All the Perl that's Practical to Extract and Report

Joel on Economics

ziggy on 2002-06-17T16:49:46

Today's Joel Spolsky article deals with the fundemental microeconomics behind open source. The gist of it is this: you need to examine the ideas behind substitutes, complements and commodities to understand any market, including software. Commodities are pretty easy. Substitutes are interchangeable products (Toyota Corolla, Subaru Legacy, Nissan Altima). Complements are products that work together (cars+gasoline, hotel+airfare, computer+operating system+office suite).

I think this is a breakthrough paper. Not because Joel is making exceedingly brilliant observations, but because he's describing what we already know (and have been talking about for years) in reasonably clear language.

Among his observations:

  • StarOffice / OpenOffice / KOffice / ApplixWare / Gobe Productive are not substitutes to Microsoft Office. They will keep being inferior solutions until they provide a zero-cost migration.

    (Apple is aware of this; their strategy is to show that switching to MacOS X is not a zero-cost switch, but a very low-cost switch, and claim MacOS X offers a lower TCO after you switch.)

  • Web browsers and Web servers are complements; Netscape gave the browser away to sell expensive servers. But web servers are a commodity, so that backfired in the long run
  • Open Source software is not zero-cost, especially when looking at the overall implementation cost (as measured by incompatible switches to the linux device driver API)
  • IBM is throwing a billion dollars at Linux because it makes good business sense. How? Because their primary revenue comes from selling enterprise-grade hardware and consulting. Anything that increases their enterprise-level sales is good for their business
  • Sun is in trouble

Dissent

pudge on 2002-06-18T14:38:56

I think this is a breakthrough paper. Not because Joel is making exceedingly brilliant observations, but because he's describing what we already know (and have been talking about for years) in reasonably clear language.

Why is it brilliant/breakthrough if it is something we already know and talk about? If it is because no one else has been able to state it in reasonably clear language, maybe rather than being brilliant, he is just competent, and we are all incompetent.

Really, though, I thought the paper was OK, but not especially interesting, partially because I dislike discussion of business models and practices and decisions, and partially because it didn't really say anything new. Maybe he said it a little better than it's been said in the past, but I don't think it qualifies as a breakthrough, even if it is interesting, which I disagree with anyway. :) But the point is that I don't see it being a breakthrough if it is restating what we have all been knowing and saying for years.

Oh well. Would you like some garbage collection with your IPO?

Re:Dissent

ziggy on 2002-06-18T14:59:48

Why is it brilliant/breakthrough if it is something we already know and talk about?
First, it's not a brilliant piece. I thought I made that clear.

Second, it's a breakthrough in the same sense that Cathedral and the Bazaar was a breakthrough. ESR was the first to link together enough pieces of the puzzle to explain why open source works, even when the conventional wisdom couldn't explain it. Joel's piece explains how and why open source fails when a project seems to be doing everything outlined in CatB.

Re:Dissent

pudge on 2002-06-18T15:37:17

Sorry, I misread about "brilliant."

I thought ESR's piece was breakthrough, but probably because at the time, I was new to Free Software/Open Source (I had been a programmer for less than two years when I attended the first Perl Conference and heard him give the paper, and had only just begun doing any work with FS/OS). Looking back at it now, in its historical context, I think far less of it than I previously did.