Having written a site alone a couple of days ago, I’ve realised I have to update my opinion. It looks like, since I last seriously searched for personal sites some time ago, that search engine results have significantly improved. Then, I only got junk. Now, in amongst the marketeers’ screeching egos, there are some real results.

image: parc

Bing, and its hangers on at yahoo and the duck, returned one set of results; Google another. Beyond the specific pages returned, there was nothing much to distinguish between each set of results. Both had a similar odd mix of genuine home pages with the same unexpected things in common. My preconceptions were way off: I expected but did not get certain common properties amongst the pages returned:

  • They in no way represent a cross section of the population.
  • They are not contemporary. One of the earliest results states it was last updated in 1998. Indeed, another was made with Front Page, which I had thought had been abandoned when the Germans invaded Poland in 1939.
  • Some were very basic, probably handwritten in HTML. Others were produced by tools, such as Word. The simplicity isn’t surprising—these are personal pages—but I was a little underwhelmed when I understood that every page used a stock design. There was no flair.

The results do have the obvious thing in common: I searched with English words on an English language search page, and got anglophone results. But they also have two other things in common that really surprised me:

  • The discovered home pages were all very clean, technically. Some were basic, some were generic, none were beautiful, but all were freshly washed & ironed, written in perfect HTML. This is unsurprising for generated pages, but it is impressive for the hand–coded.
  • All the home pages found were for professors at élite universities, professors of the sciences at that. Unsurprisingly, computer science did very well, thank you, but some other sciences, such as biology, were represented too.

Clearly, the fact that all the home pages returned in both results sets belonged to professors shows those results are strongly curated. I don’t know whether they’re curated by man or machine, although it’s probably both, but I can think of no other reason for the skew to clean–coding professors.

So what does this tell me to do with arts & ego? Well, those professors were genuine, at least by my cursory search, so there’s no much point in pretending to be an ace academic at an invented university. I can’t compete.

I need to find my level, in search engine terms: I need to know where the site’s been plonked before I can plan a route to somewhere better. Similarly, I need to better understand search engine curation. Perhaps I can start to search for particular types of home pages, for example poet’s home pages, and see just how skewed those results are. Eeek—there are two kinds—professors and web horrors!

I’ve more to do, and so have the search engines.