- From: Sylvain Galineau <sylvaing@microsoft.com>
- Date: Thu, 27 May 2010 23:56:07 +0000
- To: Dave Crossland <dave@lab6.com>, Tal Leming <tal@typesupply.com>
- CC: Jonathan Kew <jfkthame@googlemail.com>, Vladimir Levantovsky <Vladimir.Levantovsky@monotypeimaging.com>, Christopher Slye <cslye@adobe.com>, "www-font@w3.org" <www-font@w3.org>, "public-webfonts-wg@w3.org" <public-webfonts-wg@w3.org>
I must admit I've stopped understanding the logic by which we should both specify a metadata format that people should respect but random XML can also be included because it is desirable and easy to render such random data; if the output thus produced is sufficient for all metadata and users then rendering arbitrary XML in a canned generic manner is all a UA ever needs to bother with as this will, by definition, also render whatever metadata format the spec cares to define in a reasonable manner. So either we specify a format or we don't. If we do specify one, I would like us to either: 1. Say that this schema is all browsers will render. Any extraneous unknown XML is ignored. 2. Define a clear extension mechanism for font vendors who want to add their own metadata, metadata that all browsers must render in addition to the elements specified by the spec. Options that include both a well-defined schema and the total absence of one are, imo, meaningless from a conformance standpoint unless, maybe, everyone here considers a raw View-Source as appropriate. If it is, then I have no objection. But if it is beneficial for font vendors that the metadata be accessible to users in a processed form distinct from its raw format, if UAs have no way of telling which XML is more or less beneficial to the user - and they don't - then UAs really should write a generic XML rendering control. At which point we as a WG are wasting our time defining a format that every browser can and will treat as just any other piece of XML.
Received on Thursday, 27 May 2010 23:56:43 UTC