> He knew his stuff and could write and read English well,

Right, I really don't have any way to assess that of a grad student I meet in the halls if they can't speak to me. They may be brilliant, they may be the best student there, and be totally walled off from everyone who doesn't speak their native tongue. I feel bad for them... but I don't think they should be in a grad school where English is the lingua franca.

I would vastly prefer "adequate English proficiency" be a hard criteria for attending a US graduate school. Not only are grad students part of a cohort themselves, they often are TAs and provide assistance and guidance to undergraduates. English adequacy is generally the de facto requirement to adequately assist others who are in college (in the US).

It is a requirement, it's just that apparently the test of English as a foreign language (TOEFL?) either are too easily cheated or aren't adequately checking for that. I should add I knew many excellent students in grad school, from dozens of countries, and good and bad us students :-) I never achieved any fluency in other languages, it would be incredibly hard to learn Mandarian or another non-romance language,and I appreciate the challenge for them.

At my grad school, there were students from two countries that there seemed to be widespread believe that there was widespread cheating. This included (not really cheating things) like copies of all tests given by any professor in the past. I didn't believe it at first, but then someone from said country showed it to me.

It's hard to believe if you really were cheating, how could you make it through a bachelors in CS? Someone should notice, right? I was a TA of a class where the prof eventually figured out one student was copying from another student. He gave him an F.

I have same problem. :-) But I am developer, so I can just avoid direct contact with customer and always use text for communication.