Re: MIME vs. DIME vs. ...

Larry Masinter <[email protected]> writes:
> But I was thinking of a different metric: the probability
> that a typical software engineer would use an incorrect
> algorithm. I think that probability is high -- maybe 50%.
> (Or at least, I've seen a lot of stupid MIME boundary
> generators.)
Sure, but once again this seems to me to argue for a
clarification/profile of MIME.

> Similarly, while fast string searching algorithms are
> well-known, they're hard to get right.
Actually, you don't really need a fast string searching
algorithm here, because the standard guarantees that the
boundary prefix may never appear in the boundary.

> So here's a use case for DIME: programmers
> are more likely to get it right!
OTOH, they have to reimplement from scratch, meaning they're
more likely to make programming errors that wouldn't have
happened if they'd used tested code.

-Ekr

--
[Eric Rescorla                                   [email protected]]
                http://www.rtfm.com/