W3C home > Mailing lists > Public > public-grddl-comments@w3.org > January to March 2013


From: Vipul S. Chawathe <Engineer@VipulSChawathe.ind.in>
Date: Thu, 3 Jan 2013 02:11:41 +0530
To: <public-grddl-comments@w3.org>
Message-ID: <000801cde929$96dd5130$c497f390$@VipulSChawathe.ind.in>
Use of profile attribute by grddl is oft compared in this thread with that
by microformats.

HTML pages are composite from text, images and so forth. The "spec" always
over-ruled highly respectable differences of opinion for user agent web
browsers to present data with comprehensible consistency.

By performing hands-on usage with microformats my conclusion was the
profiles it has deal only with the coupling of above (text, image, etc.)
types such that they are commonly referred within human semantics. So
multiple rel="profile" links are relevant to identify the presentation
entity. This has domain-specific implications such as marketers might look
through pages with hcards, whereas recruiters might apply css filter
highlighting hresumes.

Contrast this with gleaning resources, if a marketer is pestered by resume
importing for SPARQL being advertised by profile attribute then its spam.
XML implies meta-markup so XHTML calls for being verbose with xmlns. But
gleaning might be applicable to non-meta dialect where principle purpose is
different, like HTML might imply presentation designed to target human. I'm
afraid being verbose through profile serves unwarranted overhead. Hixie is
correct imho to remove profile from html, despite concern for microformat,
as profile is susceptible to promotional spam that's otherwise unrelated
with html's presentation goal.


Received on Wednesday, 2 January 2013 20:40:43 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 19:55:03 UTC