> ... somewhat arbitrary and the related tower vs pole problem. These were all introduced in the > early days of osm without providing some rigid guidelines on how to distinguish them (the result occasionally being some strange tagging). ... tags are not "incorrect" in that they refer to correct English terms. ...<div>
<br></div><div>I'm still really new to OSM, so please excuse me if I am treading on some sort of crowd source heresy. And this question is not specifically directed at power grid tagging, but more toward resolving creating / assigning tagging attributes in general in OSM. </div>
<div><br></div><div>For reference, see the International Electrotechnical Commission ( <a href="http://www.iec.ch/about/">http://www.iec.ch/about/</a> ) Electropedia ( <a href="http://www.electropedia.org/">http://www.electropedia.org/</a> ) or from the Glossary search at ( <a href="http://std.iec.ch/glossary">http://std.iec.ch/glossary</a> ), to the Overhead lines / Towers description page ( <a href="http://www.electropedia.org/iev/iev.nsf/display?openform&ievref=466-08-08">http://www.electropedia.org/iev/iev.nsf/display?openform&ievref=466-08-08</a> ) where we see a diagram ( <a href="http://www.electropedia.org/iev/iev.nsf/master/466-08-08:fr/$FILE/466-fig2.gif">http://www.electropedia.org/iev/iev.nsf/master/466-08-08:fr/$FILE/466-fig2.gif</a> ) and nomenclature in multiple languages: (English) double warren redundant support, (French) triangulation en losange, (Arabic) تركيبة عل شكل المعين , (German) Ausfachung mit gekreuzten Diagonalen und Sekundärfachwerk, (Spanish) <span class="Apple-tab-span" style="white-space:pre"> </span>celosía doble, (Italian) tralicciatura doppia con rompitratta, (Japanese) ダブルワーレン補助材, (Portuguese) reticulação em losango, (Swedish) dubbel diagonalvandring med knäckavstyvning. </div>
<div><br></div><div>It is true that these standard vocabularies drill down to excruciating detail which may not be relevant to OSM ( characteristics which are visible (or inferred, like voltage from the in out conductor spacing) .</div>
<div><br></div><div>For example see Area: 415: Wind turbine generator systems ( <a href="http://www.electropedia.org/iev/iev.nsf/index?openform&part=415">http://www.electropedia.org/iev/iev.nsf/index?openform&part=415</a> ), the plural "wind power station / wind farm" as a "power station comprising a group or groups of wind turbine generator systems" is distinguished from the singular "wind turbine generator system" as a "WTGS, system which converts the kinetic wind energy into electric energy", the wind turbine itself as "wind turbine: rotating machinery in which the kinetic wind energy is transformed into another form of energy", to the more detailed type "horizontal axis wind turbine" or "vertical axis wind turbine". </div>
<div><br></div><div>Some of these characteristics would be visible to the person, while the other attributes at that level like 'shutdown', 'blocking', 'control system' are not. The other vocabulary category entries for 'Wind Turbines' such as 'Design and safety parameters', 'Wind characteristics', 'Power performance measurement techniques', 'Acoustic measurement techniques', are not 'visible' ( although 'Acoustic' could be debated because of cell phone microphones :-), However, 'Electrical interconnection', attributes might be inferred. </div>
<div><br></div><div>While I get that crowd generated attribute tagging has some unique advantages, huge flexibility, allows whatever level of detail in any language to be incorporates, at the other end of the pipe are editing tools, maintenance bots, and rendering engines which do expect some sort of conventions. There might be some advantages to at least examining these vocabularies ( like the IEC). For instance, it might be revealed that a proposed tag might actually be several additional tags ( I usually can't see every possible variation when looking at a specific case). As dedicated user communities seek to add their own tags, there would be a path to add more level of detail without breaking downstream tools. Expanding tag sets to other languages is somewhat easier because the basic objects and concepts are already translated. There also seems to be a growing space of communities and applications outside of OSM that would benefit from interoperability ( 'routing' as a quick example ). </div>
<div><br></div><div>There seems to be continuous churn about tag definitions, and I was curious why there wasn't more use of the international vocabularies to resolve conflicts and establish attribute spaces. And the power grid probably isn't the best example (because it appears to be so detailed and technical), but any man made structure in the environment probably already has a well established naming convention in place ( for architecture, probably since the Roman Empire :-). There are several excellent one for LULC (Land Use and Land Cover) for example.</div>
<div><br></div><div>Random thoughts, not a specific proposal. </div><div><br></div><div>Michael Patrick</div><div><br></div><div><br></div><div> </div>