[OSM-talk] Suggestions for recycling schema

[Andy Robinson]

Joerg Ostertag (OSM Munich/Germany) wrote:
>Sent: 19 September 2006 10:59 AM
>To: talk at openstreetmap.org
>Subject: Re: [OSM-talk] Suggestions for recycling schema
>
>On Tuesday 19 September 2006 11:13, Andy Robinson wrote:
>> This type of extension raises a question about how far the "Map Features"
>> tagging should go and whether we should start splitting off some features
>> and attributes beyond those needed for maps and map rendering to separate
>> sections outside the Map_Features list itself. I can understand the wish
>to
>> have a recycling annotation or sign on a map but I can only see that you
>> would want to know the types of materials recycled if you were making a
>> very specialist map or if you were using the information outside of a
>> rendered map (ie information used for other purposes).
>>
>> I have no problem putting all of this into Map Features, but since I'm
>> slowly getting around to working up the next version it would be useful
>to
>> have feedback. The next version is expected to focus on the primary key
>> types as a means of indexing so extended lists of keys and values should
>> not be a problem but for the more casual user and those new to OSM it may
>> be a bit daunting if the structure is too deep. I'm for keeping the
>system
>> simple but with extensibility.
>
>I see more and more tags arising. But for a renderer this makes it harder
>and
>harder to catch up with all these new tags. If the renderer gets an idea
>what
>kind of element to draw in some cases it wouldn't be really necessary to
>know
>which particular icon or color to use. So I would suggest to try to use a
>scheme which a renderer can easily parse and see if it is:
> - a POI: --> Draw an icon and name it
> - a Street: -> Draw a line
> - a Lake/Sea: -> Draw a blue area
> - a river: Draw a blue line
> - an area: Fill it with a default color
>This would make things much easier for rendering and Splitting Data. Even
>things like osm2csv can get much benefit out of this. It would be much
>easier
>to split between the different type of features.
>This would be kind of an extension to the existing features. But might make
>a
>lot of things easier.
>

Absolutely, distinguishing between the types of feature is important and I
see this approach as the top level of indexing in any revision of
Map_Features tagging practice.

I also think there is a benefit in considering an XML type structure since
that is currently the way everything else works in and out of OSM and would
promote a much better logic to the extension for tags but would permit some
basic rendering and use of the data just using the core XML types.

But does this just move us closer to GML, which I had big problems
understanding the basis of when I looked at it a year ago.

Cheers

Andy



>-
>
>Joerg
>
>_______________________________________________
>talk mailing list
>talk at openstreetmap.org
>http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
>
>--
>No virus found in this incoming message.
>Checked by AVG Free Edition.
>Version: 7.1.405 / Virus Database: 268.12.5/450 - Release Date: 18/09/2006
>

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.405 / Virus Database: 268.12.5/450 - Release Date: 18/09/2006