- From: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>
- Date: Fri, 31 Jan 2014 17:51:46 +0100
- To: Steve Faulkner <faulkner.steve@gmail.com>
- Cc: "Jukka K. Korpela" <jukka.k.korpela@kolumbus.fi>, HTMLWG WG <public-html@w3.org>
Steve Faulkner, Fri, 31 Jan 2014 15:52:06 +0000: > > On 31 January 2014 15:47, Leif Halvard Silli wrote: >> role="layout" > > you mean role=presentation right? Right. Isn’t <table role="presentation"> synonym for ”layout table”? > the advice in the spec has an normative MUST > > " If a table is to be used for layout it must be marked with the > attribute role="presentation"" Your initial letter was a little unclear to me. It indicated, to me, that you had edited the spec. But a change from SHOULD NOT to NOT RECOMMENDED is not really a change since, as you pointed out, it has the same meaning. So the following is valid both for the old and the (to be) proposed text: A MUST rule for role="presentation" does not prevent that it could be NOT RECOMMENDED/SHOULD NOT to use it,[*] and thus does not prevent conformance checkers from adding a warning, to the degree that role="presentation" is synonymous with ”layout table”. It would perhaps be odd to single out the table element as the sole (?) element for which role="presentation" should be 'NOT RECOMMENDED'. But it looks compatible with the NOT RECOMMENDED clause. However, I cannot ”use” this against your particular change it seem, since even HTML 5.0 says 'SHOULD NOT about tables used for layout. And thus I cannot see that your change would affect tools anymore, conformance wise, than the current spec text. [*] The border attribute on img element is one for which HTML5 has a MUST rule, if it is use. However conformance checkers warns against its use always, anyhow. -- leif halvard silli
Received on Friday, 31 January 2014 16:52:14 UTC