Re: Problems with the current user interface

I guess this relates to scoping as much as UI - do we consider
user agents on devices like phones to be in-scope or not?

(Sorry if I've missed where this is already covered, but a
quick wiki-flick didn't turn up an answer.)

If phones are in-scope, then a lot of the UI stuff gets even
harder, but, OTOH, it might force us to consider an abstraction
rather than getting bogged down in details of the usual desktop
GUI which could be good.

(If phones are in-scope, then what about fridges? We'd probably
want some kind of limit, so that we don't have to deal with
e.g. motes running TinyOS. I guess something like devices with
a generic web browser and with >N-million deployments might be
about right, with N~=10 or whatever covers blackberries/PDAs?)

If phones (etc.) are out of scope, then a) I think that's maybe
a bad idea, and b) we should probably include them later on,
after the initial work for desktops is done.

Note: I'm not saying that I think this group should develop
guidelines for phone security UIs, but rather that we should
not assume that the context information we consider will only
ever be displayed on a desktop, nor that preferences can
always be entered using a normal filesystem, keyboard, mouse,
etc.

I think we can however easily handle this for now, if we
just add a use-case like "same as use-case <<foo>>, but
with the client running on a mobile phone". Properly
handling such devices later on might be trickier of
course;-)

S.

Received on Monday, 11 December 2006 12:48:46 UTC