GTFS is the de-facto standard for transit service information--first defined by google when Portland's TriMet asked: why isn't online trip planning as easy for transit as it is for driving? At current count there are 1000+ public feeds on 6 continents. Wide adoption of the specification allows anyone interested in looking at, analyzing, or mapping transit service information to all communicate in the same language.
Fundamentally, GTFS is a set of (deceptively simple) tables, organized to relate to one another using unique identifiers. Together, a GTFS data set describes transit service in terms of: the geographic distribution of stops (using lat/long coordinates), the routes/trips offered by a given agency, and the schedules and frequency of those services. In diagrammatic form it looks something like this:
At EDR Group I've been spending a fair amount of my time lately using spatial data and other "join-able" data sources (demographics, economic activity, census journey-to-work patterns, freight flows, etc.) to understand the geography of access provided various transportation systems. GTFS is one cool ingredient in this wider world of merging many information sources into one single spatial framework.
After all-- the whole point of transportation is to give people and businesses access to opportunities.