[Tagging] motorcycle:scale

Martin Koppenhoefer dieterdreist at gmail.com
Mon Feb 4 12:08:27 UTC 2019


Am Mo., 4. Feb. 2019 um 12:24 Uhr schrieb Christoph Hormann <osm at imagico.de
>:

> While i agree on this particular case your distinction can be easily
> misread to mean that mappers must not invent new keys or that they have
> to write a proposal for them.
>


writing a proposal _is_ documentation, and I cannot see how it would be
more complicated than setting up a tag definition page in the namespace for
established tags, it is basically the same operation.



> The reason i am emphasizing this is diversity.  People in
> underrepresented parts of the world will much more often run across
> things for which no established tagging exists than we do - yet they
> have comparitively little chances to have such tag established or to
> bring through a proposal for them.



Why would it be more complicated for them to get their proposal approved?
If the thing makes sense, sooner or later you will get the numbers to mark
it "de-facto" even without any voting or tagging maliing list interactions.



>   And creating a additional hurdle
> for this by banning documentation of their tag from normal tag
> documentation (and therefore from showing up in taginfo, possibly also
> in editors etc.) is counterproductive.
>


if you use the redirect you can make the proposal visible in taginfo (IIRR,
all my old proposals seem to be de-facto now, so I don't find an example),
although one could also argue that this is a shortcoming of taginfo and it
should fallback to proposals if no tag-definition page is found. This is of
course only a fair option if there is only one proposal for a certain key
or k/v combination, but otherwise I would say there is a bigger problem
anyway and you should consider using a different word for your tag.




>
> We already have way too many cases where mapping in parts of the world
> with a geography very different from that in Europe and North America
> people cargo cult a European geography - essentially drawing the map to
> create a look-alike of a European setting.  Telling people they either
> have to use established European tagging or they have bury the
> documentation of their tags somewhere where no one can accidently find
> them is not helping with that.
>


putting a tag in the proposal section is not hiding it where noone can
accidentally find them, that would be the user's sandbox. Tags in the
proposal section are found easily.



>
> Note in the vast majority of cases we are talking about new tags, not
> new keys.  But allowing documentation of new tags but not of new keys
> would be kind of weired of course.
>


>From my perspective it doesn't make a difference if a new key or a new
value for an existing key is proposed, with regard to how the procedure to
do it should look like. Obviously, with the current ecosystem, it is easier
to convince people to add a new value for an established key rather than a
new key



>
> My suggestion:  Create a new status value for the info box "not
> established" that is to be used whenever a tag has less than 500 uses
> and less than 20 active users.  And highlight such tags with a
> prominent warning that this tag is a new invention not yet broadly
> accepted.
>


and that its definition might change? Think about people "occupying" common
words with their own "strange" definitions as new keys, in an unilateral
effort not backed by the community. This happens not too rarely, and if
these things were inside a proposal section it would be much easier to
establish different definitions, as if they were in the "normal" wiki, and
when they finally get discovered, other mappers will have based their work
on them, and there will be an outcry towards any change.



>
> There by the way is also another side to the whole subject - that is
> established tags with lots of uses for which there is only a proposal
> page.  That is bad because a proposal page by definition describes an
> idea how a tag is supposed to be used while a tag page should describe
> how a tag is used.  And even if at some point a wiki editor creates a
> tag page this is often with content copied from the proposal without
> checking if actual tag use is in line with that.  Encouraging to create
> proposal pages instead of tag pages when inventing tags essentially
> encourages wishful thinking about the meaning of tags instead of actual
> documentation of the reality of use.



it is very difficult in general to get a good picture about "the reality of
use", because you would have to know all the places that are mapped in
order to judge the tag application. In practice it means you will check the
objects you know, and these will usually be in a very limited spatial area.
Tags that are used a lot can only be checked by random sample, you can't
check 100.000 uses, and you can't be sure your sample is representative for
the tag quality. IMHO we should emphasize more on the idea of what the tag
shall represent, and while it makes sense to point out  "abuse" of a tag in
the tag definition page (if significant), the aim should mostly be to retag
the outliers, rather than "losing" the tag by making its defnition more
inclusive (unless there is agreement on a new definition, which would be
more in line with actual usage).

Cheers,
Martin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20190204/db5d82b3/attachment.html>


More information about the Tagging mailing list