If it’s about the users, then it’s about independence, stupid.
Name the four or five online services/web sites you use daily. I bet among others you’ll find Flickr, Yahoo!, Google, blogspot and wordpress.com (for those blogging/publishing online) in most peoples’ lists and in some, the most ridiculous of them all, Twitter.
What’s wrong with all those sites? They’re centralised. And for no good reason. It’s not like they are providing original information, in the way the BBC, CNN or blogs do. If Web 2.0 finally gets a meaning then according to many it’s probably user-generated content, social media etc. Yet those very same services that more or less represent the leading examples of Web 2.0 technology and phenomenon have — ironically — nothing to do with users, but with corporations.
And that’s the problem. The technology is there to do amazing stuff with the Web, to provide more information, in medium-independent ways, with rich meta-data and interactive presentations. And all that could be done today with open technologies. Yet something stops us: We’ve got stuff like W3Cs RDF, but we opt to thoroughly abuse RSS creating yet another mess. We could already have a comprehensive open RIA framework based on Javascript, XML and SVG, but we rely on proprietary Flash and Flex and the kludge that is AJAX. We could have fully tagged information and a truly decentralised web, hyperlinked to the hilt and scoured by search engines, yet we rely on a few centralised databases while some companies make millions billions.
Take for example Flickr or YouTube: web sites that hold media. These are web sites that typically allow you to specify your friends and let them see some of your data, while others are publicly viewable. Something like Facebook, but limited to media. They also allow tagging. Suppose you’ve already spent many hours uploading and tagging images on Flickr and that tomorrow Flickr decides to charge for its service because it’s not doing well or you wish to migrate to another provider by ImageCorp Inc. Can you do it? Now, you might want to move all your images to your brand new web site under your own personal domain JoeSixpack.info. You may want to accompany the information with text (a blog entry) or video, tag everything properly, and add all the meta-data you want so that a search engine could spider everything later.
In another example, consider Facebook. How do you feel letting a company know of all your personal information for free, when they can easily change their terms of service and decide to sell your information later. Have you properly read the ToS before using it? And, more to the point, do we need Facebook considering we could use FOAF in our own web sites in so many different ways? So it boils down to services providing space and letting the users free to choose what to do with it.
Chaos? You bet. And that’s the point. Individual, unique presentations, innovative, decentralised, independent web sites, linking to each other, providing very rich meta-data for spiders and search engines. The Semantic Web. RDF finally used for something more than FOAF, DOAC, DOAP, SIOC and other currently irrelevant four letter acronyms.
Decentralisation (also) means Profit
‘Profit’ for Everyone. The need to decentralise is not at all about freedom and independence at the cost of profit. It’s about freedom and independence, period. And, sure, a decentralised Semantic Web would probably diminish the importance of places like Flickr or YouTube as they stand today, but it’d still mean big money for more people. It just requires a different way of making it. Search engines could still make money. Ad providers could still make money. Even a service like Flickr could be part of it and be profitable. A decentralised Semantic Web would mean that while you could choose to have Flickr as your service provider, there’d be no restrictions in place prohibiting you from taking your data from Flickr to Googlr or Microsoftr or cosmixr if you wanted to. It’d mean you could setup your own domain and web site and keep your data + meta-data there.
We need the Software
Back in 2000 I spent one night writing a file-based photo gallery php script, mostly to play around with php. I used the program for two years before ditching it when my university changed its policy on research associate pages and forcing me to remove all non-academic material from my site. Since then I’ve been meaning to rewrite it, properly, so that it provides RDF meta-data of the pictorial information available. I started doing so about a year ago using RoR (again in order to gain a better understanding of the language and framework better). Perhaps this time when it’s completed later this year it’ll be ready for release.
We need good, easy to use software for people to use and we need to make sure that our software leverages existing technology in a way that allows the Semantic Web to come to life. That would really be Web 2.0. Until then, all those flashy fades, scrolls, AJAX/Flash displays you keep seeing, all this noise about centralised services like Digg, Twitter, Flickr, etc. leave me cold. It’s just Web 1.01 if you ask me. Open data + meta-data is the future of any service provider and RDF + XML is probably the means to do so. Google’s Eric Schmidt seems to understand this need for openness. The days of centralised services may not be numbered (people will always need service providers for one thing or another), but I hope that the days when centralised, incompatible services where the user cannot retrieve his data in interpretable form are soon to be over.