Catching up on indexing our resources
Archive - Originally posted on "The Horse's Mouth" - 2007-09-10 00:51:13 - Graham Ellis
The amount of material that gets added to this site in each month is incredible, and if I let it go much longer that that before catching up on indexing my blog articles, images, and forum posts the task can become almost unmanagable. A hectic time left me wonding how I would manage to get all these updated, important but rarely urgent, done; my car distaster and missed Pycon weekend gave me an unexpected window.The illustration here - and over 100 others - have been labelled up in our image database. You can see those images in six pages ... here, here, here, here, here and here. Source code examples - the bread and butter of our training business - have expanded to nearly 1800 examples (that's an increase of nearly 50), and the web site as a whole comprises 8994 pages - just another couple of days of blogging and we'll be over the 9000.
Is it important to have all these resources? And to index them? Yes and no; there are pages here that are rarely read and are forgotten in the mists of time within hours of being posted - lost never to be viewed by anyone as a useful tool. But them there are pages that are accessed hundreds of times a day, from all over the world ... and I can't predict which pages those will be. A frankly flippant piece on Bratton and Edington New Town has been picked up as a significant warning about getting the future development of this area correct, but a following piece on the security issues of hotel room keys appears to have sank without trace. It would be really great if I knew which way any particular post would go ahead of time!
But there's not only the question of having the resources there for people to find, but also having the other associated pages / examples / resources all indexed under the same banner too. And that's a lot of this current indexing job - helping people who land in almost the right place to take one or two steps and find all of exactly what they need. And this means that I look back at all my posts here and decide what topic they relate to. You may call the system old (it is), and crude (it is) ... but it's also effective. For example - you'll find all reources about Tcl's packages and resources indexed in a single location whether they were written in October 2004 or yesterday (9th September), and you'll find that the index page also links you to other associated Tcl/Tk pages such as those that deal with Geometry Managers.
For none technical readers, the same strategy applies too .. you'll find an around Melksham group of resources, and another bundle labelled Business accommodation in Melksham. You'll even find behind the scenes if you want to go back over some of my more revealing posts.