The Strange Case of the Disappearing Open Source Vendors
Subject:   The Clothesline Paradox
Date:   2002-06-30 22:32:55
From:   timoreilly
Victor Yodaikan of Finite State Machine Labs: The RTLinux Company (, sent me an email message, which he said it was OK to post here, along with my reply.

Victor wrote:

    "I can't believe that people still take the Eric Raymond article seriously. Scanning the employment section of a newspaper will show you that there are many more people working in construction and maintainence than designing and selling building materials, so what? Software is both a construction material and a final product. All those people who build software for "use" depend on software products, from Java and Perl to Visual C++. In fact, much of the software "use" industry is like the huge machining industry that used to dwarf the manufacturing industry in the early days of automobiles. It is large in proportion to how much extra work is needed to make the products actually usable. You can look at the vast army of programmers needed to maintain and customize Oracle and believe that this proves that Oracle has little actual value as a product, but it seems pretty far fetched."

I replied:

No one said that software has "little actual value as a product", just that
if you only measure the software that is sold, you miss a great deal of the
market. This is a variation of what solar activist Steve Baer called "the
clothesline paradox" in his 1975 book Sunspots:

    "If you take down your clothesline and buy an electric clothes dryer, the electric consumption of the nation rises slightly. If you go in the other direction and remove the electric clothes dryer and install a clothesline, the consumption of electricity drops slightly, but there is no credit given anywhere on the charts and graphs to solar energy, which is now drying the clothes...."

We need to move away from absolutist thinking. I'm not saying we should
have only user-contributed software, and no software vendors. But I'm
trying to counter the idea that only software vendors are in the software
business. Microsoft has been working this idea pretty hard, arguing that
the GPL is bad for business because software vendors can't take the
resulting code and re-use it. My point: why should we care? Other vendors
can't take Microsoft's code and reuse it either! If the users of that
software can take it and work with it, more power to them. And if the users
want commercial vendors behind their software, more power to them. I'm for
choice and the marketplace. But I want a marketplace that recognizes all
the players, not just those that fit into the familiar models.

Victor then replied in turn:

    I'm ok with that. Microsoft's argument is nonsense for many reasons: including that fact that a very large percentage of computer/software/telcom technology was both developed and proved out at taxpayer expense and the lack of innovation in the proprietary market is striking. What I think is the main story of Linux, however, is that the core Linux developers showed far greater tenacity, ambition, and sense of the market than did a generation of pathetic corporate "competitors" who queued up to be pummeled by Redmond.

    On the other hand, the theory that software developers should live by support and customization alone has been shown to not work well in most sectors. In many cases, the "support/customization model" turns into "your software that makes my product more valuable should be free". And I think that this subsidizes a profoundly broken model of how technology business should work.

I'm in agreement with Victor that it's a non-starter to think that open source businesses can subsist only on revenue from support and customization. (See for example my article Lessons from the Layoffs at LinuxCare, written at the height of the open source VC frenzy.) But that isn't my point here. My point is that vendors are only a small part of the total software ecosystem. And claims such as those that Microsoft has been making, either directly or through shills like the de Tocqueville Institution, that what's bad for Microsoft is bad for the entire software industry, need to be contradicted.