[OSM-talk] tagging maxspeed restricted to time span

[Peter Miller]

GDF allows attributes ('tags' to you and me) to be optionally restricted by
any combination of Time Dependency, Vehicle Dependency and Lane Dependency.

I haven't used GDF for real yet, but I do now know my way around and I am
impressed with it. I recommend that other put a wet towel over their heads
and also give it a read. I find it good to print that sort of standard out
with 2 sheets to a page and double sided if possible which keeps it quiet
manageable. Do book-mark the important sections as you find your way around
as it is laid out in a particularly unhelpful format and I found it really
hard to locate the important sections again when I wanted to refer back to
them.

There is also a page on GDF on the OSM wiki:
http://wiki.openstreetmap.org/index.php?title=GDF

My company will be doing some experimentation with GDF and OSM prior to the
SOTM conference and will give an update when it is done.

Btw, I think that we should not ditch Segments until we have looked in more
detail at GDF. Personally I think the GDF links/segments model ('Level 0'),
may be useful when defining parks and building that butt up to roads.



Regards,

 

Peter



-----------------------
> 
> Message: 7
> Date: Tue, 26 Jun 2007 01:02:47 +0300
> From: Tapio Sokura <oh2kku at iki.fi>
> Subject: Re: [OSM-talk] tagging maxspeed restricted to time span
> To: talk at openstreetmap.org
> Message-ID: <46803B87.6020800 at iki.fi>
> Content-Type: text/plain; charset=UTF-8; format=flowed
> 
> D Tucny wrote:
> > Unless it's changed recently, the 'Map Features' restrictions tags are
> not
> > sufficient to tag restrictions in enough detail to accurately reflect
> many
> > real world restrictions... It's something that will be complex to get
> > right,
> > and it's unlikely to be a simple solution... So, we should probably
> > start by
> > detailing some real world restrictions that we will need to document so
> > that
> > we can discuss and devise a solution...
> 
> I've been following the discussion on turning restrictions etc for a
> while now. Every now and then someone mentions GDF, the data model most
> commercial road data producers and consumers use. If most of the data
> model problems have already been solved by GDF, what's stopping us from
> using (parts or all of) it? Sure it's an ISO standard and they cost
> money, but the specifications seem to be freely available as well (for
> example pdfs at
> http://www.ertico.com/en/links/links/gdf_-_geographic_data_files.htm).
> 
> In any case I'd rather we develop a very good model or use an existing
> model. I'm afraid anything made out of bubble gum by using the current
> node/segment/way/area key-value pairs will just lead into further
> trouble and extra work later on.
> 
>    Tapio
> 
> PS. I haven't actually read the GDF spec so I don't know how braindead
> or not it is, but at least it is something that probably should be
> looked at.
> 
> 
>