Microformats panel at W3C Plenary Day
Session 3: Microformats
Description: Leveraging the widespread adoption and understanding of CSS and semantic (X)HTML among web designers and publishers, microformats are a set of simple, practical, open data formats that are designed for humans first and machines second. Microformats are designed by researching and adapting to current human web publishing behaviors and usage patterns, and then reusing bits of existing widely adopted standards. This session will demonstrate the capabilities that have been quickly developed with microformats on today's Web, review current microformats, and open discussion on what microformats and related efforts mean for the future of the Web..
Moderators: Tantek Çelik (Technorati) and Don Connolly (W3C)
Presenters and Topics:
- Ian Hickson (Google) - "A billion documents and no semantics anywhere"
- Tantek Çelik (Technorati) - "What are microformats?"
- Håkon Wium Lie (Opera) - "Cascading Markup Languages — boom!"
- Rohit Khare (CommerceNet) - "Where Angle Brackets Fear to Tread"
- Dan Connolly (W3C) - "Microformats for practical Semantic Web deployment"
"A billion documents and no semantics anywhere"
The "semantics" in the title refers to the "semantics" given to the Web by HTML, as opposed to the "semantics" that are actually put in the Web by the humans. Microformats can document existing practices (especially based on the data from studies of existing documents) so that the existing content can be given "official" "semantics".
Basically my talk will be something like "ooo look pretty shiny billion documents lots of data that shows people use HTML not like it was intended but we can reverse engineer microformats out of existing practices to obtain useful information".
"What are microformats?"
How did we get here? Picking up (hopefully) where the previous talk left off, a brief review of recent web authoring trends, a timeline of the ideas that formed microformats, and demonstrations how hCard and hCalendar are being put to good use today by publishers, browsers, and end users alike.
Håkon Wium Lie
"Cascading Markup Languages — boom!"
Cascading in CSS allows descigners to not have to provide the full presentation of a document, they only need to supply the deltas from the html default style sheet; cascading does the rest. In the same way, microformats builds on the (admittantly shallow) semantics of HTML; you don't have to create a new laguage to describe more semantics, you just provide the deltas on top of HTML.
This is one of the design principles that CSS and microformat share, others inlcude: simplicity, author-friendliness, an evolutionary approach.
Boom will be provided as an example.
"Where Angle Brackets Fear to Tread"
XML has been a wild success almost everywhere in the information technology universe *except* for adding semantically-rich information to ordinary Web pages. Once upon a time, we were supposed to wish for
<price> elements sprinkled about our HTML -- why is are we supposed to be so much more excited to see class="price" this time around? Dr. Khare will speak about his experiences with both xml and microformats in light of a recent project, hListing (for classified ads), and writing a parser and search engine for microformats in general.
"Microformats for practical Semantic Web deployment"
Microformats provide just enough structure to use semantic web technologies like RDF, OWL, and SPARQL on data in ordinary web pages. Dan will relate a few case studies and demonstrate some tools.