What I love about Google is they consistently think 'Web Native' when developing web applications. My favorite example of that is Gmail, which changed the whole paradigm of web email. And that's nothing against the new Yahoo Mail Beta, which has a desktop UI and is a very fine app too. But when you build a browser-based app, then why not optimize the design for the Web and introduce new UI paradigms that weren't possible in the desktop era? That may be the early adopter in me speaking, but the innovation you do now will be the norm in the long-term.
Carl Sjogreen said that the Google Calendar team focused "on what the web can do that paper can't" - e.g. collaboration and access anywhere.
...it's important to remember that with new technology comes new functionality. A term I use for this is 'Web native', meaning that the next generation of office software will not necessarily be the same as the past PC-based generation (typified by Microsoft Office). The new generation will have Web native functionality - including, but not limited to, collaboration. Rod Boothby likes to say that blogs and wikis are the first major 'office 2.0' apps, but I think a web-based suite will be so much more than publishing and collaboration features.
One new feature that I think will be common place is 'mashups', whereby data is sourced and combined from a variety of internal and external sources. Imagine an online spreadsheet for a marketing report where you gather data from all over the Web and across internal business units too.
There will be other Web native functionality too, things that we can't yet predict.
I referenced Zimbra, DabbleDB, Morfik, Zoho and ThinkFree as some companies that are building Web Native office applications. If, as Nick Carr suggested recently, office suites will be fully web-based by the early 2010s - then I'm betting that Web Native functionality will underpin this evolution.
Original link via Greg Linden