December 24, 2002
More to and fro Norlin
Eric is continuing his experiment in thinking out loud. He’s refined his original argument about the Net’s effect on the economy. It’s too rich a chunk to chew all at once, so I’ll just nibble at it. (Even though I’m about to disagree with him, this isn’t a “He’s Wrong!” sort of disagreement but an attempt to understand better by pushing back a bit.)
Eric begins by saying that the Internet is inefficient when it comes to managing reputations. Maybe, but so what? The Net’s strength is its inefficiency. A more efficient network would queue bits in order of importance, pre-compute routings, etc. So long as the bits are getting there, who cares if it’s the most efficient route? I worry that strong digital ID systems over-build in the name of efficiency and completeness.
The fact is that the Net has actually done an impressive job of building reputation management systems where they’re needed: eBay, Amazon, epinions all do a good enough job of it. Home pages and weblogs are another sort of reputation system. So, if there’s a market waiting to happen if only we had a reputation system, then why hasn’t someone already built one on the edge of the Net?
And the same goes for digital IDs themselves. If the need is that great, then why hasn’t it been solved? (I don’t mean this in a neener-neener way. I mean it as a real question.) Yes, we need strong digital IDs to enabling online voting. But I can already buy anything I want online by using my credit card. I have passwords at a zillion sites and a little password note pad that reminds me of what they are. My fear is that by trying to build systematic ID systems that don’t spring from particular applications, we’ll over-engineer a solution to a problem that doesn’t need that much solving.
And since over-solving this problem would benefit powers that would recentralize the network, there are additional reasons to aim for inefficient, minimal digital IDs.
Also, Eric makes a passing reference to an idea that sounds fascinating: “… property rights exist (at some economic level) to simplify the exchange.” He credits this to Frank Field. I hope Eric expands on it.
Doc is waxing wise in his response to Eric, although I think that in Doc’s terms I don’t want identity services to be “Net native.” I want them on the edge. I suspect Doc does, too, and we’re disagreeing only verbally.