Plucker already handles some things of questionable acceptability. I don't think it's so terrible if it fails to handle some incorrect things and junks the occasional page that uses them. It's not like the entire download fails. Even ripping out the current parser and putting in something like HTMLPrag (no, I know that's not feasible for Plucker) wouldn't handle the case of some site failing to encode a URL correctly. The best fix is for the site to encode correctly.
"Best fix" depends on perspective. "Best fix" for the Plucker developer is for everyone to follow the documented specifications, but the documented specs have a loose correlation to real-world practices. "Best fix" for the web site authors is probably to do what they're doing; they aren't alienating many people since every major browser supports it fine (as I've already tested and proven.)
But those are both myopic views. "Best fix" for the Plucker user is for Plucker to work on whatever browsers work on, not for Plucker to refuse to work with things that all their browsers work with.
(Again, this is mostly philosophical for me. If I was bugged by it, I'd simply fix it like I did the things that did bug me last autumn. None of the sites I visit manifest this problem.)
_______________________________________________ plucker-list mailing list [EMAIL PROTECTED] http://lists.rubberchicken.org/mailman/listinfo/plucker-list
