Google Base
If you were to design the World Wide Web all over again, what would you do?
I know what I would do. Every piece of information would be first based on structured data and only then transformed to a visual representation of the data (read: XML and XSL). Every piece of structured data would be tagged with meta information, indicating semantics beyond the trivial ones found directly in the data. Tags used in the metadata would also be structured.
And we would make it real easy for anyone to post their little pieces of value - content, goods, services - in a structured, metadata rich manner. And... we would make it easy for them to get compensated for the value that they provide, if they wish so.
No lost bits and broken links. A central system to track all these little pieces.
What a fine online world this would make. And imagine, how much easier it would be for a search engine to locate information in such a world, where instead of scanning meaningless HTML blurbs, it would instead be able to access structured, tagged and semantics-rich data.
Tim Berners-Lee's Semantic Web is one attempt to solve this. So far it didn't really work out.
Enter Google Base. Clicking on the link won't do much good, since it's already offline. It was online for a few hours only, just enough to generate lots of buzz and now-rare screen captures. Gotta love the Google PR machine.
Google Base is Google's database into which you can add all types of content. WeÂ’ll host your content and make it searchable online for free.
What can you put in the database? Some examples from Google docs:
Blogosphere-meet-eBay-meet-Ning-meet-My Web 2.0-meet-Flickr-meet...…
- Description of your party planning service
- Articles on current events from your website
- Listing of your used car for sale
- Database of protein structures
Read more about it, enjoy the hype at its height, and invent your own speculations:
Subscribe
3 comments:
Note: I didn't say Web 2.0 even once :)
Yes you did:
My Web 2.0
A problem with structured data is that it is not very flexible. You can of course transform a record of library information into a visual representation, but the actual "structure" of the world wide web is very loose. You would have to write XSLT style sheets for every possible form of content.
Strictly structured data would limit the possibilities of the content creator. And I think the WWW would not be what it is today: Of course it is a mess, a heap of "meaningless HTML blurbs", but that's what makes it so great.
I'm a computer scientist and of course I sometimes think that it would be much easier for us if the world would be structured in a well defined way. But that's not how the world actually looks like. Although it is a hard way. We have to design machines that can understand how humans work. It would be a bad idea to make the humans think like a machine.
In my opinion metadata like tags and rdf may be a good approach to solve this problem.
To answer your question: If you were to design the World Wide Web all over again, what would you do?
I would make a very strict HTML standard, where all invalid documents are rejected. Web browser could then be much lighter, faster and perhaps even more secure as they don't have to guess what a website may look like. That would be one step to a better WWW.
Henrik - thank you for your thoughtful comment. Some notes -
1. Generally I agree with you - tags & rdf go a long way toward solving these issue. See my post on this http://yanivg.blogspot.com/2005/09/go-tag-yourself.html.
2. There is a lot to be said for providing pre-defined structures ("templates") for people to use when submitting content. Note that the actual templates do not have to be defined centrally.
Post a Comment