg was inspirational and a great new-found friend. We talked about just about everything and he is just the type a person that you would want to help up at Roger Maris cancer center in N.D. What was really amazing was watching him and Tom Sargent (Synergy-Tucson) talk forever about everything from Cancer research to spelunking to circus maneuvers in vintage aircraft to man-eating viruses found in the depths of Africa. These guys need to have a talk show on cable.
P.S. Linus Torvalds deserves great credit for the revolution of free OS. What is his status now and how does he get support for the continuing effort? I am also forwarding this to Dr. Greg. I would like to see how his efforts are paying off up there in what would otherwise be a drain on the spirit of a mere mortal.
Dr. G.W. WettsteinOncology Research Div. Computing FacilityRoger Maris Cancer Center, firstname.lastname@example.org I enjoyed the opportunity we had to visit at the conference (and the beer). I think the Conference on Freely Redistributable Software was important for its implications about the entry of free software into commercial venues. This was probably demonstrated most appropriately by a sampling of the selected papers. Of particular interest was the paper from the Apple/OSF consortium as well as the papers from the group working through SBIR and Peter Deutsch's thoughts on marketing. Hopefully my discussion of our accomplishments at the RMCC were helpful in this area as well.
The impact of redistributable sources on the commercial arena was also underscored by the presence of commercial enterprises at the conference. Of notable interest was the presence of Apple and representatives from HP and Synergy Systems. Apple's presence was extremely important with their announcement that Linux would assume a role in their strategic presence and would be marketed as an Apple offering. Who would have ever thought that an organization such as Apple, which considers software as strategically central to their business plan, would consider free software as a viable marketing tool.
What I found extremely interesting were discussions with Steve Imlach and Tom Sargent from Synergy Systems. Their interest in Linux is specifically for the strategic advantage that it will confer on their hardware marketing efforts. Their group feels that Linux represents a high quality solution which will allow them to more effectively penetrate markets due to the cost advantage it imparts to their hardware platform.
The implications of this are extremely important for the freely distributable software movement. Steve Imlach indicated that the leadership of Synergy is committed to the philosophy of free software. This commitment includes monetary donations to the effort as well as a commitment to release sources back into the Linux kernel development stream. In my mind this is representative of the wonderful synergy that can and should develop between commercial enterprise and the free software movement.
In business parlance this is a win-win situation for everyone involved. Commercial enterprises gain a strategic advantage and the free software movement reaps monetary advantage. An even more intrinsic advantage and one that must be carefully considered is that the software remains free. Capitalism flourishes and ultimately incubates the growth and development of freely distributable software.
This conference came at a particularly important time for the free software industry. If the concept is to move forward, it is imperative that advocates join hands with commercial endeavors wishing to exploit the benefits of free software. Niche markets must be targeted and developed so as to promote commercialization of free software with the ultimate benefit of subsequent re-investment.
Phil Hughes, publisher of Linux Journal probably phrased the imperative most correctly when he said, “It is time for Linux and free software to have applications and markets which allow it to DO something.”
Finally, here I am on a late night flight on my way back to Seattle so I guess I am done with what I need to tell you about the conference. We intend to keep up on plans for the next Freely Redistributable Software Conference and will let you know about it in plenty of time so you can attend. If you want to rub elbows with some of the important people in the free software movement, get genuinely inspired by the high interest level in Linux, learn something new and maybe drink a real or virtual beer or two, it is the right conference to attend.
The following papers were presented at the Sunday conference. Note that a printed copy of the proceedings is available for $25 (postpaid in the U.S., add $10 for foreign airmail) from Free Software Foundation, 59 Temple Place, Suite 330, Boston, MA 02111. They can be reached by phone at +1 617 542-5942 or fax at +1 617 542-2652.
Automated Management of a Heterogeneous Distributed Production Environment by Ph. Defert; CERN, European Laboratory for Particle Physics, Geneva, Switzerland
Freely Redistributable Software across the Internet—Current Practice and Future Directions to Overcome the Bandwidth Crisis by Neil Smith; HENSA Unix, University of Kent at Canterbury, UK
Cheap Operating Systems Research and Teaching with Linux by Victor Yodaiken, New Mexico Tech
Freely Redistributable Instead of Commercial Software—Yugoslav Experience by Radivoje Zonjic; Department of Electrical Engineering, Belgrade University, Yugoslavia
Linux on the OSF Mach3 Microkernel by Francois Barbou des Places; OSF Research Institute, Grenoble and Cambridge
Internationalization in the GNU project by Ulrich Drepper; University of Karlsruhe
Perceptions—An Implementation of a Medical Information Support Environment with Freely Distributable software by Dr. Greg W. Wettstein; Oncology Research Division Computing Facility, Roger Maris Cancer Center
The RPM Packaging System by Erik Troan; Red Hat Software
Coordination Joint Cost/No-Cost Rights for Software Developed with SBIR Funding by Philip A. Wilsey; Computer Architecture Design Laboratory, Department of ECECS, Cincinnati
Licensing Alternatives for Freely Redistributable Software by L. Peter Deutsch; Aladdin Enterprises