You're right that stand-alone frames are much easier to edit. In my experiments with Squeak JPEG movies, JPEG movie are generally 1.2 to 3 times larger than the original MPEG movie at similar quality levels. So MPEG is definitely preferable for compactness in final distribution.
Ed Lazowska (Chair of U. Washington CS department) just gave a talk here at Georgia Tech where he pointed out that CPU speed is doubling every eighteen months (with some arguments saying that there is maybe 10 years left on Moore's Law), but disk space is doubling every NINE months and the backbone bandwidth is doubling every SIX months, with no limits in sight for those. Ed argues that in 10 years, things like streaming video protocols will make no sense at all: The CPU will be the bottleneck, not disk or bandwidth.
If he's right, then a less-compressed but more-easily-accessed and more-easily-constructed format makes more sense in the long run.
Mark
-------------------------- Mark Guzdial : Georgia Tech : College of Computing : Atlanta, GA 30332-0280 Associate Professor - Learning Sciences & Technologies. Collaborative Software Lab - http://coweb.cc.gatech.edu/csl/ (404) 894-5618 : Fax (404) 894-0673 : guzdial@cc.gatech.edu http://www.cc.gatech.edu/gvu/people/Faculty/Mark.Guzdial.html