[OSM-talk] Tag proposal/approval system is too heavyweight

[Christian Linder]

IMHO, this great project only makes sense if tagging converges to one
logical and clearly defined schema in the end. The whole point of the
project is making one aspect of reality (geography) computer-readable. If
the tagging doesnt converge to a common schema, it is not understandable by
computers. I would then rather call it "reduced feature aerial photography".

In addition to that, IMO it is absolutely required to have a recommendation
for beginners. And this RECOMMENDATION is all we talk about, right? This is
what people on the proposed features page vote on. Everybody can tag
whatever (s)he wants, but what is the point in presenting the beginner 8
different possible tagging schemes? Some people just want to tag, they do
not want to read 4 pages of discussion just to make a conscious decision on
how to tag something. What we need AT LEAST is a list of whats rendered in
which renderer, but since there might be un unlimited number of renderers,
this will get very tedious.

I clearly believe that

> Frederik wrote:
> that we didn't have a fixed ontology from the start, because it
> would have (a) cost too much time to create and (b) turned out to be
> insufficient at the earliest opportunity

What I dont believe is

> Frederik wrote:
> The "evolutionary" way we
> tag the world is one of the key distinguishing factors between OSM
> and anything else, one of the big pillars of our success

If this means the evolution of the map, I agree. If it means the evolution
of the tagging schema, I dont.
In my opinion, if there would have been a good and complete tagging schema,
and adherence to this schema would have been enforced, the project would
have had even more success.
But this is a pointless argument, as it would have been impossible to start
off with a perfect schema.

> Richard wrote:
> Unfortunately Map Features may be diverging from the community. Tags
> are proposed and voted on by a very small subset of people

If they are diverging from the mapping community, I dont see a problem. As a
matter of fact there are people that enjoy discussing map features and some
others that enjoy looking up how something has to be mapped and just mapping
it. If the renderers adhere to the recommended map features, the mapping
community will either have to get involved in discussion or accustom to the
changes. Plus, in a project in its exponential growth phase, the number of
new members adhering to the recommendation will soon be higher than the
number of old taggers that havent realized the change.
If map features are diverging from the renderers *and* from the mapping
community, I definitely see a problem here.

I can see how voting is a PITA and not representative either. Still, it is
IMHO better than not voting, because it requires at least *some* agreement.
And after that, there is ONE recommendation. Ofcourse, the decision can be
challenged any time later, but thats the whole point of an open project.
Then, again, there would be ONE recommendation.

The only alternative I see is a page with a couple of *well-defined*
alternatives (that have to be voted on?), each accompanied by usage
statistics and information about which renderer supports it and which
doesnt.

Best regards
Chrischan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20080320/f56f2c2f/attachment.html>