Discovering Magic

by Glenn Jones

10 Reader Comments

Back to the Article
  1. Great read, I’m a big believer in RDF and Microformatting. I was disappointed to see the Web moving away from XHTML back to HTML because of the flexibility RDFa provides in communicating and sharing data
    Copy & paste the code below to embed this comment.
  2. While I very much appreciate the technical aspects of this well-written article I do question the complete lack of reference to data protection and overindulgence in Web2.0-ness. As for microformats - CSS classes are not the place for semantic metadata. That just turns them into “keywords” and we all know what happened with meta-keywords. While there is definitely something to be gained by minor extensions and “namespaces are a honking great idea” and as a recent article showed RDF shows some of the possibilities, the semantic web is a dodo.
    Copy & paste the code below to embed this comment.
  3. Charlie I don’t agree that microformats are putting keywords into css classes. The HTML class attribute has many roles; it can be used to help add interaction to a page with JavaScript or presentation style with CSS. This is reflected in the “HTML 4 spec”:http://www.w3.org/TR/REC-html40/struct/global.html#h-7.5.2 which refers to the attribute being used “For general purpose processing by user agents”. The example given for general purpose processing:  “identifying fields when extracting data from HTML pages” describes exactly how microformats use the class attribute.  It is a common misconception that classes are just for CSS.

    I think I know where you are coming from with the meta-keywords comparison.  The reason why the use of metadata in HTML can fail is that it is often hidden from view. The meta-keywords are out of sight and because of that errors go unnoticed. Microformats don’t have the same problem as most of the data values are in clear sight.

    If I am honest I would say that building the semantic web is taking a lot longer than most would hope. There are real points of traction starting to shine through; “Google’s parsing of Microformats/RDFa”:http://googlewebmastercentral.blogspot.com/2009/05/introducing-rich-snippets.html in its search results being the most obvious example. I also hoped the demos with this article would go some way to showing that designers /developers can practically use semantic data today.  It may not yet be the grandiose vision, but it’s not a dodo.
    Copy & paste the code below to embed this comment.
  4. Glenn - if anything like the semantic web is to take off then any data attributes have to be unambiguous. Namespaces are great for this as can be seen by the extensions to RSS and I can see some advantage in doing something similar in HTML with RDF for unambiguously handling copyright, data protection or similar common requirements. But overloading HTML namespace really isn’t going to fly - to overextend my dodo metaphor. There was a recent article on “Opera Dev”:http://dev.opera.com/articles/view/styling-and-extracting-hcalendar/ about the calendar microformat which really made me think about the problems with the approach. Why on earth isn’t the ICS available directly? Or, pack use a namespace so that the browser can decide what do with the calendar information directly as happens with the date-formatting. Apart from the technology there are very important data protection reasons for letting the browser and the user decide what happens with this kind of information.
    Copy & paste the code below to embed this comment.
  5. Most people dont even realize their own spread on the internet and then there are some who seek to build on it. This is a nice way to consolidate all the data and measure the overall weight of web presence of an entity.
    Copy & paste the code below to embed this comment.
  6. As usual, _A List Apart_ seems to have read my mind again! ALA articles have a spooky habit of co-inciding with whatever I’m working on or thinking about at the moment… I was just working on turning my (currently rather bare) “personal home page(The personal website of Jordan Clark)”:http://www.jdclark.org/ into a collation of the various bits-and-bobs that are dispersed all over the web, then up pops this! @Glenn: Fair play, I must say that the “Ident Engine”:http://identengine.com/ is very impressive to say the least. It’s great for web designers to see the benefits of using semantic markup – along with microformats, RDF, RSS etc. – pay off with the emergence of tools like this. It also the practical benefits such as this that will encourage the further use of standards-compliant (X)HTML, as opposed to theoretical pipe-dreams. What we need now is an “inverse Ident” – some sort of centralized web service that does the same thing for online profiles as what “OpenId”:http://openid.net/ is trying to do for the log-in dilemma. (Please excuse me if something like this already exists!) I’ve always found it slightly annoying to have to create a profile on “Facebook(Jordan Clark’s profile on Facebook)”:http://www.facebook.com/clarky.y2k then an almost identical one on “Elance(Jordan Clark’s profile on Elance)”:http://xclarky.elance.com/ - and yet another for “LinkedIn(Jordan Clark’s profile on LinkedIn)”:http://www.linkedin.com/in/xclarky and… (I’m sure you get the idea!)
    Copy & paste the code below to embed this comment.
  7. I’m impressed, this is a great tool—a great tool many people should be frightened of. It shows how your own identity has become public online, what is especially interesting for people who don’t know what can happen to their personal data once they publish it.
    Copy & paste the code below to embed this comment.
  8. I love this tag, I think it’s a great tool and have used it in the past.  Every day more and more people are realizing how public the information they put onto the web is.  I have never really been afraid of what I’ve put online because none of it could ever come back to hurt me.
    Copy & paste the code below to embed this comment.
  9. I’ve been thinking about this stuff for awhile, first playing around with Google Social Graph/Profile search API and now with Ident’s.  Its very interesting stuff, and fills in the gaps to questions i’ve often had when following hashtags for conferences on twitter.  I’ve often wondered- who are all these people.  Leveraging these new API’s I’ve put together HashParty, a twitter hashtag explorer that does just that.  http://hashparty.com/ Give it a look see and tell me what you think.  The team and I working on allowing people to claim and clarify their id’s as well.  This is a rapid concept coming together in about 2 weeks after digesting what all I could find online to help us connect the dots.  In testing the concept, Google Reader/Profile id’s appear to mess up Indent the most, some users get incorrectly reported with having numerous ids for some strange reason.  Overall cool tech though and always trying to improve our results.
    Copy & paste the code below to embed this comment.
  10. Good points you got there Glenn… It’s really important and i gotta tell you, it’s an amazing tool…
    Copy & paste the code below to embed this comment.