|
I have been hunting through RFCs and Internet Drafts for previous work related to my proposal. I found a recent Internet Draft by Sam Sun and Larry Lannom, "Handle System Overview". It points to an already rather mature Handle System project, with a software distribution.
There is a lot in common between my proposal, Frankston's, and the Handle System. I haven't had time to read all of the details of the Handle System. At first glance, it seems that:
- It commits to accommodating meaningful names as handles, except at the root level, which may make it vulnerable to tradename disputes.
- It contains a level of complexity in managing read and write permissions that I hoped to avoid.
- The project doesn't seem to propose any immediate deployment through under the current DNS, but depends on adoption of its software.
Like fnord, I am primarily concerned with the fact that there is no widespread deployment of some sort of handle system, even though the desirability appears to me to be a no-brainer, and independent of the particular ideology that motivated me to think of it. I expect to eventually find some discussion of it in an early RFC, once I guess what other word or phrase it used instead of "handle." It may be that the idea has always been bogged down in disagreement over how much service to build in to the basic system. My emphasis here is to identify the minimum service that could be provided by a strategic handle-resolution system, while allowing useful additions to be implemented independently, figure out what resources we need to get the thing going, and then start sending proposals to potential supporters.
Mike O'Donnell
|
