It's great to see the newly named Mozilla Messaging moving along. As I've said before, this is a tremendous opportunity, but as many people have found out, one that is incredibly challenging. Well, not challenging to do something, but challenging to really effect a significant change.
One thing that I think puts it in a better position than most to be successful is the openness. Not so much the open source aspect of it, though that contributes. But more the open architecture. You look at the number of plugins and add-ons for Firefox, and then you think... how many different ideas do people really have about web browsing? But email... ahh, you want to talk about different approaches and different needs, now there is a domain for you!
The traditional self-contained / monolithic mail application has either aimed to tackle one or two things, or tried to do a bit of everything. Neither approach works very well, for the obvious reasons. Doing an email application this way inevitably leads to the "Microsoft Word" problem - everyone uses only 20% of Word's features, but everyone has a different 20%. What to include? What to exclude?
With a plugin approach, there is the potential to get around this. But it's really hard. At the technical level you need to provide the deep roots that make a truly wide variety of plugins possible, allowing you to add new functionality and new user experience. But that challenge I think pales in comparison with the user experience for plugins. How do you know they're there? How do you find the right ones? How do you manage them? How, in effect, do you make your own personal email client, or even better, how does the right email client for you end up in front of you, by whatever means? Firefox's success with plugins is a good start, but email is going to need a few more layers of magic happening to make it truly pay off.
Starting with an existing product (Thunderbird) may make this a bit difficult, because you're always pulling against existing code... "but, but, but I don't want to break these 500 plugins that are out there". But it's also the only way to keep your feet on the ground enough to get something done. Way too easy to keep going up yet another level of abstraction and away from code in this domain. Having good UX people involved and with a significant role is hugely important.
In any case, glad to see this soft launch, and glad to hear that there have been many people who've been energized by this and have committed themselves to make it happen. In many ways, it can be very tempting...
I agree wholeheartedly.
Let me know when you first extension is ready! =)
Posted by: David Ascher | February 19, 2008 at 09:53 AM