This post is at http://gnodal.livejournal.com/57614.html … but it can also be reached at http://tinyurl.com/s5pgc … is that a bad thing? Good?

There is no such space as gnodal.livejournal.com (which, BTW, can also be reached at livejournal.com/~gnodal/); that’s a handy representation that makes a long number human-readable. But what happens when things move?

In the early days of the web my work creating user docs got me a 1Meg account on our community free net. After years of inactivity a sysop decided to cut that back to 500K. So files that had for years sat in chebucto.ns.ca/~ab006/ had to move. Likewise the is.dal.ca/canid/ account I had used while at Dalhousie University (programming VRML for the ethology lab in the psych department): the box was decomissioned and everything had to move.

Even when things are done well enough to ensure that nothing gets lost (not the case here) whole sets of documents end up 404. The better and more popular the site, the more profound the ramifications.

Raising this issue with Alexander Johannesen (Know many folk who would blog “The epistemological implications of Topic Maps for librarians“?) he replied, in part, “one of the policies at the library is that all links are permanent, meaning we get to hold on to all sorts of redundant crap and can’t introduce new paradigms in tools and practices of our content.*sigh*”

I’m suspicious; when things are persistently wrong I get to feeling that we really don’t know what we’re doing.

We’ve had DNS a long time … why is it so hard to index entities?