I'm kind of wondering which comment you are answering, you seem to be angry at my post for some slight that I didn't even say in my comment (or even in the quote). If you have a chip on your shoulder about google, fine, but the things you seem to be upset about seem almost wholly unrelated with what I said.

I didn't say that those that quit are stupid or somehow inferior, and if you read that in the comment, I really don't know what to say about that. I don't think that is the case (except here and there, naturally) and that those engineers that left the company did so because they found an opportunity that fit them better. Turnover happens, and you can read some great deep meaning into it, I suppose, but sometimes people just stop being interested in a job after 2 or 3 years (or less)

I think that you are right in that you have to be persuasive (or political, for some definitions of political) to get new projects off the ground or to direct a project in some direction, but that's kind of academic, isn't it? Where isn't that true?

As to the great programmers getting trivial programming tasks: I'm certain this happens. Our readability guidelines almost require that people do something basic or trivial. Some people hate that.

Here's your quote again: "if you love algorithms, scaling problems and the rest, you are likely to find a group where you can be pretty happy, intellectually speaking. But, if you aren't into that kind of thing, it may not be the place for you."

The point is, the people who left Google did not do so because they weren't into algorithms! So why did you even bring up such a point when discussing attrition at the company? Almost every engineer Google hires likes this stuff and is good at it.

The reason I am so aggressive in decrying what you said is because it's both misleading and insulting to the people who left Google. Not only is your post at odds with what I know, but it's also at odds with the mailing lists posts people leaked to techcrunch.

Please show me one email in that whole sheaf in which a person says "I left Google because I didn't like working on algorithms and scaling problems."

So, look again at the words "If" and "may" and maybe you'll read that quote with less vitriol and see its not so different from what you are talking about. Also, note that nowhere in what I said did I say this is the only reason people leave.

I was relating my experience and trying to qualify it carefully, but you are choosing to read what you like and ignore the rest. Which is fine for an internet argument, I guess.

Actually, the way you read it almost exactly contradicts most of the reported reasons that the engineers who left did so. It seems far more likely and believable that you'r exactly wrong -- and that the engineers who are leaving google are leaving BECAUSE they like working on algorithms and solving scaling problems, and aren't able to get opportunities to do that at google.

Just for the record, since I think the feedback might be useful: I had the same read. The way you phrased things there did sound if not condescending, at least undeservedly elitist.

From what I've gathered from contacts of mine that have been through Google it is a good place to work, but it's good in that sense that big companies are good places to work, not because it's the computer science promised land.

When I hear ex-Googlers (or Mircosoft) talking about being in Google, it sounds a whole lot like what I felt in my 4 years at SAP, but that doesn't seem in tune with the recruiting rhetoric.

It isn't always a lack of technical fit if someone isn't interested in algorithms and scaling issues. What I am about to say is as a "devil's advocate": I, myself, deeply interested in these issues myself, to the point that I'd never want to work in a company that doesn't deal with these issues.

"Wanting to work on algorithms and scaling issues" is, ceteris parabis, a cultural issue. There's more to building a product than scalability and algorithm design: there's used experience, there's capturing requirements, there's the whole development lifecycle (testing, scoping). It may be true that people who are good at Computer Science may be on average more interested in algorithm design than in (say) requirements gathering or UI design -- but the converse doesn't follow (there are plenty of people who would prefer on the "simpler" issues of software development who are just as competent at algorithms).