[OSM-talk] Voting

[Nigel Magnay]

I haven't been paying too much attention of late, but it's always
struck me that the simple solution to tagging is to allow tags to
exist within namespaces (akin, say, to XML namespaces).

That way, if you want to have one group going around having votes on
things, they can, and they can have their own sandbox to play in. They
could then have core:railway=station [1], and I could have
nrm:conkers=many [2], and everyone carries on. No need to get
'permission', but groups that _really care_ about, say, dataset
quality, can look up tags in "their" namespace and 'correct' them.

Good features to have would be the ability to subsume tags into a
namespace (hey, your nrm:conkers=many is my core:tree=chestnut, so
lets merge, or "Ive tagged every boatyard in the UK, I think it's
useful, would you like it in the boating database?), and possibly even
being able to clone / branch a namespace (e.g. I disagree with the
cabal running core:, so I'd like to have mycore: based on the tag set
as of xx/yy/zzzz, and I shall continue from there).

Because it seems to me that both sides are right. The people wanting
some kind of process are right because in order for some kinds of tags
to be useful needs some kind of consistency. And the people wanting
the freedom are also right - 'cos voting is cumbersome and I wanna
just Get On With It, and "who elected those guys over there anyway".
It's just the same as an OSS project - if you disagree, you can take
the branch and go off on your own.

[1] where, say, core=http://openstreetmap/tagset/core, or
core=http://openstreetmap/the-tagging-foundation/tagset/core, or
whatever
[2] and perhaps nrm=http://openstreetmap/users/magnayn

On Tue, Apr 8, 2008 at 2:30 PM, Frederik Ramm <frederik at remote.org> wrote:
> Sven,
>
>
>  > I can't remember that ULFL ever claimed that.
>
>  Ok. There we go again. Nobody has claimed anything, but the fact of the
>  matter is that a number of people seem to think that those who vote make
>  a decision that is "a decision of the project" rather than "a decision
>  of those five people who voted".
>
>  I've been critcised for not suggesting an alternative. So here's my
>  suggestion:
>
>  * Continue your discussion and voting as before
>
>  * Give yourselves a name ("OSM Tagging Task Force" or whatever) and
>  create a mailing list.
>
>  * Do not talk about "approved", "rejected", or "deprecated" features;
>  instead, if something is voted in favour, it becomes a "recommended by
>  OSMTTF" feature.
>
>  * Be very clear that any feature *not* voted upon, or any feature which
>  got less votes than something else, or any feature that a majority of
>  voters didn't like, is still perfectly valid to use - you just don't
>  actively recommend it.
>
>  * Never try to keep people from using tags you didn't recommend (i.e. do
>  not add a big message to the Wiki saying "THIS FEATURE IS NOT
>  RECOMMENDED!").
>
>  * Be very clear that the group you form is a small subset of the
>  project; you create recommendations based on today's knowledge and on
>  what you like and dislike. There may be any number of *other* groups in
>  the project who also create recommendations and who have the same right
>  to exist that you have. You are not special, the project has not asked
>  you to please give recommendations, and has not given you any special
>  powers that others don't have. (Much as the project never asks anyone to
>  please write software and be the project's premier software contributor
>  - anyone can do it and if it proves to be good, it is used.)
>
>  * Be very clear that your recommendations create no obligations
>  whatsoever on the part of renderers and editors; your tags are not
>  better or more important than anyone else's.
>
>  Do all this and I will stop complaining. I might even actively refer
>  people to you ("better talk this over with the guys on the tagging task
>  force list, they usually have good ideas" or so).
>
>
>  > Will this discussion only end when Ulf, Robin, me and several others set
>  > up a separate wiki for those who want to agree on and use a consistent
>  > tagging sheme because they believe it's a good thing? When this project
>  > is so open, why are we always blamed for what we do?
>
>  I'll draw a parallel to the licensing debate here. Over on legal-talk, I
>  constantly advocate PD, saying that nothing can ever be more "free" than
>  PD because it has no restrictions. I am then routinely criticised by
>  share-alike advocates who say that the freedom of PD might be abused by
>  people further down the line to actually *reduce* freedom.
>
>  In this discussion, I find myself on their side: Our project is so open,
>  and I have the impression that you are trying to *reduce* that openness
>  by setting up a voting process. I have the suspicion that in the end you
>  want a project where new tags aren't even allowed unless they underwent
>  discussion and voting. And that's where my fierce opposition comes from.
>
>  Bye
>  Frederik
>
>
>
>
>  _______________________________________________
>  talk mailing list
>  talk at openstreetmap.org
>  http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
>