- From: Tantek Çelik <tantek@cs.stanford.edu>
- Date: Mon, 05 Apr 2004 11:06:13 -0700
- To: Ian Hickson <ian@hixie.ch>, <Matthew.van.Eerde@hbinc.com>
- Cc: <www-style@w3.org>
On 4/5/04 10:46 AM, "Ian Hickson" <ian@hixie.ch> wrote: >> 3) Force the table-ness of things to be determined by the markup language. >> >> I like 3) myself. > > There are many situations where the table layout is purely a presentation > thing, and also some where tabular data needs to be rendered in > non-tabular ways. So that isn't a workable solution. I'm not sure I agree. Or rather, I think I lean towards agreeing with Matthew. I think there can be many examples where the table-ness of elements (i.e. their table semantics) MUST be determined by the language (e.g. say you were marking up a sparse matrix -- the mathematical kind), and thus it would be possible to create pseudo-classes that selected the semantic table-ness of those elements rather than their presentational table-ness. For example, consider how you would a speech interface would read aloud the various rows and columns of a sparse matrix. Clearly there is no visual table, and yet, the interface would likely still do something presentational (perhaps different pauses) between rows and columns in the matrix. Matrices in mathematics can also get quite large, and rather than forcing them to be rendered exactly as they mean (in as many visual cells as wide and tall as the matrix), it may make sense to render the cells one per "line", and use some other form of styling (perhaps different thickness borders) to distinguish column/row breaks in the matrix. Nonetheless, I don't think the idea of semantic table-ness (and corresponding pseudo-classes) should be dismissed so quickly. Tantek
Received on Monday, 5 April 2004 14:06:18 UTC