Some very interesting responses to my CLR editorial today including some good details that I haven't seen elsewhere. I'm still thinking that we're missing a lot of examples that might convince people to give the CLR a try, or implement it in their applications and I'm going to try and contact some posters to see if they'll write a bit more.
In the meantime, it seems that most people are still down on the CLR.That's definitely my first reaction, but I'm less concerned, especially as I hear about some small successes.
I'm sure that some people will make mistakes, and it's a little scary, but I also see many people implement poorly written T-SQL code all the time. It doesn't do well, so is it really that bad an idea to try and write something with the CLR? Wouldn't it make sense to implement some trivial, perhaps string splitting routines, in the CLR to gain some practice?
I'll draw an analogy to the building trades since we seem to like doing that. If someone offered me a nail gun, I'd be slower at using it than a hammer initially. There's a certain amount of skill that I'd need to build up before I could easily, and quickly, start using the nail gun in an efficient manner, at which point you'd be sure that I'd be more productive.
I'm not sure if the CLR falls into that category, but I do wonder if it's worth investing some time in it to see what can be done, build some skills, and then see if can be something you can use.