- From: Anton Prowse <prowse@moonhenge.net>
- Date: Fri, 20 Jul 2012 18:23:31 +0200
- To: www-style@w3.org
- CC: fantasai <fantasai.lists@inkedblade.net>
On 20/07/2012 17:27, fantasai wrote: > On 07/20/2012 05:05 AM, Anton Prowse wrote: >> >> I do have a gripe about the following sentence in the ED: >> >> # Authors /must/ use ‘order’ only for visual, not logical, reordering >> # of content; style sheets that use ‘order’ to perform logical >> # reordering are non-conforming. >> >> The "must" and the threat of non-conformance are toothless tigers >> since how can a UA determine whether the author has used >> 'order' as an unwise alternative to logical reordering? My impression >> is that this sentence is intended to be an authoring >> recommendation. That's valuable, but it needs to be a note and it >> needs to get rid of the RFC2119 keywords and the >> non-conformance claim. > > Author conformance requirements don't need to be machine-checkable. > It's handy when they are, but they don't have to be. See, for example, > the author conformance criteria on the use of tables: > > http://www.whatwg.org/specs/web-apps/current-work/multipage/tabular-data.html#tabular-data Interesting, thanks. Somehow that example (tables in HTML) makes sense, because the document being declared as non-conformant is the same as the document where the non-conformant behaviour occurs. I find it odd to declare a stylesheet non-conformant based on a particular instance where it has been applied to a document, though. (That said, I accept that stylesheets and rulesets can be embedded into a document in various ways, thereby tying the two things together.) Ultimately, though, subjective things seem like undeserving candidates for a "must". Cheers, Anton Prowse http://dev.moonhenge.net
Received on Friday, 20 July 2012 16:24:01 UTC