The Blog of Harald

A common issue discussed in weblogging (and many other WWW environments) is the permalink.

It is often necessary to redesign web pages, or change layouts, in such a way that previous URLs no longer work. I’ll have to do this soon; the initial base URL I chose for my weblog is incorrect. But if I do, I’ll invalidate all of the existing links to my pages, unless I keep the old name or old hierarchy around.

Here’s a potential solution; the Persistent URL from the Online Computer Library Centre.

Functionally, a PURL is a URL. However, instead of pointing directly to the location of an Internet resource, a PURL points to an intermediate resolution service. The PURL resolution service associates the PURL with the actual URL and returns that URL to the client. The client can then complete the URL transaction in the normal fashion. In Web parlance, this is a standard HTTP “redirect”.

In MovableType, for example, the permalink for an entry depends on the default archive format you choose; if you need to change that format, all your permalinks change. I’m currently using Individual as the default, but as my website grows that might require too much disk space (a 512 byte entry wastes 1532 bytes of disk space on my server).

However, every blog entry has a blog ID and a six-digit entry ID. One could conceivably construct a PURL with this information and use it as the permalink; then register with an existing resolver (or run your own) to map the PURLs to your current blog archive layout. That would allow me to change my archive format to a more efficient weekly or monthly page without invalidating permalinks.

I’ll play with the software a bit at CFRQ, and report back.