A few months ago I hacked (I mean literally hacked, it ain't pretty :) up a little GreaseMonkey script called TripIt For GMail. The idea was to play around with the concept of embedding TripIt functionality inside a mail client. Since the GMail team are supporters of GreaseMonkey and have created an "API" for GreaseMonkey scripts within the GMail UI I decided to start there.
The script is extremely simple. All it does is embed an 'Add to TripIt' button in the right hand column of the GMail web UI.
Currently the script does not even attempt to decide whether or not the email you are looking at is a travel confirm (that's something one could imagine adding pretty easily). Also, because it's just a GreaseMonkey script and not a true application running inside of GMail it doesn't have access to the source of the mail message and therefore needs to automate the actions of a user clicking on the 'Forward' button, typing in firstname.lastname@example.org and hitting 'Send'. Clever, but not how I really want to build this app.
I haven't had much time to work on the script since January as I've been pretty busy but I did just post it to UserScripts here for anyone who is interested in playing around with it. It's only been tested on Firefox 2 on Linux, Windows, and Mac so if you're using FF3 or FF1 your mileage may vary.
More important than the actual implementation of this script is the direction I see email clients and platforms heading. Imagine an email system that is smart enough to detect what you were looking at and automatically do intelligent stuff for you. In the use case that is currently near and dear to my heart the system figures out that the email you are looking at (or maybe you don't even have to look at it :) is a travel confirmation, sends the source of that email to TripIt (ideally via a RESTful API authenticated via oAuth) so that a beautifully formatted itinerary gets created for you. Take that one step further and imagine the email platform re-consuming that data from TripIt in any one of a number of forms (e.g. iCalendar, RSS/Atom, API, etc...) so that all of it is re-integrated into your suite of productivity web/desktop apps in semantically intelligent ways. The system should do all the work for you so you can focus on where you're going and what you have to accomplish once you get there (i.e. your "real" work).
In the travel space TripIt has been driving the use of email as a critical component of what I humbly think is a very useful social travel organizer for folks who travel. Beyond travel however, there are tons of useful applications (both social *and* non-social) for this kind of intelligent platform and trying to inspire folks to create these applications was a focus of my talk at the Web 2.0 Expo here in San Francisco.
I think the message is getting through and I think folks are finally realizing that email always has been and continues to be a great interface to systems, humans, and your data. Folks such as Boris Veldhuijzen van Zanten over at NextWeb, Jake Kuramoto at Oracle, and Mark Hendrickson at TechCrunch understand and have talked about the possibilities.
The time is right, and long overdue, for intelligent apps built on top of good old fashioned, ubiquitous, and highly scalable email to begin seeing the light of day. In the coming months I hope we start to see companies like Yahoo and Google open up their mail platforms and turn them into application platforms. We've already seen Yahoo begin opening up their search results page with SearchMonkey and the best information is that he rest of Yahoo, including mail, is going in that direction as well.
In smaller ways it appears as if this trend is already beginning with the beta launch of webmail provider Zenbe. I haven't actually seen the Zenbe yet but if it works as advertised, this is a great model for where mail platforms should be going. There are also companies such as Xobni who are nicely positioned to chip away at this problem from the client side of what is probably the most widely deployed email platform in the world (at least in corporate settings), Outlook/Exchange. It would be nice if Microsoft's Exchange group were focused on this but so far I've seen no indication from the sleepy giant in Redmond that they are thinking along these lines. Of course there are a lot of smart folks up there so I wouldn't be surprised if someone is trying to fight what must be a significant amount of corporate inertia to get this done, who knows. It's definitely a huge opportunity for them given their dominant market position in corporate email infrastructure.
I can't wait to see this continue to unfold, it's going to be an interesting year!