- From: Jukka K. Korpela <jkorpela@cs.tut.fi>
- Date: Wed, 5 Nov 2003 23:59:58 +0200 (EET)
- To: www-validator@w3.org
On Wed, 5 Nov 2003, Emlyn Addison wrote: > Yet another XHTML validation snafu on the road to spiritual creaminess... I would appreciate it if people got to the point, without trying to be poetic. > I'm sure that I'm not the only designer out there who has discovered > that opening <form> and closing </form> tags have a nasty habit of > adding vertical space. There's nothing nasty about it. And it has nothing to do with markup validation. > This is especially difficult to deal with when the form has been > carefully constructed within a table for layout purposes You have carefully constructed yourself a problem, alright. > My trusty workaround to this predictable HTML snafu has been to open > <form> and </form> tags *outside* of a table's structure You misspelled "unreliable hack". And from your description, it is obvious that you are intentionally using invalid markup. So what are you complaining about when you ask a validator to check and report the validity of your markup, and it does exactly that? > Does the XHTML validator honestly expect an entire block-level form to > appear within a single table cell, or none at all to be considered > valid? Or am I just missing something horribly obvious? You are. A validator is neither honest nor dishonest. It simply does what you ask it to do. The question is: why do you use a validator in the first place, if you _don't_ want reports from invalid constructs? Suggested reading on the myths and facts of validation: http://www.cs.tut.fi/~jkorpela/html/validation.html Regarding the extra spacing, there is a very simple solution: put form { margin: 0; } into your style sheet. If you are worried about the small amount of browsers that don't support this yet, then it's not a validation problem, and then the real problem is your page design that relies on details of layout, instead of being flexible. -- Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/
Received on Wednesday, 5 November 2003 16:59:59 UTC