I also think it sort of waters down Computer Science to wedge it into engineering and treat it as such. Ultimately it's more about theory, formalisms, and mathematical/logical foundations than it is about "making it go". I think when we start treating a CS program like it's a EE/ECE program, then the CS program is impeded by getting too caught up in the nitty gritty. Don't get me wrong, I think the hardware portions are very important, but I like writing compilers. I'll learn as much about a system as I need to get a good compiler going, but the physical nature of a system generally are of no interest to me. They should be somebody else's job (who enjoys such things and derives pleasure in doing them [more power to them for it ;-)]).

I've replied to similar sentiments on HN a few times:

http://news.ycombinator.com/item?id=968013

http://news.ycombinator.com/item?id=1131606

http://news.ycombinator.com/item?id=690798

People who do systems research don't do much with theory and formalisms, but they are decidedly doing computer science.

that divide works both ways. for example, most of the cmpe students i know weren't exactly well versed in the innards of a compiler.

Is there any major, including CS, which reliably imparts upon its students a proper understanding of the innards of a compiler? It probably depends on the school.

I guess it's the same as in Argentina.

Here, you can choose either Computer Science at the exact sciences university or Informatic/Systems Engineering at the engineering university.

Both offer courses on algorithms, discrete mathemathics, and such. The thing is that to earn an engineering degree, you must sit for advanced physics, chemistry, economics and other complementary studies.