[Disclaimer of bitterness: I'm not one of the smartest people in the world. I get rejected after one to three interviews all the time. The author of the post is right: nobody ever looks at your online code and your resume only gets read once by HR to set up the initial pre-interview.]
Testing for concrete intro-to-CS level drivel may give you a company full of people who spend 30 hours a week doing topcoder or project euler, but... you don't need that (if you do need that, good for you, but most companies don't).
It also tricks the interviewee a bit -- "Oh, they're asking algorithms questions! Maybe this'll be a fun CS research job." [six months into the job] "Uh... all I've done is write SQL and fully library-driven backend website code. Boring."
Being able to test the ability of somebody in algorithms to a high ability level is a useful late-stage signal.
I always thought it would be interesting to just ask "What do you want to be interviewed on?" and to tease out their strengths from this. You'd then be able to question away to whatever depth necessary without worrying that they were being forced to dance in another man's frame.
I guess the problem with that would be that the interviewer needs to know the area he is testing an interviewee on in order that he may judge it accurately. The communal ground which is intellectually testing is often the academic experience they (possibly) shared.
I also work for a company that does these types of interviews. While I wholeheartedly agree that asking these algorithms/data structures or other CS fundamental questions is not perfect, precisely because it may induce quite a bit of false negatives, the engineers who does end up successfully completing the process at my company are very competent ones. So from an anecdotal perspective, the process works fairly well in the fact that it does not produce too many false positives.