[Tagging] Feature Proposal - Voting - historic

[Minh Nguyen]

Vào lúc 18:16 2022-11-04, Martin Koppenhoefer đã viết:
> 
>> On 4 Nov 2022, at 13:17, Marc_marc <marc_marc at mailo.com> wrote:
>>
>> our "sister" project (wikipedia) has no problem defining what is an anecdote and what is "relevance from a historic viewpoint",
>> I don't see why we should have any issue doing it.
> 
> 
> Mappers are working fundamentally different from wikipedia authors, because they are recording observations, first hand study, while wikipedia work means working with sources. Original research is explicitly frowned upon in wikipedia while it is at the basis of mapping. We do not have relevance criteria as a hurdle for inclusion of things, we only require them to exist. I do not say relevance does not exist, but it is less important for our mapping. We are creating “categories” of things by applying tags, and I do not believe it would be helpful to have different main categories for the same thing, depending on its historic relevance, hence I do not believe redefining the “historic” key in this direction would be helpful for the project.

I think this point would be every bit as applicable to our other sister 
project, OpenHistoricalMap, which works with both published sources (for 
the past) and observations (for the present). So far, OHM has hewn to 
OSM's tagging schemes, including the historic=* tags that are almost 
always anachronistic.

For example, most historic=manor weren't historic the moment they were 
completed. A historic=battlefield may not have been historic until the 
tide turned in a larger campaign. I would expect historic=citywalls to 
be less common in OHM than barrier=city_wall.

Both historic=* and heritage=* make a value judgment, but heritage 
designations are made by historical societies and government agencies. 
Who are we to make a similar designation as mappers?

Maybe both OSM and OHM could live with historic=* being a misnomer as 
(wait for it) a historical accident, but if someone is coining a tag and 
they're choosing between historic=* and another suitable key, I'd advise 
them to pick the other key.

-- 
minh at nguyen.cincinnati.oh.us