<div dir="ltr"><p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial">Sorry for not answering for so long time. </span></p><p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial"><br></span></p>

<p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial">Only recently we have a meeting of a small local
group of OSM mappers that are preparing a land cover data import. We have
analysed the “problematic” tags (natural=heath, natural=moor and natural=bare_rock).</span></p><p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial"><br></span></p>

<p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial">We have been looking to already imported data
and we have generally agreed that existing tagging might be problematic. We
have concluded that natural=scrub corresponds to raba:id=1410  much closely. Not really exact – but the
estimation is than more than 80% of existing elements of raba:id=1410 could be
classified as natural=scrub. We have concluded that these elements should be
imported with tag natural=scrub.</span></p>

<p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial"><br></span></p><p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial">Situation with raba:id=5000 and raba:id=6000 is
much more confusing. We have been agreed that these two elements should not be
imported.</span></p>

<p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial"><br></span></p><p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial">Prior to importing new areas existing data
imported will be improved:</span></p>

<p class="" style="text-indent: -18pt;"><span lang="EN-GB" style="font-family:Arial,sans-serif">-<span style="font-stretch:normal;font-size:7pt;font-family:'Times New Roman'">       - </span></span><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial">Elements with raba:id=5000 and 6000 will be
deleted</span></p>

<p class="" style="text-indent: -18pt;"><span lang="EN-GB" style="font-family:Arial,sans-serif">- -<span style="font-stretch:normal;font-size:7pt;font-family:'Times New Roman'">    -  </span></span><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial">Elements with raba:id=1410 will be changed to
natural=scrub</span></p>

<p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial">We are intending to use JOSM with query
operation (like: <i>user:the_one_who_import
and “raba:id” =1410</i>) for each area (tile) already imported. </span><span style="font-family:Arial,sans-serif">If there is any better technique to perform
such operation, please let us know.</span></p>

<p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial"><br></span></p><p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial">Best regards,</span></p><p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial"><br></span></p><p class="MsoNormal"><span lang="EN-GB" style="font-family:Arial,sans-serif;background-image:initial;background-repeat:initial">Martin</span></p>

<span lang="EN-GB" style="font-size:11pt;line-height:115%;font-family:Arial,sans-serif;background-image:initial;background-repeat:initial"><div><span lang="EN-GB" style="font-size:11pt;line-height:115%;font-family:Arial,sans-serif;background-image:initial;background-repeat:initial"><br></span></div><br></span></div><div class="gmail_extra"><br><div class="gmail_quote">On 12 July 2015 at 18:36, Christoph Hormann <span dir="ltr"><<a href="mailto:chris_hormann@gmx.de" target="_blank">chris_hormann@gmx.de</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">On Sunday 12 July 2015, colored stone wrote:<br>
> Please find clarifications/explanations on the Slovene agricultural<br>
> land use (shortly RABA-KGZ) to OSM tag translations. For the<br>
> clarifications we have mainly used the “methodological paper on<br>
> agricultural land use” or “the interpretation rules” available at<br>
> <a href="http://rkg.gov.si/GERK/documents/RABA_IntKljuc_20131009.pdf" rel="noreferrer" target="_blank">http://rkg.gov.si/GERK/documents/RABA_IntKljuc_20131009.pdf</a>.<br>
</span>> [...]<br>
<br>
Thanks for the additional information.  I won't argue the points based<br>
on the methodological paper in detail since i can not read it in the<br>
original language and i am not sufficiently familiar with the data to<br>
judge how well the actual data complies with the specifications.<br>
<br>
A few general points though:<br>
<br>
- use of landcover tags, in particular things like natural=scrub and<br>
natural=heath in OSM is frequently quite inprecise.  This is not a good<br>
reason to be less strict with tagging in an import.<br>
- the OSM landcover tagging unlike the classification scheme of your<br>
source data is not a closed system.  Not every area on earth matches<br>
one of the OSM tags.<br>
- individual positive examples do not mean much - no one questions that<br>
each of the source data classes also contains areas that match the<br>
planned OSM tags.  The question is how much of the data does *not*<br>
match the planned tagging.<br>
<br>
Therefore here a few examples from the already imported data of what<br>
does not match your planned tagging:<br>
<br>
natural=heath:<br>
<br>
<a href="http://www.openstreetmap.org/way/353487056" rel="noreferrer" target="_blank">http://www.openstreetmap.org/way/353487056</a><br>
<a href="http://www.openstreetmap.org/way/353231651" rel="noreferrer" target="_blank">http://www.openstreetmap.org/way/353231651</a><br>
<a href="http://www.openstreetmap.org/relation/5271375" rel="noreferrer" target="_blank">http://www.openstreetmap.org/relation/5271375</a><br>
<br>
As far as i can see most of these are grassland (or other herbaceous<br>
vegetation like ferns as well as Blackberries, other Rosaceae and<br>
similar plants) with a varying amount of larger scrubs and mostly<br>
smaller trees.  In many cases this is land that is in the process of<br>
being reclaimed by trees - either previously cut forest areas or former<br>
farmland no more used.  natural=heath specifically means dwarf scrub<br>
vegetation (that is species that naturally do not grow tall, not young<br>
trees).  In Slovenia's climate this is rare outside high mountain<br>
areas.<br>
<br>
Interestingly most areas that would qualify as natural=heath are<br>
probably included in class 5000.<br>
<br>
natural=moor (to be natural=fell):<br>
<br>
<a href="http://www.openstreetmap.org/way/353333307" rel="noreferrer" target="_blank">http://www.openstreetmap.org/way/353333307</a><br>
<a href="http://www.openstreetmap.org/relation/5274011" rel="noreferrer" target="_blank">http://www.openstreetmap.org/relation/5274011</a><br>
as well as all other areas with this tag except for the Alps - none of<br>
this is close to the alpine tree line so natural=fell does not match.<br>
Use in the Alps is questionable as well but since natural=fell is a<br>
fairly vague tag you can't really say it is wrong.<br>
<br>
natural=bare_rock:<br>
<br>
<a href="http://www.openstreetmap.org/way/351883914" rel="noreferrer" target="_blank">http://www.openstreetmap.org/way/351883914</a><br>
that is a great example for natural=scree - and here it is specifically<br>
excluding the rock area above<br>
<br>
<a href="http://www.openstreetmap.org/relation/5252002" rel="noreferrer" target="_blank">http://www.openstreetmap.org/relation/5252002</a><br>
mixture of bare_rock, scree and sparsely vegetated areas<br>
<br>
<a href="http://www.openstreetmap.org/way/353153325" rel="noreferrer" target="_blank">http://www.openstreetmap.org/way/353153325</a><br>
that would be natural=shingle<br>
<span class="HOEnZb"><font color="#888888"><br>
<br>
--<br>
Christoph Hormann<br>
<a href="http://www.imagico.de/" rel="noreferrer" target="_blank">http://www.imagico.de/</a><br>
<br>
_______________________________________________<br>
Imports mailing list<br>
<a href="mailto:Imports@openstreetmap.org">Imports@openstreetmap.org</a><br>
<a href="https://lists.openstreetmap.org/listinfo/imports" rel="noreferrer" target="_blank">https://lists.openstreetmap.org/listinfo/imports</a><br>
</font></span></blockquote></div><br></div>