I've seen people who bookmark the google search results page. The search they have bookmarked is the URL they want. Not the company name, the complete URL.

These were very smart, world renowned people doing this. Just not computer smart.

This is a great point. I've heard raging arguments about URL rewriting (or routing, or whatever one wants to call the feature that creates "semantic" URLs) and why it's essential because users "insist on" friendly URLs. For many years, I was able to point to Amazon as having the world's ugliest URLs (possibly beat out only by SharePoint), with no appreciable effect on traffic. (They have recently tinyfied somewhat, so this is no longer the compelling case it once was.)

People just do not care, because one person in 100 (or a 1000, or make up your own number) actually types in, or even reads, a URL. They click links. The people who really care about semantic URLs are people who are designing RESTful Web services, and they actually have a reason to. The rest of us ... phhht.

I think it's worth mentioning that at time of writing, Readwriteweb's comment thread for the article in question is 28 pages long.

Thanks for posting this, as a developer it's quite easy to forget who we're really developing for and what they actually care about. Sometimes we get stuck in our own little idealistic worlds that don't really matter to the majority of our users :)

Do you think they feel lucky now?

Try this: go to the Google home page and in the search box type "http://a" and watch the autocomplete suggestions.