With all the multitude of ways of storing infromation, both structured and un-structured, we still dont seem fully liberated even when we store what is essentially simple data-sets, lists, hierarchically-structured information... on the Web.
Our information, like bookmarks, to-do lists, recently listening to music, wishlists, birthdays, calendars are all sitting on different peoples websites....
Google has realised that there are many companies (Web 2.0 style) that are offering just that... But in reality... there are only a couple of ways to skin a cat...
And whilst people are getting excited that Google Base is supposed to kill e-bay, we just need to realise that there's just a bunch of journalists and Marketing research Analysts out just getting a bit excited.
Its not about killing e-bay. (far from it)
But it is about thinking about solving a problem implemented many times before, in a simple and easy way and (as Garret Rogets over at ZDNET says...) about making money through search on this structured data.
Google wants to build a hosted database service, which relies on Google's highly scalable, fault-tolerant infrastructure, to provide an easy way for users to create their own data. That is searchable through its Search engine, that can provide targetted ads to subscribers of that data.
Think about it, you could expose that data, using RSS/Web Services (cos its SOA world out there), and you can then skin your own application using AJAX or Ruby on Rails...
Google has spent most of the last few years trying to organise the worlds data, most of which has been unstructured in web sites. Google is trying to find ways to organise structured information in a way that has never been done before. But also to unlock data held in websites which have been closed to Googles spider, in short this is a great monetization opportunity for advertisers who will be able to aggregate better quality information about subscribers.
And what we will be witnessing are the first steps of unleashing a new paradigm for building data-centric web applications.