Why do Corner permalinks have such loooooonnnngggg URLs? Here's the one for Derb's latest "Darwinism" post:

http://corner.nationalreview.com/post/
?q=OTNkZTFkMDdmZTE5ZTc4MDE5ZTVmNTNiZGUzNTBmMjY=

Do you want to ensure uniqueness of them to the end of history? If so, that's real farsightedness. By my count, with that "q=whatever" parameter, you can provide 62^43rd unique post IDs (upper/lower case plus the ten digits, with 43 characters). We know you Corner guys (and gals) can be prolific, but consider that this may be overkill. Unless there's something going on that requires a unique fixed field length up front, it is possible to do it by simply going sequential (I'm only up to 9000 or so on my blog). I haven't done the math, but I suspect that with your scheme, you could assign an ID to every particle in the known universe, with more than a few left over. The sun would start to dim long before you got out to even the twentieth field for your post IDs.

What's even worse is that that's utterly meaningless to anything but their web backend. Not only does that tell a human being nothing about the content, but it's very hard to tell two of them apart without having them side-by-side.

A better format would be date-time-author_shorthand, which would both uniquely identify posts, and contain information that is both useful and readable to human beings.

It couldn't be much harder to implement and maintain, either.

Why not just bookmark "javascript:void(location.href='http://tinyurl.com/create.php?url='+location.href)"

Then hitting that bookmark generates a small URL for whatever.

Why not just bookmark...

Why bother? It's not a problem for me--I just bury it in my links. I'm just amused.

Maybe they use it to keep track of who refers to their site. That much data could also include information about which user (by session id) copied the URL. Ie, they couldn't tell what you had for dinner, but they could say that someone read this story last night and linked to it in their blog.

I suspect they are using WebObjects or some similar system. The long URL is because it's an object reference, probably encoding somewhat more than just which article is being referenced.

It's a right-wing conspiracy, if you ask me. The leftie sites only use short, helpful, people friendly, culturally appropriate, environmentally sensitive, non-competitive, vegetarian, carbonless, righteously freaked out URLs. Damn, they are smart.