[Osmf-talk] Tagging standards

[Craig Allan]

PART 1 : Tagging

I've been thinking about the tagging discussion...

    It is often a problem when a specific tag is used for either different
    purposes or even worse when it's used for the same purpose but the
    values are interpreted differently. (Peter Gervai)

The LITTLE issue is when Person Romeo and Person Sierra use different 
tags for one type of geographic feature. It's a little issue because a 
computer provided with a list of such duplicate tags can successfully 
render a map. It's a bit more work than using a strict unique key=value 
standardised coding schema, but quite possible, and is normal practice.

The BIG issue is when Person Tango and Person Quebec use one tag for two 
different types of geographic feature.  It's a big issue because the 
tagging cannot be resolved by any (affordable) software. It is just 
impossible for a map rendering application.

One more example of the BIG issue is "natural=wood" for (mostly) 
untouched natural forests. Examples are in the Congo and Amazon basins.  
Simultaneously "natural=wood" is used for man-made single species tree 
farms which to my mind are  "landuse=farmland crop=fast_growing_wood".   
I don't want to debate these particular tags, I'm just giving an example 
of impossible to resolve confusion generated in this way.

Many people support free tagging, so we should accept that as status 
quo. The implied rule for free tagging is "Any one mapped feature may be 
identified by many different tags".

To avoid the same tag identifying different features we need to 
collectively set just one rule. "No two different mapped features may be 
identified by the same tag"

That's not standardising at all, but it achieves similar resolution of 
the BIG issue. If the second rule is applied, and enforced, and cleaned 
up across the database as best we can, then there is no pressing need 
for standardisation to a unique key=value schema.

And yes, Rule 2 aligns with the Wiki "Any tags you like" page. See the 
second paragraph:

    "Though it is not 'feel free to ignore existing tagging schemes and
    start marking pharmacies with unicorn=parking_lot' ".

This sentence opens the door to some light-touch rule making.

Any thoughts?

PART 2: OSMF Talk

osmf-talk is a great place to discuss this kind of question. The list 
has been dead for a while and suddenly people are taking an interest. Good.

tnx, Craig

> I believe the tagging should contain all the information required to render a map according to local standards (if this is desired), but it should not “hijack” globally established tags to make every “global” standard renderer produce maps according to local expectations. This would not work because there are also other uses of the data than rendering a map image.
>
> Not sure if osmf-talk is the best place to discuss this kind of question
>
> Cheers Martin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/osmf-talk/attachments/20221019/a642f517/attachment.htm>