Digital Phoenix

by Henry on August 23, 2005

[This is the first of a few book reviews that have been piling up on my desk – next up is Chris Mooney’s The Republican War on Science, and then sometime in the not too distant future, Glyn Morgan’s The Idea of a European Superstate].

Digital Phoenix: Why the Information Economy Collapsed and How It Will Rise Again by Bruce Abramson, (the MIT Press 2005). Available from Powells and from Amazon (deprecated).*

Bruce Abramson’s’ Digital Phoenix is a smart read – it combines an excellent overview of the recent developments of the digital economy, with some important insights into how it works. The writing style is pacey, the stories (the Microsoft-Netscape battles, the MP3 wars, the birth of open source) are well told, and the quite substantial intellectual content is delivered in a user-friendly format. It’s the best non-technical account I’ve read of how network economies do and do not work in the information age. I’ll be assigning it to my students – as far as I can see, it’s the best and most complete account available. This said, there are two problems. First is its slightly breathy enconium to the new economy. All we need to do, says Abramson, is to renew our faith in the “corporate innovators and entrepreneurs who make growth possible,” and we can achieve the original promise of the information technology revolution. It isn’t that simple; the New Economy was never everything it was cracked up to be, and the future, insofar as we can discern it, seems likely to be considerably weirder than Abramson gives it credit for being. Second is the concluding section which feels a bit tacked on, jumping into an argument over the fight for control of the information economy between terror movements and authoritarian governments on the one hand, and democratic liberals on the other. It reads like the conclusions of a very different book, and a substantially inferior one.

When Abramson is good, he’s very good indeed. His description of the dot com bubble and the flawed thinking behind it is both lucid and convincing. He shows both (a) how network economies aren’t as simple as many investors thought they were, but (b) how they still can bite – especially when they are reinforced by a nasty intellectual property regime. Abramson’s account of the relationship between network economies and lock-in alone is worth the purchase price of the book. As he says, many investors assumed that network economies, tipping effects, increasing returns to scale etc would create a hundred Microsofts – firms that had achieved early dominance in their market sector and were able to translate that dominance into monopoly profits. The problem, of course is that network effects only produces a real advantage for incumbents when there is some mechanism that creates lock-in. Otherwise, when a Microsoft-wannabe raises its prices so as to reap monopoly profits, it finds its customers disappearing to other competitors who can offer the same goods more cheaply. The failure of investors to realize this very basic fact explains much of the irrational exuberance of the dot-com economy.

So what explains Microsoft’s (continued) dominance? Here, Abramson refers to Brooks’ argument on modular design. Programmers write applications that sit on top of ‘platforms’ (operating systems/interfaces)- as the application becomes more stable, it should eventually be migrated downwards, so that it is integrated into the platform. But platforms also generate switching costs, and allow the platform owners to use APIs strategically to maintain their market dominance. Microsoft’s control of the Windows platform has allowed it to see off potential competitors in the operating system/interface market, and to achieve partial control of key applications markets.

Abramson argues that these problems are aggravated by the US IP law system, which tends increasingly to systematically favour incumbents. Like Lessig, he worries that a combination of monopoly power and draconian intellectual property rights is liable to strangle innovation. He illustrates his case with chapters that describe the Microsoft v. Netscape case, the battles over downloading of music and movies, and the origins and promise of open source software. Abramson argues that the solution to many of the most pressing problems of the information economy lies in a radical rethinking of intellectual property law. Here, he combines a Zittrain-type solution of limited copyrights for software, with an incentive system that would privilege open source solutions. In Abramson’s ideal world, firms would have a choice. Either they could keep their source code secret – and enjoy no copyright protection – or reveal their source code and have copyright protection over the object code for a few years, so that they could enjoy the fruits of their innovation. It’s an interesting proposal – as Abramson acknowledges it is a little rough around the edges – but it’s a line of thinking that has some promise.

Abramson’s justification for this proposal is rather dicier. He claims that reforming the intellectual property regime would be a statement of faith in the entrepreneurial spirit. When the dot-com bubble burst, “we lost our trust in corporate America ” (italics in original); we now need to trust it again so that the wellsprings of innovation will be renewed. But this analysis doesn’t seem to me to be supported very well by developments in the information economy over the last few years. The new patterns of innovation that he (rightly) points to as key drivers – in particular the open source movement – aren’t corporate in nature. While they’re supported in part by canny corporations such as IBM, they don’t really work according to a market logic, as Steve Weber has argued. Instead, they involve a quite complex set of social motivations that only in part have an economic rationale. Abramson fudges this by invoking Eric Raymond’s metaphor of open source production as a bazaar.

With one little metaphor, we’ve moved from the quasi socialist sounding ideals underlying free software to the rawest known form of capitalist markets to describe open-source software – and we haven’t changed a damn thing

Indeed we haven’t. Raymond’s metaphor is inapt and misleading – while open source isn’t a directed process of innovation, it doesn’t involve markets, or anything like markets. Instead, it involves something a lot closer to dialogue and non-market driven exchange; rough consensus and running code. That isn’t to say that open source isn’t embedded in the wider market economy. Corporations can support open source solutions that further their interests (or weaken their competitors’ monopoly power) by donating programming time and other resources. Open source programmers can build and burnish economically valuable reputations through successfully working on open source projects. But this isn’t the whole of the story, and doesn’t explain how open source software itself is created – it’s produced in an informal social space, with norms and intrinsic rewards, that indeed resembles a kind of quasi-communism.

Abramson, by emphasizing the role of corporate incentives, is arguably missing out on one of the key lessons of open source and related phenomena. It may be that there isn’t any way for corporations – or other entities – to reliably capture the profits that should flow from innovations and valuable new content. But it also may be that this doesn’t matter as much as it should – some of the most important new innovations aren’t motivated by the desire for profit, and for those that are, even a tiny portion of the rents that are potentially there in the abstract may be enough to motivate innovation. Blogging is a good example. A very small number of people make the equivalent of a good living wage from blogging or more; perhaps several hundred or a thousand make a moderate-to-decent-ish supplement to the salaries from their day jobs. The closest thing to a blogging publishing conglomerate, Nick Denton’s mini-empire, has a small fraction of the turnover of your average suburban supermarket. But several million people seem willing to pursue blogging, without the sniff of any real financial reward from it, sometimes providing very substantial and interesting content. The “New Economy,” or bits and pieces of it at least, seems exactly to run on quasi-socialist principles. From each according to his or her ability; to each according to his or her needs. The future may turn out to be weirder than we expect, and trusting in corporations may be exactly the wrong thing to do. Instead we should trust in consumers, who are in many cases becoming non-market oriented producers thanks to massive reductions in the transaction costs of, say, publishing on the WWW, or contributing to an open source software project. While Abramson’s prescriptions for a more flexible IP regime would have positive consequences, they might not be the positive consequences that he anticipates.

Which leads to my final disagreement with Abramson. In a final section, which is somewhat disconnected from the main text, he argues that we are seeing a battle between the forces of authoritarianism and terrorism on the one side, and democratic liberalism on the other, to exert mastery over the Network. Shades not only of The Matrix but of Highlander; for one to succeed, the other must fail. There Can Be Only One. But it’s not quite clear why this should be true – and those indicators we can see seem to suggest that say, China, can maintain a fair degree of control over the Internet side-by-side with an anything-goes attitude in the US. The future is going to be mixed – and freedom and control are two sides of the same coin of technology. To consider a slightly different aspect of this: as I’ve argued before, there’s something enormously attractive and exciting about the new technologies which are making consumption and production blur together, so that individuals can produce or remix existing content freely, and distribute it. BoingBoing socialism, as it were. But even if the IP-zealots with their DRM, DMCAs and the like lose, these new technologies have a double edge. The ability to remix is also the ability to censor – it’ll be a lot easier for groups to isolate themselves from “mainstream culture” by chopping out the bits of it that they don’t like. Sometimes this will be good, sometimes bad (it may undermine basic values of tolerance). In short, we may be seeing a very complicated order emerging, in which techno-hipsters and fundamentalist Christian Sponge-Bob-bashers both make out like bandits. Abramson’s conclusions seem a bit simplistic – especially in light of the quite sophisticated analysis elsewhere in the book.

But enough criticism – it’s a smart, well-written, challenging book, which should start an interesting conversation. Go read.

  • I recommend that interested readers buy from Powells rather than Amazon – the former is union friendly and treats its workers rather better. All income that I receive from both sources will be donated to charity.



Barry Freed 08.23.05 at 9:56 am

That’s all well and good but let’s cut to the chase. It certainly sounds pretty interesting but what I and I’m sure many others will really want to know before shelling out our hard earned money for the book is whether or not the author supports our President, the troops and their mission in The Greatest War Ever Fought or not? It’s the only thing that matters these days. Well, that and sitting on my ass and inscribing snark in the comments on blogs.


Tom T. 08.23.05 at 7:42 pm

This is an excellent review; you make the book sound quite interesting.

One question: Isn’t Abramson’s proposal for limited copyright for software essentially describing patent protection? Disclose one’s innovation publicly, and get a few years’ monopoly? Perhaps that’s the central problem with software law: At the outset, software code should have been recognized as a patentable device, rather than a copyrightable literary work.

One small quibble: I’m not sure that your point about blogging goes too far, because I don’t think blogging represents any sort of innovative economic activity. Rather, it’s a hobby, like skateboarding, beading, or plain-paper diary writing. Some few individuals make money at those things, but as with any hobby, most people do it because they like it, not because they expect to get paid.


John Quiggin 08.23.05 at 8:57 pm

Tom T, I think your quibble is actually central to the argument.

Blogging is non-economic like a hobby, but it’s still a source of economically relevant innovations. RSS is a big example of something that’s spread from blogs to the Internet in general and RSS is significant not only in itself but as the first big application of XML.

And what’s true for blogging now has been true for the Internet as a whole. It’s the hobbyists/consumers/household sector who’ve driven the advances and the corporate and government sectors that have followed them.

The dotcom boom was supposed to show that hobbyist development (with government backing) was an immature phase and that the corporate sector was now ready to take over. But what it showed instead was that corporate business had no idea how to turn leading-edge innovation into profits, or even into a business plan sustainable long enough to deliver substantial real innovation.


paul lawson 08.24.05 at 7:33 am

Thanks, Henry and John Q. But let me ask four questions. (I can anyway, so the ‘let’ is courteously redundant, or an oblique part of my point.)

1. If, by initial design, ‘packet switching’ can survive nuclear war, how can it be controlled?

The late, and reflexive, capitalism struggles with the anarchy, makes a short-term ‘quid’ and moves on–to the next quarterly earnings report.

2. If ‘hobbyism’ is one of those increasingly affordable ‘affordances’ for the dis-employed, can they not ‘organize?

The ‘Talking Tree’ in pre-revolutionary Paris comes to mind.

3. Is this not the ultimate ‘subversion’?

(Himanen has made some arguments to the Finnish Parliament that innovation and a welfare state are intertwined. Perhaps this is the real ‘terrorism’ threat to the ‘maddies’ runnning, in some ways, their own ‘pension programs’, or anti-social ‘security’, elsewhere.

4. Does anyone have the least idea?

I don’t think so, and am with Henry on “weird”.

The weird are going ‘shopping’–and they are…we don’t know.

Comments on this entry are closed.