Surname Musings

Surname Musings

As part of my job as a systems manager, I’m often looking at long lists of names of people who are on the network. When I originally set the network up, I decided to put half the 1300+ users on one server, and half on the other, to reduce the load and stuff. My criteria on how the split was to be done was by surname.

And so it came to pass that surnames A-M are on one server, and surnames M-Z are on the other.

When it was first set up this way, I realised that it wasn’t a true 50-50 split. In fact, it was more like 60-40 in favour of A-M. If I’d have sat down and actually counted, I’d have known that, but it didn’t really matter. Of course, you can see why this is the case, as not so many people have surnames starting with letters like U, X and Z. However, the sheer number of Smiths have some sort of balancing factor, so 60-40 isn’t all that bad, and so I kept the system with this split.

That was almost five years ago. In that time, 95% of the users have left, and so compared to then, virtually all of the users have “replaced”. And here’s the odd thing – the split is still in favour of the A-M users, but bizarrely it’s more like 80-20. Not only that, but we now have twice as many Smiths as we had five years ago, and fewer Browns. I’m at a loss as to how to explain it.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.