I too care very little about MMS, but that's always been the beauty of Maemo. The diversity of tablet usage patterns quite frankly blows my mind, and I could go on about this, but I won't.

Anyway, let's say a decent implementation of MMS happens relatively soon, imagine someone records a video and wants to send it to one of their contacts. With possible further community development, the delivery mechanism of that video could be based on that contact's address profile(s). And furthermore, if MMS is one of the delivery protocols for that contact, the same community software could automatically send multiple MMS messages based on the video's length. And so on, and so on...

[click]
video recording starts
[click]
video recording ends
[click]
user chooses "group(s)" and/or "contact(s)" to send to
[click]
video sent to youtube, to email, to MMS, and/or to ???. Profit.

The above won't happen out-of-the-box, and the above won't happen by one app. But it could happen in a timely fashion with a community effort.