Disclaimer: the views expressed on this site do not nessecarily reflect those of my employer
Sunday, November 17, 2002
Dave has this to say about the current state of RDF:
If you can't explain it to me so that I understand what you're doing -- you've got a big problem.
That may be true, but I know Radio has made more than one of us feel like a moron. Admitedly, I haven't tried to do so much with this stuff, but part of the reason for that is that I'm too busy and life is too short, and everytime I click on the "Help" link I can't find any. I think its probably not that uncommon. Even though we have all been soaking in computers for years, some people just get things differently than others. Ok, that being said, we all know its way too easy to create ambiguos, impossible to comprehend specs that are way too over engineered to allow anyone to use them effectively..
1:59:01 PM
Finally had a chance to read Joel's latest essay: The Law of Leaky Abstractions. While I agree with a lot of it, I'm not sure I buy the idea that because an abstraction might have a performance penalty in some cases, its a leaky abstraction. And there certainly are some that seem well cemented by now. For example, how much .asm have you written in the past 5 years? How about floating point assembly on the PIII? How many people can read assembler and translate to machine code? Ever since VS 6.0 SP3 I really haven't had to worry about how the compiler generates my code. There have been a few times when I was convinced that it was screwing something up, but 100% of those times it turned out to be me. And I don't know anyone who can hand code floating point Pentium assembler better than the compiler. We tried to be sure, but couldn't do any better.. And I hear VS C++ is about to get even better.
