In January, Google announced a contest in which US and Canadian students were invited to generate their college campuses in 3D, using Google Sketchup. A few individuals or teams whose works are judged to be the best will be flown to Google headquarters in Mountain View for a special workshop with 3D modelling experts in the company and tour of the Googleplex campus. All submissions belong to Google of course, making it yet another innovative approach to exploiting the power of user-generated content to build out features through relatively simple/cheap incentives, similar in some respects to the Google Image Labeler program.
This was a clever way to build great 3D projections for college campuses, and in the process seduce college students into becoming familiar with Google Sketchup. But, could there have been another ulterior motive to begin gathering data for a type of content that has been less optimal in Google Maps?
|Folsom Stadium, University of Colorado, Boulder
Model created by Nathan King
It can be surmised that Google set up the contest in small part to draw college students into using the Sketchup tool and become familiar with it, but the bigger goal was likely to improve content for Google Earth in areas of significant value for which they didn’t have ready access to good or detailed data. As Google Earth is being enhanced, there’s been increasing supposition that the company was planning to use it as a foundation for a new virtual reality.
It seems even more commercially-applicable if they’d be using the content for increasing the worth of a primary focus area of theirs: maps and local search.
What if Google began using this data to offer improved free/interactive online campus maps to universities? Jazzy, AJAXified maps have been a trendy thing, and each of the top providers is looking to find ways to further differentiate themselves from their competition in the cartographic arms race, and to do this they’re looking to create new features or to enhance existing features. Campus maps are one significant area needing improvement in the major map search interfaces.
After street/road maps of cities and the country, college campuses are where there is another substantial consumer demand for good maps. There are over 4,000 colleges and universities in the US, with around 17 million college students (source: NCES). Add to this the amount of other university constituents such as faculty/staff, and visitors each year and you have a whopping big number of individuals that need to know where to find locations on campuses. All of the universities and colleges try to provide some sort of printout maps of their campuses which pinpoint building locations in relation to surrounding streets, parking lots, and other landmarks. But these maps are challenging to maintain for larger institutions, and many universities don’t have the time nor skill to keep up with much more than a diagram showing building footprints. Sometimes these college maps have north at the top of the page (as is a standard for maps), but others orient their buildings to best fit on 8 1/2 x 11 pages. With vague site-plan outlines of building footprints, and inconsistent north orientations, these maps are typically disorienting to users.
The top online map providers do not have good campus maps in most cases, since the US Postal Service doesn’t typically deliver directly to each building on university campuses—they deliver mail to one, central address for the university campus, and the university distributes the mail from there. Because of this, many of these buildings do not have a direct street address, making it impossible to pinpoint them using US Postal data, which is used in part for geocoding home and business addresses.
In most cases, college campuses are marked in online maps with little more than an outline of campus or a perimeter outline with the campus building footprints shown, effectively rendering the maps irrelevant once a user arrives at the campus. (see Google’s map of UCLA, for instance).
Knowing that a building is square when viewed from above, along with many other rectangular buildings in the same area doesn’t help much since most consumers are not be trained in mentally projecting how those outlines should appear to them when viewed in profile as they are walking across campus. Even being able to see the roofs of the buildings from the satellite images wouldn’t necessarily help much, either. The top mapping providers don’t even display labels for the building names.
Even with less-than-ideal graphic representation, some universities have opted to use the Google Maps API to mashup their buildings and parking info with the existing birds-eye-view maps and sat images, such as: Stanford, Texas A&M, and Tufts.
Streets, intersections and some main landmarks are sufficient for consumers to use as navigation while driving vehicles, but it’s not at all satisfactory for pedestrians who are trying to find their way across the sometimes maze-like grounds of decent-sized educational institutions where jumbles of buildings have grown up with no street addresses and sometimes only sidewalks connecting them. Universities with dozens upon dozens of buildings seem to form small cities within their host cities, making good navigational aids vital to their constituents.
Considering the need for better campus maps, and their decision to host the Sketchup contest, it’s not at all inconceivable that Google might be looking to create better interactive campus maps for universities, perhaps with an extra button for "Perspective View" alongside their standard "Map", "Satellite", and "Hybrid" buttons. Google already has a long history of providing free services to college websites such as SiteSearch. Why wouldn’t Google look for a way to enable users to select a ground-level view of surrounding buildings and terrain directly from their location maps? It’d be particularly beneficial for places like college campuses where there’s more pedestrian traffic and where navigation is less dependent upon street addresses, and more dependent upon visual orientation of landmark features and 3D views.
Opinions expressed in the article are those of the guest author and not necessarily Search Engine Land.