where "word of keystroke" begins

April 1, 1999 -- Business implications of the Linux development model

Transcript of the live chat session that took place Thursday, April 1, 1999. These sessions are normally scheduled for 12 noon-1 PM Eastern Time every Thursday. Please note that the US is now on Daylight Savings Time. So in international terms, we are on at GMT -4.

To connect to the chat room, go to

Since the chat itself happens at a rapid pace, it's often difficult to note interesting facts in particular URLs as they appear on-line. Here's a place to take a more leisurely look. I've rearranged some of the pieces to try to capture the various threads of discussion (which sometimes get lost in the rush of live chat).

Please send email with your follow-on questions and comments, and suggestions for topics we should focus on in future sessions. So long as the volume of email responses is manageable, I'll post the most pertinent ones here for all to see.

These sessions are hosted by Richard Seltzer. If you would like to receive email reminders of our chat sessions, simply send a blank email message to or go to and sign up there.

For transcripts of previous sessions and a list of future topics,

For an article on how to make "business chat" work (based on this experience),

For articles on topics related to this one, check our newsletter, Internet-on-a-Disk

Threads (reconstructed after the fact):

Today's participants


Richard Seltzer -- We'll be getting started in about an hour -- at noon Eastern Time (GMT -5), continuing our discussion about the business implications of the Linux development model. As you connect, please introduce yourselves and let us know your interests.

Bob Fleischer -- Bob Fleischer, Compaq Services, seeking alternatives

Kathleen Gilroy -- I'm Kathleen Gilroy, CEO of the OTTERGroup, a distance learning company, based in Cambridge. We are exploring new methods of developing courseware that are more collaborative in nature.

Barbara -- Hello. My name is Barbara and I'm just here to learn.

Richard Seltzer -- There's a chance that Eric Raymond (author of the seminal article about the Cathedral and the Bazaar) might join us today -- which would certainly add a new dimension to the conversation.

Richard Seltzer -- Welcome, Barbara, Jeff S, and dmackell.

Jeff S -- Hi Richard, et al. I went to one of Richard's talks at the NH isig a while back and still work at his "alma mater" (Digital, now Compaq) in Nashua, NH. I've been having trouble posting, since the refresh has been clearing my message...

Richard Seltzer -- Jeff -- real quick, log out and log back in with frames. Also set the refresh rate at zero, and to see new messages click on "new messages". Otherwise every refresh erases what you were writing. Sorry about that.

Richard Seltzer -- Last week we were probing at how you can get a bazaar started -- what are the circumstances that make such a model viable? we touched on such factors as the size of the group and also the fact that at the heart of the project their must be some goal, some holy grail that many people can identify with. Company goals would not be strong enough. But company goals might well be tied to such an over-riding mission. 

The bazaar and communications within large organizationns

Kathleen Gilroy -- I'd like to start with talking about how the "bazaar" model can be used to improve communication within large organizations. Any thoughts on that topic?

Richard Seltzer -- Kathleen -- when I think of the bazaar model with regard to communications inside a large company, I can't help but think of the VAXnotes environment inside Digital 10 years ago. Tens of thousands of people freely participated. Many of the discussions related directly to work. The discussion environment favored ideas and their clear expression rather than the position of authority of the person posting. But 1) managment (outside of Engineering) never got involved and hence rarely benefitted from the discussion and 2) in today's environment that would be difficult to duplicate because employees have many more places in which to discuss matters of concern to them online. 

Diversity and vulnerability to viruses

Richard Seltzer -- In terms of the benefits of the "bazaar" style, I saw an interesting forwarded message just a few minutes ago. It was an item from Red Herring, by the editor, Rafe Needleman. He was talking about the Melissa virus,a nd made the point that it only acts on Microsoft email clients. He notes "Undiversified biological ecosystems, jsut like all-Outlook offices, are at increased risk of ctastrophic infection because the lack of diversity makes it mroe likely that a single failure can cripple the entire system. This argues in favor of open standards (no monolithic control over a system and, hopefuly, no hidden vulnerabilities) and multple platforms (if one system breaks, you can use another.)"

Bob Fleischer -- It may seem like ancient history now, but I recall that there was at least one virus/worm (Morris?) that spread through the Unix community on the internet a number of years ago. I don't think open source per se discourages that -- it may even encourage it in the short run (since it will be easier for the less-dedicated hacker to find vulnerabilities). On the other hand I do agree that diversity is a strength against total loss from such problems. My concern there is that the computer market seems to discourage diversity -- there is a desire to have the greatest range of application/software/hardware options available, and that tends to mean one (or perhaps two?) mega-platforms. What would really be nice would be multiple implementations of the same systems, in which the vulnerabilities would lie in different areas but the interfaces would allow interchangability.

Richard Seltzer -- Bob -- Yes, I agree that the key is interchangeability. I think that that was the original intent of "open" -- to make it so diverse systems could work together smoothly, so that it was not necessary for everything to be identical (and from the same vendor) for it to work together. If we were imagining a "distance learning" bazaar, that could be one of the goals. There might be a systems goal of providing a framework in which all the diverse chat and forum and related applications could be interoperable (and indexable/searchable), at the same time as developing curricula and procedures that could be broadly adopted for online learning. (I'm still too fuzzy here... especially with regard to what can come from a "bazaar" in terms of content and curriculum...)

Bob Fleischer -- The warning here is that one mustn't assume that multiple implementations with interchangable parts will flow from an "open" process. The many Unix systems were vulnerable to that ancient worm because they were in fact variants of the same implementation. 


Kathleen Gilroy -- Perhaps people could comment on the role of "leadership" in the bazaar. What are the ingredients that make for effective leadership and do they differ from other environments?

Barbara -- It seems that leadership would vary. One person being the leader with one idea, another taking the lead with another idea. I don't see one leader, although you do need someone who moves things along in the long run.

Richard Seltzer -- "Leadership" is one of the puzzles that Eric Raymond tried to deal with in his paper. The "leader" of a subgroup tends to be in part self-selected -- someone with the inclination, and the time to do it; and someone with online social skills and online recognition at a level where this person already commands respect. This implies that there is a pre-existing online social "soup" -- many of the people who sign on to become active participants of the development activity have interacted before in less formal ways in the general online environment (in Eric's case, the Linux community). This is beginning to feel like there are layers inside of layers -- a general "soup", then a bazaar defined by a compelling mission that many can identify with, then subgroups where leaders emerge and are recognized. (Leadership style is a separate, related issue -- but the point I'm getting at here is that no one "chooses" a leader; a leader steps forward and is recognized (or not). 

Size matters

Richard Seltzer -- Another aspect of the "bazaar" that we haven't touched upon, but which I feel is important -- it needs to be large enough and diverse enough to foster numerous subcommunities/subbazaars. Eric Raymond's project was really a subproject in the realm of Linux. I am thinking of Ebay, as an enormous bazaar-style environment. No one is "developing" anything there. But there are hundreds if not thousands of subcommunities there (people who buy/collect particular kinds of things), where participants build relationships with one another and "recognition" within the community becomes a major motivator. (I sense that many people get involved to the point where the experience itself motivates them -- not obtaining a particular object, or making money by selling, but the social interaction -- I think of the Trobriand Islanders where the trading they did among one another was for purposes of building relationships rather than for any intrinsic value. He sense that at Ebay there is an aspect to the person-to-person transaction which is like an exchange of subatomic particles (that image needs more work :-). What I'm getting at is that an "bazaar" environment needs to foster a self-regulating, recognition based culture, which means there have to be subcommunities the right size for people to "know" one another and appreciate one another's accomplishments, while the whole bazaar itself may be immense.

Kathleen Gilroy -- Richard, I do think scale matters, but scale relevant to what? The size of the problem? In our distance learning environments, we plan for 65 people in a remote classroom. (It's also a good size for parties). Enough variety for diverse and interesting communication. What's the optimal size for a chat? Today's group, as interesting as it is, feels too small and is de-motivating for me.

Richard Seltzer -- Kathleen -- Yes, today's group is too small. I prefer at least a dozen, and up to about 2-3 dozen active participants, with even more viewers. Ideally, a 65-person class would constitute a sub-group. Ideally, the best of what is generated by their discussion would be posted where the other subgroups could view it (and were expected to view and react). and there would be several stages of recognition and cross-fertilization of ideas. (But how do you motivate people to actually do this? To get started? Once started, an internal recognition community can build its own motivations. But for starters, maybe some level of participation is a requirement of the course. (Like requiring students to provide critiques on one another's papers.)

Kathleen Gilroy -- Richard, I think the starter is the compelling nature of the problem. We're winding down on this discussion because it doesn't seem to be of significantly high value to a group. It may be too subtle and complex. What has allowed you to get a dozen or more going in the past?

Richard Seltzer -- Kathleen -- I believe that, in part, we seeing the effects of the evolution of the Web. When we started this chat program three years ago, there were very few scheduled, focused chat programs. Also, "business on the Web" was something out of the ordinary. Very few businesses were really using the Web for anything more than to post their brochures. Now every business of any size uses the Web. Also, now, topics that we used to hit upon for a few weeks have become specialties that thousands of people are devoting their careers on or betting their businesses on. We set this chat up as a "generalist" forum, and the online world is evolving more toward specialties. Theoretically, I believe that much is lost by not taking the broader perspective and becoming familiar with many related areas. But in practice, it is hard to attract the attention of someone with a topic that isn't in their specialty. I believe that this is a great topic that many could benefit from. but they just aren't looking in this direction...

Richard Seltzer -- Kathleen -- yes, it is frustrating when we don't have the critical mass in terms of number of participants. With the right numbers and the right mix, this can take off in very interesting and unpredictable directions. But in an open format like this, there are no rewards and incentives for participation, other than the social contact and mental stimulation. I wish I had the magic formula. I feel we're close, but not close enough.

Design principles for operating in this new bazaar-style medium?

Kathleen Gilroy -- In terms of thinking about courseware, the problem with "cathedral"-like development is that one runs the risk of creating something that is of lesser value to the marketplace. The bazaar model puts you closer to your customers. But it is more cumbersome (and therefore more expensive) to manage--if it need be managed....

Kathleen Gilroy -- I am also interested if anyone has thoughts on "design principles" that this loose configuration of collaborators both requires and enables.

Richard Seltzer -- Kathleen -- I think it's important to separate the role of the "bazaar" and the aims of the company which gives it a home. I think of the "bazaar" as an open marketplace of ideas. That's the testing ground. A distance learning company might pick and choose from among the ideas and developments that come out of a bazaar those pieces that make sense for the business; and might encourage others to adopt other ideas that are good and should be implemented but just don't fit for this particular business. This feels too fuzzy, I must admit. We need to get the discussion going, to generate the enthusiasm and creativity that is characteristic of a bazaar and then see how a company can benefit from it. This feels like too much theory at this point. How do we move from theory to practice? And how do we seed such activity in an environment where there is so much competition for everyone's online attention?

Kathleen Gilroy -- Richard, I think people are searching for design principles to make themselves more effective in the electronic universe, above and beyond an open and interoperable medium for learning (in a way the internet itself is that medium). I am thinking about the professor's point of view. Once I open up the walls of the classroom, what do I need to do to make my communications better/different? How do I modify my processes to work within a new medium? Any comments?

Richard Seltzer -- Blue sky -- I could imagine a set of courses each of which had some concrete content that needs to be mastered and each of which also has a component that deals with trying to understand the changing business environment, where this is all headed. The open-ended piece involves matters for which no one individual and certainly no textbook holds the answers. And all the courses have that open-ended piece in common. Part of that piece is actually self-directed, because on-line interaction, like that which the students are engaged in, is an important element of the future of business (building relationships through online interactions, rather than just racking up daily sales).

Kathleen Gilroy -- Richard, in a way, what you are describing is the case-study method, where the learning occurs through common discourse around a structured problem. I think the challenge comes in managing engagement, once you no longer have a "geographically discrete" community. It's being done in forums and chats all over the web, but where is it being done really well? Can some design principles be built? And how do you teach people to do this?

Richard Seltzer -- Kathleen -- I believe that that is the crux of the problem and the opportunity. Yes, the case-study method is used at many schools, but are there any instances of the insights gained from that process being shared among classes, much less schools? Yes, the instructors may incorporate what they've learned from students in the next iteration. But I've never seen an environment where the best conntributions of students were preserved and shared and improved upon in discussions that extend beyond the individual class and that continue over time. But I'm not sure how to motivate that activity -- promises of publication? other forms of recognition? the opportunity to get significant discounts on future courses or to move to the ranks of tutor or instructor? there needs to be a tangible reward system, at the very least to get this started. Priming the pump is not a trivial task.

Kathleen Gilroy -- Motivation is key. The reward system in Linux is recognition-based, as it is in the classroom. Students participate because they want the recognition of the professor and their peers--I am not sure how to extend these rewards beyond the classroom. Perhaps you are right about moving people up the food chain in course creation and management. 

Open source alternative to Windows?

Bob Fleischer -- One problem I'd like to explore is: "Is it possible to develop an open source *compatible* alternative to the Microsoft windows platform?"

Richard Seltzer -- Bob -- Is Linux heading in that direction? (I finally splurged and spent $29.95 online at Xoom to get the Red Hat version of Linux -- surprisingly they send it by snail mail rather than offering it as a download, so I don't have it yet. The promo material indicated that there's Word processing software that runs on Linux that is "compatible" with Word. I need to try it to believe it. But I suspect that that is a natural direction for open development, with or without Microsoft's cooperation.

Kathleen Gilroy -- Bob, I think I saw something in the Globe (Hiawatha's column) a week or so back on new applications and operating software that is coming out of the Linux movement. She did not think it was "ready for prime time." And there is the whole issue of interoperability. I am finally bending to the monopoly and am about to buy a Wintel machine (after being a Mac diehard) because I need common formats for document sharing.

Bob Fleischer -- I want a system that is compatible at the application programming interfaces, so that I could actually run Word (or any other software that happened to be built for Windows, which just happens to have the biggest supply of software). Remember back when software was only supported on IBM PCs, and companies like Compaq worked mightily to create truly interchangable clones, so that the software that was built for the dominant architecture would run on their systems as well? Eventually, they got so good they out-competed IBM, but they needed the start of a ready supply of software, which compatibility gave them. Why is it only Microsoft which seems like a monopoly today? IBM is being driven out of the PC business, and Intel faces stiff competition from alternative, compatible implementations. Why not Windows?

Kathleen Gilroy -- Bob, Microsoft's monopoly is a resource issue--nobody is going to "fund" the development of competitive alternatives. Software may be easy to make, but it succeeds or fails on marketing (which takes major resources). 


Kathleen Gilroy -- Richard, It looks like this topic has run out of steam. Any ideas about what's next on the agenda?

Richard Seltzer -- All -- what would you like to do next? I believe that the topic has much more to it than we have explored, but without the right numbers and diverse perspectives, we're not likely to benefit from continuing in this direction. (It would have been fun if Eric had in fact showed up today, as he said he might.)

Richard Seltzer -- Andrea Newman suggested Marketing Communications -- the impact of the Internet on how you get your marketing messages out to the world. Does that strike anyone as compelling?

Richard Seltzer -- All -- We're at the end of the hour. As usual, I'll post the edited transcript. Check I'll also post the various threads as separate messages in the forum area at Kathleen's site -- Please check that out and add your further thoughts there. Also, we need good suggestions of compelling future topics (and creative ideas for how this program can provide some tangible benefit). Please send me email at

Kathleen Gilroy -- Richard, what about talking about one of the chapters in your book, or content-based internet marketing?

Richard Seltzer -- Kathleen -- Sure, I'd be happy to talk about "content-based marketing". I did a speech on that at Web-net a couple weeks ago, and they have asked me to come back and speak again at their April meeting. But I need some hook, I need to identify a constituency for whom that discussion has real and immediate value.

Kathleen Gilroy -- Thanks Richard, I'll be in touch by email soon.

Barbara -- Good bye. Thanks for an interesting chat.

Richard Seltzer -- Thanks again to all.

Previous transcripts and schedule of upcoming chats --

To connect to the chat room, go to

The full text of Richard Seltzer's books The Social Web, Take Charge of Your Web Site, Shop Online the Lazy Way, and The Way of the Web, plus more than a hundred related articles are available on CD ROM My Internet: a Personal View of Internet Business Opportunities.

Web Business Boot Camp: Hands-on Internet lessons for manager, entrepreneurs, and professionals by Richard Seltzer (Wiley, 2002). No-nonsense guide targets activities that anyone can perform to achieve online business
success. Reviews.

a library for the price of a book.

This site is Published by Samizdat Express, 213 Deerfield Lane, Orange, CT 06477. (203) 553-9925.

Return to Samizdat Express