- From: L. David Baron <dbaron@dbaron.org>
- Date: Thu, 7 Oct 2004 17:11:37 -0700
- To: GEO <public-i18n-geo@w3.org>, www-international@w3.org
- Message-ID: <20041008001137.GA2686@darby.dbaron.org>
On Thursday 2004-10-07 16:48 -0700, L. David Baron wrote: > On Thursday 2004-10-07 15:52 -0700, L. David Baron wrote: > > > Richard Ishida wrote: > > > >http://www.w3.org/International/tests/sec-bidi-misc-1 > This actually probably wouldn't be hard to fix, assuming that the > windowing environment supports LRE/RLE and PDF. And I think I'd lean > towards agreeing that it is a bug. I should really get all my thoughts together before I respond. I think it may be a good thing to fix. However, I think doing so would be extending the HTML 4.01 specification, and claiming that failing the test is a bug would require errata to HTML 4.01. In particular, a literal interpretation of section 8.2 of HTML 4.01 [1], suggests (although the spec isn't entirely clear) that the dir attribute has an effect on block-level and inline-level elements. Although, again, the spec isn't entirely clear (but what's new!), the title element seems to me to be neither block-level nor inline-level [2]. Thus I think HTML 4.01 does not describe any mechanism for either specifying a base direction or an embedding for the title element (or for the contents of any attributes). -David [1] http://www.w3.org/TR/1999/REC-html401-19991224/struct/dirlang.html#h-8.2 [2] http://www.w3.org/TR/1999/REC-html401-19991224/struct/global.html#h-7.5.3 doesn't specify which elements are block-level or inline-level, but title is in neither the %block nor %inline productions in the DTD. -- L. David Baron <URL: http://dbaron.org/ >
Received on Friday, 8 October 2004 00:12:21 UTC