[Talk-us] directions of ways in MassGIS data

[Christopher Schmidt]

On Sun, Feb 01, 2009 at 10:33:24AM -0500, Greg Troxel wrote:
> 
> In Massachusetts osm has state GIS data, and a number of the 1-way ways
> are the wrong direction.  This is particularly obvious for divided
> highways, e.g. Rt 2 inside 128, 495 near Rt 2.
> 
> Clearly I could go edit these manually.  But I wonder if there is
> automated way to fix this, perhaps by going back to the massgis data
> (surely massgis would encode direction), 

MassGIS does not encode the direction. Oneways which are potentially
wrong are marked with a FIXME note to fix the incorrect directionality.
(I believe it is 'FIXME: Unconfirmed oneway'.) MassGIS pays NavTeq for
routing data, and does not have a 'free' source of Oneway
directionality. 

So, no amount of fixed import script can help this: the data simply
doesn't exist from a free source.

> spiffing up the import script
> and applying diffs.  Did this issue arise with TIGER data, or in other
> states with state-level data?

TIGER data does not encode one ways at all.

> Also, some of the classifications seem not quite right.  Rt. 2 is
> sometimes labeled primary, which is correct some of the time, but the
> length between Alewife and 128 seems to be to me a motorway.  Is it the
> general consensus that state and US highways that are divided and have
> only on/off ramps and speed limits of 55/65 be tagged as motorways?
> That's how I interpret the tagging guidance but wanted to ask before
> changing it.

The classifications are based on the roads data, which is maintained by
the executive office of transportation. I would assume that the
classification is based on the legal classification of the road, rather
than on usage.

Regards,
-- 
Christopher Schmidt
MetaCarta