At today's meeting, Verisign's Chuck Gomes rhetorically asked whether the conclusion should be that innovation at the network's edges should be encouraged, even when it breaks standards, and that innovation at the network's center should be discouraged, even when it complies with standards.
Things are, of course, more difficult than that.
Even innovation + edge + standards compliance doesn't automatically mean that something is a good idea and should be encouraged. Think of spam as an example: It was a new idea some years ago, it happens at the edge, and you can easily do it without violating any RFCs. Still it's a really bad idea.
What kind of innovation should be encouraged (or dicouraged), then?
I'd offer these criteria: Innovation should empower the end user. It should offer more choice to the user, not less. It should make it easier for the user to exercise choice, instead of making it more burdensome. Innovation should not cause increased cost to many for the benefit of few. In particular, it should not break existing technology or longstanding design assumptions.
Sitefinder's implementation does not pass these criteria. More generally, it's hard to fulfill them when innovating at the network's center -- but it's easy to fulfill them at the net's edge. It's also easy not to fulfill them at either the network's center or the network's edge.
Comments (1)
Posted by Alexander | October 8, 2003 10:38 PM
Posted on October 8, 2003 22:38