Linuxcare's CTO Respondsby Tim O'Reilly
Dave Sifry, one of the founders of Linuxcare and its CTO, took me to task for my article about web services. First off, he criticized me for not checking in with Linuxcare before publishing the article, since he felt that it misrepresented their business. Second, he pointed out that Linuxcare was in fact doing many of the things that I was talking about in my article, and that I should be using them as a good example, not a bad one.
Since Dave did a great job of stating his own case, I want to quote extensively from his email. Here's what he had to say:
You wrote, "Linuxcare's initial business model involved a great deal of reliance on phone-based tech support and other low level services". Did you know that phone support has accounted for less than 10% of the total tech support tickets that we handle? We created a significantly easier way to provide service for our customers, by creating an online knowledgebase that integrates information both from publicly accessible sources (USENET, mailing lists, web sites, HOWTOs, etc.) and information culled from our staff of experts. Once we solve a problem for a customer, the information goes back into the knowledgebase so that customers (and non-customers) can get access to that information - which further reduces the number of people who end up contacting us, and we end up with happier customers at a lower cost. We realized the best way to provide support to people was to solve their problems before they knew they had a problem, and let them find the answers themselves. It makes for satisfied customers, less frustrated support staff, and higher margins because we don't have to spend the time answering the same question over and over again.
I appreciate your UUNet story, because in it you make a couple of interesting points - that markets evolve, and that the revenue producing opportunities in a given market are temporal. Rick Adams started off by selling dialup UUCP access to his customers - but as TCP/IP became more predominant, he moved more and more accounts to full-time TCP/IP service. The same effect is occurring in the open source space - 2 years ago, the one question everyone asked was, "Where can I get commercial support and services for this stuff?" Just as Rick started his business providing UUCP-based feeds, Linuxcare started out answering the single largest problem that enterprise Linux customers had - and we did it better than anyone else in the industry. But in the same way that Rick saw his business evolve, we too are seeing our business evolve. Just as a man who wandered in a desert immediately sates his thirst upon finding an oasis, it is only after he has drank his fill that he realizes he is hungry. Our customers have begun to sate their thirst and are recognizing that open source software and the open source process can do a lot more for them than just email, file, print, web, and DNS! They are deploying open source solutions in datacenters, in telco closets, in ERP systems, and in embedded systems, to name just a few. They have also seen that by taking advantage of the fundamental shift in the relationship between the user and the developer of code, they can get to market faster, they can reduce development risk and cost, they can turn potential competitors into allies, and they can create open standards where none existed before.
You actually make our argument for us when you refer to Collab.net, one of your investments. You wrote, "In the end, open source isn't just about new software licensing terms but also about a new, internet-enabled style of distributed software development." Yes, that's absolutely right. Companies large and small are intrigued by the power of the open source model, but doing it right is a hard job, and requires a unique set of skills. We've seen this before - with the growth of the Internet itself. Remember that taking advantage of the Internet as a medium takes a lot more than just converting one's corporate brochure and putting it up as a web site. Some companies early on learned that the web called for a different way of communicating between buyers and sellers, that it blurred the line between producer and consumer, creating a new intimacy. Just as in 1995-1998 the question on every CEO's lips was, "What's your Internet strategy", I believe that in 2001-2004 the question will be, "What's your open source strategy". Who is going to be there to provide those services? Who will provide a trusted source of information, advice, and know-how? The companies that can effectively answer those questions will reap enormous profits and will help to influence the shape of the open source movement for years to come.
I've already apologized to Dave for not talking with him before publishing my article, so that I could have gotten more of an inside view of how they now view their business. (However, I did point out that I took my starting point about Linuxcare's initial reliance on phone tech support as its revenue stream from a recent quote by Pat Lambs, Linuxcare's current CEO, in the news.com article that sparked my original piece.) But I'm really glad that Linuxcare is in fact looking beyond pure tech support to higher-value services. I suggested as much in my original piece, and based on what Dave says, it looks like they are on a really good track to deliver.
Overall, though, Dave's comments also made me realize that I was trying to do too many things in that article! I was using the current news about Linuxcare as a starting point for some "big picture" insights that I wanted to share. That's all well and good, but to the extent that I made Linuxcare look bad by implication, that was not a great jumping off point. Further, to the extent that collab.net and Digital Creations also employ professional services as part of their business model, I further muddied the water.
My key point was that professional services need to be part of a bigger "solutions" framework. In particular, I wanted to share my enthusiasm for the idea that (as I said in my original article), "In the age of the net, 'services' doesn't just mean professional services, it means figuring out how to build frameworks for automating activities that used to require human labor."
That's the drum I really want to bang. Since I wrote my Linuxcare article, Rael Dornfest of the O'Reilly Network has written a great article about the API he's provided for the O'Reilly Network's Meerkat emerging technologies wire service that illustrates very concretely what I'm talking about.
I plan to do some more writing about web APIs shortly, but I wanted to get an apology out to Linuxcare while the subject was still "hot." And despite the bad feelings, I'm hopeful that my comments will be helpful to Linuxcare overall. This idea of web APIs and e-services is an important new trend on the web, and something that companies like Collab.Net, Digital Creations, and Linuxcare will all need to incorporate into their services vision if they aren't already doing so.
Tim O'Reilly is the founder and CEO of O'Reilly Media, Inc., thought by many to be the best computer book publisher in the world. In addition to Foo Camps ("Friends of O'Reilly" Camps, which gave rise to the "un-conference" movement), O'Reilly Media also hosts conferences on technology topics, including the Web 2.0 Summit, the Web 2.0 Expo, the O'Reilly Open Source Convention, the Gov 2.0 Summit, and the Gov 2.0 Expo. Tim's blog, the O'Reilly Radar, "watches the alpha geeks" to determine emerging technology trends, and serves as a platform for advocacy about issues of importance to the technical community. Tim's long-term vision for his company is to change the world by spreading the knowledge of innovators. In addition to O'Reilly Media, Tim is a founder of Safari Books Online, a pioneering subscription service for accessing books online, and O'Reilly AlphaTech Ventures, an early-stage venture firm.