Ihmo there isn't so much left to innovate in an email client and Thunderbird is already a solid client so personally this message won't deter me from using it
The alternatives -- including Thunderbird -- are bulky and poorly designed. I believe the wide migration to gmail is in no small part due to the extremely poor showing of e-mail clients on Linux and Windows.
No ads (and no need to try to avoid them). Nobody systematically reading all my mail to custom-tailer said ads.
Easier integration with client-side encryption tools to facilitate end-to-end encryption.
Better support for plugins and other modifications.
I'm sure others have variations on this theme.
I have several GMail addresses, and a Google Apps email address for work, and variously-hosted other email addresses, all managed from within Thunderbird (with several extensions found over the years).
I don't want to merge personal stuff with my work email; I don't want to merge my work email into a personal gmail account; and above all I don't want to do either of those things then find that some robot in the system has locked me out of my account.
I also really need to have full offline access, so I can work in spite of unreliable or infrequent net connections. I.e., in 20 minutes of internet access in a crowded coffeeshop, I can upload dozens of responses to yesterday's mail, and collect today's mail, and fit in a few rapid realtime messages, then go. I'm not often forced to work this way, but when I am this is a lifesaver.
Spam protection is a more serious issue, I think. Both mail servers and desktop clients do it. Mail servers, especially if they get feedback from clients, have a huge advantage though in that they can learn from all their users. The desktop client has to work with a single user's input; unless you implemented some kind of p2p collaborative approach (which would be really cool).
if Mozilla wanted to do "good things" and not "lets try to be another corporate company!", they' buy off postbox and release that open source, and keep the devs working on it.
They're abandoning TB because it sucks. Frankly. It's buggy, leaky, etc. No one really cared for it, for too long.
However -- at least for my primarily work and OSS-focused e-mail usage -- Sparrow's simplicity and focus is misaligned with my requirements and preferences.
There is interesting innovation going on, for instance <http://notmuchmail.org/>; (and satellite projects such as <https://github.com/teythoon/afew>; and <https://github.com/pazz/alot>.
Any further features for Thunderbird can be developed as an extension or as a fork (like, say, Postbox -- which is actually really more of an extension rather than a fork, as far as I know).