RE: SIMILE Research Drivers

> [Andrew Fitzhugh]
> ... it may just seem quaint that we were once so enamored of clicking on
> hyperlinks in HTML documents ...
>
> [wmj]
> A *crude model* would be: selecting named-pages/screens one at a time.
>
> [Andrew Fitzhugh]
> ... in my own work the vision of future information networks is more
> decentralized than today's web.
>
> [wmj]
> How your "vision" is different from *crude model*?

I'm afraid this might be getting a bit off topic so I'll try to stay brief.

When there is semantic markup, links may be authored implicitly or
automatically.  A photo marked with contextual information such as time and
location may be published with no explicit hyperlink.  When I see the photo,
an application could, however, offer me access (via some sort of navigation)
to content that is related in time and/or location.  In addition, the links
offered may depend on my local environment (museum, friend's home, train,
...).

My group (Nomadic Computing Department) also focuses on bridging the
physical and virtual worlds, where information access is accomplished via
non-browser mechanisms, often through physical interactions.

The constrained browser navigation model and highly centralized services
aren't well suited to these scenarios, though I've always thought SIMILE and
its precursors were more general and quite relevant.

-- Andy

Received on Tuesday, 8 April 2003 14:38:51 UTC