Ed Juskevicius from Nortel asked Glorianna the following question:
"How much bandwidth will early adopters/pioneers of camera phones need (and how will that need to change with time) to enable pictures and then (later) movies of what I am experiencing (while I walk around) to be shared with other people over my wireless connection? Today, I heard it said that existing technology makes the sending of a video take a long, (too) long time. This is, I believe, a function of available bandwidth over the airwaves. So, how much will be needed (i.e. what will be the minimum amount) to fuel the next cycle of technology adoption/innovation/adoption, etc?"

==========
Glorianna asked us to answer. Here's my take:

The poetic answer is "as much as is needed to make the experience of making a story with a phone be no different from making a story with a canvas and a brush - you find the paint, you place a stroke, and you see the result. so think of story with the phones as many people painting"

The applied answer is "in any technological innovation that tries to present a new interface to an existing media form (e.g. phones being used as devices for making audiovisual stories) there are two ways of going about it: one is to try and give the users a maximum amount of bandwidth hoping that it will help them ignore the inconvenience of the new medium; the other way is to accentuate the imperfect nature of the technology thus allowing for a different type of end artifact to emerge (i.e. the stories you are to make with the cell phone are to be inherently and uniquely different from the stories that one makes with an ordinary camera). Assuming the latter as our objective (since the unlimited wireless bandwidth is not yet feasible), three conditions seem to be reasonable:
(1) amount of bandwidth at the minimum is to be in an inverse proportion to the amount of interface interaction possible within the device. In other words, the more in-device processing is possible, the less demand for bandwidth is required. A more precise ratio, taking into account the speed of the cell's CPU, monitor clock, and most popular functions, can be derived if necessary.
(2) amount of bandwidth is in direct relation to the following axis: on the least-bandwidth side of the spectrum we have the use of a phone as a control mechanism; on the most-bandwidth side is the use of the phone as a media-gathering device. It's a scale; the closer you get to using the cell as a control device for a variety of media, the less bandwidth you will require. As we know, creative process is not only about creating your own but also changing someone's - and for that many narrow-bandwidth scenarios might be derived.
(3) i envision the biggest consumer of bandwidth in the cell networks of the future be not the content itself but the axillary traffic resultant from the group nature of creative interfaces. Call it the ICQ syndrome - when too many people are on the network, the mere pinging of the participants becomes a burden. This is significantly more so in diverse media applications where the amount of different types of exchange between audiovisual cell clients is bound to increase dramatically. Therefore, the third condition is: the amount of bandwidth is to be proportionally dependent on the amount of group interaction to be found within the cell media networks of tomorrow."