David Weinberger, in his Second Sight article for The Guardian talks about domain names and identity, or as he notes in his blog, it is about the DNS mess and Google URLs.

The "the net" part of the iCite net includes iCNS (iCite Name Service), a name service something like a higher-level DNS. And, essentially, this is meant to enable a similar approach to what David is suggesting: since there are lots of people with the same name, and (second level) domain names are too limited to be uniformly distinguishing between all of them, a search / directory tool (like Google) can be used to build identifying URLs.

I think another similar approach is evident in blogrolls and other personal directories. For example, I know that hyperorg.com is where I get first hand info on David Weinberger (and davidweinberger.com isn't) because people link to this hyperorg.com domain name and call the link "David Weinberger".

In a more general context, this matter is similar to what Tim Bray talked about in his Bye-bye homepage? post, which I comment on in my bye bye homepage, bye bye website? post. I should have added: bye bye domain name.

The deal is: home pages, websites, and even domain names are less and less "bags" in which people contain themselves online. David is a good example: his article is on The Guardian, and he mentions it on his blog, both of which are on different domains from his homepage on evident.com, all of which are under different domains than his posts on ThreadsML which appear on the www.quicktopics.com domain.

In Tim's piece, he is talking about people going first to RSS feeds where they used to first go to homepages. I am further imagining people moving to first use directory organized RSS feeds (or, in general, lists of information packages) where they used to first use websites or even domain names. David's article suggests this: if you want to find Tony Blair online, you need to search him on Google because the Tony Blair domain names aren't the place to look. Google search results are a list-oriented of view of what I call lists of information packages, whereas RSS is more a package-oriented view.

For example, I would like to see a David Weinberger list / RSS feed that links to: his blog entries, his articles, his posts in discussion forums, his books, his issues of JOHO, and his speaking engagements. Wherever I come across a David Weinberger who may be this guy who does all these interesting things, I would like to be able to see this kind of view in which I might discern if this is the same guy or what.

iCites should enable exactly this kind of thing. They are both list-oriented and package-oriented. And, iCNS makes it possible to create unique URLs to any list or any package. Altogether, it also represents a decentrally created / managed directory with uniquely identified entries, which should be a nice thing to do searches on, by Google or otherwise.

As a final note: with iCites, a list of information packages about Tony Blair or David Weinberger can be built by multiple people in multiple locations (domains) and joined together. So, David Weinberger might create a list of links to his writings, and someone else might create a list to his writings as well. Those lists can link to (and thereby, effectively, include) each other, or a third-party like me could create a third list that joins them together.