My URI scheme
After reading Mathias’ rant about his URI (URL, URI, IRI, what the hell?), I’ve decided to choose something simple for my site as well.
When I first started out blogging, I was using the most heavily used and most clunky structure known to blogland, that of /archives/yyyy/mm/dd/post-name/. I mean, it was huge. I believe every blog worth its money at that time had this awkward structure going for it. It was around this time that the slew of URL snippers were born. But that’s not the point of this.
The point is, what part of the scheme is important? Obviously, /archives/ did little more than add the notion that this was, infact, a blog. The next part, /yyyy/mm/dd/ is the important one, cause it helps categorise your entried by date posted, and lets people easily remember then. The /post-name/ part is obviously important, being the name, after all.
But I figured, people surf around 40~60 sites daily, on an average. Right? So, is it really feasible for someone to remember the dates his bookmarked articles were created? Moreover, is it really necessary for him to remember them? Thanks to bookmark managers, no-one remembers the URI anymore. The URI’s serve the simple function of filling up the address bar, nowadays.
Hence, I decided to go with a clutterfree address bar for both of my readers. I thought about using /blog/post-id, but I found the word “blog” everywhere. Literally. So, I went with the simple, unobstrutive, easy to use /article/post-id
March 14th, 2006 at 12:02 pm
So, those links to your posts (especially search engine cache) will lead to 404 now.
March 14th, 2006 at 9:41 pm
Well, yes. Of course, this’ll also mean that my ranking will probably suffer. But, what’s to be done has to be done.
March 15th, 2006 at 6:20 am
However, Mathias is going to use 301 to redirect the pages. You may contact him on how he goes about with it.
March 15th, 2006 at 8:53 am
I do know how to redirect using 301. However, I just prefer not to, as I am trying to go a different way with the content of this site.
April 16th, 2006 at 9:54 pm
Of course you should redirect using 301, permanent. URIs should be permanent and unique, and if you move something you should point it to the new place.
You say “archives” relates to a blog, but I really don’t see how that is a direct relation. A general news site has archives. Infact anything with date-specific posts has an archives section.
I think they URL scheme which you’ve changed to is a lot worse. First of all, you shouldn’t expose the database unique identifier to the public. Also, just using a number remove the ‘nice-url’ ethos which should also be ‘hackable’. Tom Coates talks about this, and I certainly agree.
I think the /yyyy/mm/dd/article is a perfect URI scheme (the /archives/ bit isn’t needed, fair enough) and it certainly isn’t “clunky”. And it’s hackable, you can remove the ‘article’ and get posts under that day (usually just the one on many blogs, but removing the day should show an archive of the month. And finally, an archive of that year, whether it be a simple list or pagnation extracts.