On mapping: where we are, where we're headed, and a starting use case

James William Yoon james.yoon at utoronto.ca
Wed May 27 21:27:07 UTC 2009

Hi all,

Some of the chatter around mapping, its possibilities, and what we're doing
with it has been a bit scattered, so I'm writing up this email to
consolidate some of that and give an overview of the landscape, as well as
provide a basis for some ongoing dialogue.

If we were to map out the space of our mapping ideas (pun unintended), I
think we might find that they lie on two axes: location awareness, and
spatial vs. conceptual.

Maps can be location-aware (map knowing the current location of the
map-user), or location-unaware (map oblivious to the current location of the
map-user). Note that it's not entirely clearcut; there's a spectrum, or
rather, varying degrees of awareness: maps could be acutely aware of the
user's precise position at all times (full tracking), aware of when the user
is in a particular zone (e.g., proximity to an object), aware of when the
user interacts with a particular object (e.g., tagging of an object), etc.

Additionally, maps can either be spatial (mapping the physical layout of a
space, as in a floor plan) or conceptual. Conceptual mapping, in our case,
largely involves relating objects (to each other, to concepts, etc.). This
could be, say, semantically (e.g., DRESS worn by a Belle used at The
Uber-Grand Gala of 1942, attended by photographer Bob, who photographed
Belle in this PHOTO), or otherwise (e.g., objects related to each other by
subject, geography, time period, etc.). It's probably most natural to think
of conceptual maps as node-link graphs (e.g., concept/mind-maps), but they
could conceivably take on other, more 'interesting' forms (e.g., arc
diagrams, treemaps, etc.).

Also, like location awareness, there's conceivably a spectrum of solutions
between spatial and conceptual mapping. One could imagine, for instance,
various ways that a spatial map might overlay the relationship of objects to
each other (e.g., a node-link graph overlaid upon the spatial map; or,
highlighting objects on the spatial map that belong to the same
period/theme/etc.; etc.).

So, our mapping ideas thus far lie somewhere on space like this:

                  spatial mapping

location-unaware --------+-------- location-aware

                 conceptual mapping

Another thing to consider is whether we are looking at maps from the desktop
or mobile experience. For instance (and most obviously), location-aware
mapping is a moot point in the context of the desktop experience.

All of this is maybe an unnecessarily longish preamble to talk about the
following simple use case:

Consider an individual who's returned from a museum visit. A few objects
caught her interest, and she'd like to learn more about them or revisit the
content. She goes to her desktop computer, opens up a browser to the
museum's site, and finds the interactive museum maps. The maps show the
exhibition space, divided by zones, which are labeled by their theme. She
clicks on the general area she remembers the object to be, and is given some
textual content about the zone in general (i.e., zone main text), and a
shortish list of objects (image and object name, perhaps) organized by
subtheme/subzone. Clicking on the object gives detailed information on it,
including related texts. From here, she can click on various different paths
to related objects: by time, artist, etc.; clicking a path highlights zones
on the map which contain those related objects. She can then click on those
zones, and find the relevant objects highlighted in the objects list.

Alternatively, she might use a search feature to find an object that she
doesn't remember the general location of, or simply to explore for more
information. The map would highlight zones with relevant object results, as
well as give a standard listing of object search results to the side of the
map. Hovering over or clicking on an object in the side results would
highlight the correlating zone.

This use case has been built in part from some data we've collected from our
McCord visit, which I'm compiling and will send out soon(ish).

I'm hoping this use case can serve as a starting point for dialogue about
our first designs and prototypes for a mapping solution. Naturally, our
dialogue may very well lead us into an entirely different direction.

Let's meet soon to talk about going from here.

(also, I think Antranig has some notes he'll be putting up soonish about the
implementation side of mapping; technological options and so forth; yes?)

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://fluidproject.org/pipermail/fluid-work/attachments/20090527/56054f79/attachment.html>

More information about the fluid-work mailing list