iGoogle
I recently learned about the ‘personalized homepage’ (read ‘Portal’) that Google has in beta [screenshot]. I won’t go into whether they should/should not be doing this but I will say that like all things Google, I was impressed with the coolness factor. I really like the simple clean lines and lack of clutter. The best part is re-arranging the various ‘portlets’: it’s as simple as ‘Drag-Drop’ the individual portlet into the correct space. Don’t like the Wired News at the bottom of the page? No problemo, just drag it to the top of the page and see everything slide over to make room. Really nice.
(Disclaimer: Non-geeks continue at your own risk)
Aside from the usability, iGoogle illustrates how distributed computing has come full circle from client-server to thin client to rich-client. The latest breed of Google offerings (Google Maps, igoogle, Gmail etc.) all use the Rich Internet Application/AJAX model to deliver data to your favourite browser.
Unlike ‘traditional’ web applications that require a reload of the entire page to refresh data, AJAX (short for ‘Asynchronous JavaScript and XML’) applications only get the data you requested without a refresh of the page. How do they do it you ask? Well, instead of using a Form submit, AJAX uses the ‘XmlHttpRequest’ object to request data over HTTP and then updates the page using its DOM (Document Object Model). This simple idea is potentially revolutionary and given the lowering costs of bandwith, the day may not be far when we see browser-based applications running at client-server speed.
You can read more about AJAX/RIA in this whitepaper.

