You are fond of proposing keys with arbitrary numbers as the value, or part
of the value. This would be fine if we were using a relational database,
where a mapper could select one of a list of human-language descriptions,
which would then get translated to the magic number for storage. However,
we do not have such tables, and presenting a mapper with a list of choices
such as photo1, photo2, and photo3 is likely to result in corrupted data,
due to a mapper picking one at random, or misremembering what means what.
As long as the descriptions of the values aren't shown at the time the
value is being selected, we need to stick close to natural language, not
magic numbers.
--
John F. Eldredge -- j...@jfeldredge.com
"Darkness cannot drive out darkness; only light can do that. Hate cannot
drive out hate; only love can do that." -- Martin Luther King, Jr.
On June 6, 2015 5:31:48 PM "pmailkeey ." <pmailk...@googlemail.com> wrote:
On 5 June 2015 at 10:33, David Fisher <djfishe...@gmail.com> wrote:
> On Fri, Jun 5, 2015 at 12:33 AM, pmailkeey . <pmailk...@googlemail.com>
> wrote:
> >
> > The issue with the 'oneway' key is that the key itself contains 'data'
> > relating to the value. Oneway without a value would imply =yes whereas
> > building without a value (or =yes) would give data independent of the
> value,
> > IYSWIM
> >
> > building=
> >
> > hospital=
> >
> > The latter describes the building without the need for a value.
> >
> > I note your TIAL v CE above. Why do we need to know what the landuse is
> in
> > any case ?
> >
>
> I do see what you mean. I think the difference is that "building = x"
> in some sense defines the presence of the object, as does "highway =
> x" on a way. So, if "building = x" is not set (presumably on a
> "circular" way), or if "highway = x" is not set (presumably on a
> "linear" way), then those ways are just collections of nodes, nothing
> more. But "oneway = x" defines a *characteristic* of a way. A way
> must fundamentally *be* something (e.g. a building or a highway), but
> it may nor may not have any number of characteristics which don't
> alter that fundamental *being*. The only sensible way to deal with
> *characteristics* (other than insisting that every way has hundreds of
> tags) is to assume defaults.
The problem with oneway is the key name - it's 3 letters too long:
way=1
way=2 - maybe ways=1, ways=2.
>
> As for "landuse=residential" -- I agree that we could probably do
> without it. But it does add to the readability of the map, especially
> at low zoom levels, as it enables you to see at a glance where places
> are and how big they are.
Again, the issue is the letters in the tag. landuse=place or
landuse=settlement - landuse=town rather than landuse=residential,
place=town - combine the two.
> Personally I'm an advocate of covering the
> majority of the map (not necessarily 100%) with some form of landuse
> area, e.g. residential, industrial, grass/meadow/parkland, farmland,
> etc. -- though I appreciate that not everyone shares that view.
>
>
Anyone that doesn't share that view should be nowhere near OSM ! Sounds
like a military cover-up to me ;)
I do feel the default map not covered should be black though - to make the
unmapped areas stand out :)
--
Mike.
@millomweb <https://sites.google.com/site/millomweb/index/introduction> -
For all your info on Millom and South Copeland
via *the area's premier website - *
*currently unavailable due to the country's ongoing harassment of me, my
family, property & pets*
T&Cs <https://sites.google.com/site/pmailkeey/e-mail>
----------
_______________________________________________
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging
_______________________________________________
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging
