Don't Back Down

As you certainly know if you've been reading for a while, I tend to introduce more and more categories over time. I shift things when I feel it is necessary to better get the information into the right place. These traits alone indicate that I am not someone who cares if things fit into the right places (such as a librarian, who is tied to the Dewey Decimal System to some degree).

You may also have read that I don't really care that much for "social" services all that much. I even set up my own Scuttle server for storing my bookmarks (in fact, I joined the development team). It just isn't something that I really use. That all changed - at least to some degree - when I read an excellent piece yesterday about how Ontology is Overrated.

Ontology is, loosely defined, the filing of things into their appropriate classifications. The problem is that the things that fit into my classification system may not fit into yours. An excellent example from the article is that some people may tag items as "movies", others "film" and still others "cinema". I'm sure there are many other such examples.

A point I've made before is that this makes it impossible for categorization to work, because I might be looking in "movies", while someone has filed their item in "film". The point I overlooked was that I may not care for the items filed in "film", and that someone, somewhere, is going to share my system, so using "movies" should be fine.

In the end, the index (in this case, del.icio.us) doesn't care which tags are which, or even that three (or more) tags represent the same exact information depending on who is using it, and that is the point. Using such a system allows people to connect with similar-minded people. Whether the index thinks you are of similar minds is irrelevant - it just shows you related information.

Still, I don't much like the del.icio.us interface, so I'll probably keep my Scuttle install running. And chances are that I'll probably keep adding and changing categories here (I have a few such alterations in mind, in fact). And while on the surface it seems a good idea, I probably won't do much in the realm of allowing users to tag particular entries - there are other services for that.

At least in MySQL, all you have to do to return random data from your query is add ORDER BY RAND() to your query, for instance:

      SELECT * FROM table ORDER BY RAND() LIMIT 1

Read "Random Rows in SQL" »

So it isn't really much of a brand. Yet. But I'm hoping that perhaps one day it will be the one-stop shop for all things Movable Type. Okay, maybe not all things. You'll still need to visit David and Tim from time to time as well.

But in the meantime, I figured it was time to add a little promotion to the company site, and I have done so with not only this entry, but the new site. While it's a bit sparse at the moment, it should serve to get the point(s) across.

Read "Building on the Brand" »