A possible Game Changer
Posted: Tue Jun 07, 2011 6:17 pmThere have been a lot of attempts to add machine readable meaning to web pages. Microformats, RDF, those EDI schema, etc. (At the same time there has been a counter tendency to publish a lot of content with little, no, or negative meaning. 
)
These efforts are pretty much still born. That might now change, for a small part of the web. Google, Microsoft and Yahoo have agreed on formats that search engines will use. That extra incentive (one ring to rule them all) might just make this stuff more popular. See http://j.mp/lBRHbG for a brief summary. This is riding on the back of HTML 5.
The approach is interesting see http://j.mp/miniRW for some detail. It's intended not to break browsers and it's extensible if you have your own requirements. (Though the search engines probably won't understand what you're doing to any significant degree!) If you look through the base types and the existing schema you'll no doubt see some that don't look right, but it is still genuinely useful.
A lot of web writers are going to be unaware of this for a long time, and many of those that do understand it are not going to bother (in the short term at least). There will be a group that embraces and uses this. They will need ways to make sure that their non visible markup is right.
I'm not sure what impact that will have on CSE design etc. However I'd like to suggest that whatever is done, an extensible mechanism is a good idea. Schema.org has an extension technique, our validators should have something analogous. That way anybody who creates these frameworks-for-meaning will be encouraged, and be able to use them without too much hassle.
One pipe dream level hope is that a uniform mechanism is available. With such a thing I could develop my own schema, including a definition (with exception messages defined!) and plug that into CSE and any other tool I use that validates markup. I'm not holding my breath, but that would be an example of the software world actually working well, for a change, if it happened!
)These efforts are pretty much still born. That might now change, for a small part of the web. Google, Microsoft and Yahoo have agreed on formats that search engines will use. That extra incentive (one ring to rule them all) might just make this stuff more popular. See http://j.mp/lBRHbG for a brief summary. This is riding on the back of HTML 5.
The approach is interesting see http://j.mp/miniRW for some detail. It's intended not to break browsers and it's extensible if you have your own requirements. (Though the search engines probably won't understand what you're doing to any significant degree!) If you look through the base types and the existing schema you'll no doubt see some that don't look right, but it is still genuinely useful.
A lot of web writers are going to be unaware of this for a long time, and many of those that do understand it are not going to bother (in the short term at least). There will be a group that embraces and uses this. They will need ways to make sure that their non visible markup is right.
I'm not sure what impact that will have on CSE design etc. However I'd like to suggest that whatever is done, an extensible mechanism is a good idea. Schema.org has an extension technique, our validators should have something analogous. That way anybody who creates these frameworks-for-meaning will be encouraged, and be able to use them without too much hassle.
One pipe dream level hope is that a uniform mechanism is available. With such a thing I could develop my own schema, including a definition (with exception messages defined!) and plug that into CSE and any other tool I use that validates markup. I'm not holding my breath, but that would be an example of the software world actually working well, for a change, if it happened!