[OSM-dev] Ponderings about an improved tagging scheme

Steve Hill steve at nexusuk.org
Fri Feb 20 13:31:07 GMT 2009

On Fri, 20 Feb 2009, marcus.wolschon at googlemail.com wrote:

>> Well, I'm not sure that the kind of objects found on a map can't be
>> categorized by a sufficiently flexible system.
> Like....
> * objects belonging to multiple "types"?

This was already covered by the proposal (and is indeed one of the 
driving factors behind it).

> * different ways of categorizing the same information?

I'm not sure what you mean.

> Given the imagination of people it was proven beyond any doubt
> that the collection is infinite.

There are a finite number of objects in the OSM database.

In any case, no tagging scheme is going to place any restriction on the 
number of different types of object - there is nothing stopping you 
deciding to use a new value for the "type" tag if no one has ever mapped 
that type of object before.

> Strange, I was under the impression that our system was quite chaotic
> and worked under the 2 rules:
> 1. "just do it or be ignored"
> 2. "don't break what others have build or they will get very angry"

These 2 rules are at odds with eachother though.  Over time, tags get 
depricated, new tags get created.  If I'm going to get angry because you 
"break" the roads I've mapped by migrating them from depricated tags to 
new tags then all the software is eventually going to become a nightmare 
to maintain.  You *can't* support a mess of depricated tags forever, at 
some point you have to draw the line and say "we need to fix the data so 
we can simplify the software".

>> How about an orderly collection of attributes (which would cover most of
>> what's going on now with things like "highway=" or "natural=") WITH the
>> option of a free-form "tags" attribute for object types and attributes
> that
>> fall through the cracks?
> What about you start working on such an order so that it encompasses
> everything
> people have already done or are currently doing and we see each other again
> in
> say... 15 years?

I'd like to point out that if everyone said "that job's so big, it's 
impossible", we'd never get anything done.

I mean, a bunch of people creating a map of a good chunk of the world is 
such a big job that it is practically impossible... right?  Oh wait - 
that's exactly what we've done, precisely because people actually got on 
and did it instead of telling eachother that it was such a big job that it 
wasn't worth attempting.

Steering a project as big as OSM is like steering a supertanker - you 
can't just stop and change direction.  But if no one ever takes a step 
back and looks at possible directions that it *might* be an idea to go in 
then the project is doomed.  You can't keep taking the path of least 
resistance forever, at some point you have to say "what can we do to make 
this *better* instead of just continuing as we are", otherwise you just 
end up heading into a dead end.

> Yes and we achieve consistency by consense.
> People do like other people have already done.
> Everything else has failed due to being ignored by an army of mappers.

That's not true at all.  Over the years the project has evolved - it is 
nothing like it was when it started.  If everyone just stood by and said 
"we're not going to try and change anything because we will just be 
ignored" we would still be using constructs like segments, relations 
wouldn't exist, we wouldn't have spinoff projects like the cycle map and 
the piste map, etc.

  - Steve
    xmpp:steve at nexusuk.org   sip:steve at nexusuk.org   http://www.nexusuk.org/

      Servatis a periculum, servatis a maleficum - Whisper, Evanescence

More information about the dev mailing list