Hi Todd,
See below...
On Apr 23, 10:58 am, "lee.t" <
3le...@gmail.com> wrote:
> This is interesting JP, but I'm thinking that, while the real-time
> info is certainly novel,
...actually, it isn't at all. The first real-time passenger
information system for bus fixed-route services with wayside displays
along the route - that I am aware of - is route 54 in Stockholm,
Sweden. This system was built around 1992. Many systems of this kind
around Europe and North America were built since (I worked on a
handful) and probably other continents as well...
> the majority of agencies either don't have
> it,
The bigger ones typically do for quite some time (called CAD/AVL
systems) and many in Europe have second generation refresh's behind
them for quite some time now, while the first transit agencies which
implemented CAD/AVL in North America are now also looking at
refreshing their systems with new technology.
> or if they do they're not sharing it externally...
which is absolutely correct. Initiatives like Dabnab or Trillium's
project in San Francisco illustrate that need - at this point, they
are scraping arrival estimates off transit agencies web sites. Less
than ideal, as the format of the presentation layer will change more
easily than the underlying data feed.
> What is generally shared (the regularly scheduled routes/stops in a given
> service area) is useful enough in it's own right and could certainly
> benefit from an open Web Service/API so that developers don't have to
> 'reinvent the wheel' in order to get access to up to date data on
> mobile device.
With growing maturity and broader implementation of CAD/AVL, real-time
data indeed are available. These are multi-million dollar systems, so
you can see that offering the relevant data to the public is by far
and large a cultural and "political" issue rather than a technical or
fiscal one, and the benefit over plain schedule data outweighs the
extra effort IMO.
>
> The idea here would be to have a standard, hosted solution based on
> the google transit data that allows structured queries or specific api
> calls of ALL available agency data, rather than a specific solution
> tailored to one specific transit provider (like trimet).
You hit it on the head - it would be more than feasible that other
transit agencies offer feeds in this format. Development of standards
however has been a longstanding issue. You've got to start somewhere,
as they say.