[WIP] Borders for archeological site polygons.

[StyXman]

Fixes #3958

Changes proposed in this pull request:

• Add historic=archeological_site to #tourism-boundary.

Test rendering with links to the example places:

Athena: https://www.openstreetmap.org/#map=15/37.9742/23.7297
Italy: https://www.openstreetmap.org/#map=11/42.4126/11.9428
Roma: https://www.openstreetmap.org/#map=16/41.8915/12.4839
Messini: https://www.openstreetmap.org/#map=14/37.1780/21.9276
Olympía: https://www.openstreetmap.org/#map=13/37.6417/21.6304

After:
Athena:

Italy:

Roma:

Messini:

Olympía:

Roma becomes a little bit more busy, but that's because every ruin has been mapped independently instead of mapping the whole Forum. I would like also to see how Petra (16/30.3280/35.4458) renders, but I don't have it in my import.

Finally, I should also include names in labels.

[kocio-pl]

Sounds like really a kind of tourism attraction site. Almost half of them is mapped as an area/relation (27 088+943), so it makes sense to show the shape of the place.

[StyXman]

OTOH it will highlight places like https://www.openstreetmap.org/way/476382208 which look more like actual archeological excavations than tourist attractions.

[kocio-pl]

4.52% of all the historic=archaeological_site has a tag of tourism=attraction, this sounds to me like reasonable combination to make it different from typical excavation area.

[imagico]

I am fairly inclined to be against this - for a number of reasons:

• historic=archaeological_site does not in any way imply any sort of touristic importance.
• it has not been demonstrated that mapping of polygons with historic=archaeological_site has been done predominantly with meaningful and verifiable geometries w.r.t. the meaning of the tag. At a quick look most polygons with this tag indicate either (a) some physical structure like ruins of some sort, (b) some abstract wrapper polygon drawn around or (c) some sort of low level administrative boundary or land ownership but not a meaningful perimeter of the actual archaeological site.
• the proposed asymmetric outline rendering already causes a lot of issues in cases where it is used right now and it is questionable IMO if it makes sense to extend this to other application. This is essentially the swiss army chainsaw of digital map design: Yes, you can in principle apply it to almost everything but it is loud and smelly and creates a real mess.

[StyXman]

Let's add tourism=attraction, then. The actual style can be discussed, of course. As for the polygons themselves, maybe this is a way for people to take a second look what they mapped.

[StyXman]

Many polygons have disappeared, making Roma cleaner. Also, only one of the huge polys in Italy survives, and it's probably a bad tag.

Messini:

Athena:

Olympía:

Roma:

Italia:

[mboeringa]

Also, only one of the huge polys in Italy survives, and it's probably a bad tag.

If you mean the last picture near Lago di Bolsena, that is Parco Naturalistico Archeologico Vulci, and I don't think it is a bad tag. According to this site:

https://www.italiaparchi.it/parchi-di-arte-e-storia/parco-naturalistico-archeologico-di-vulci.aspx

the park itself is at least 9 km2, and encompasses one of the largest Etruscan cities - Vulci - and multiple necropoles with some 30.000 tombs. The city was destroyed by the Romans in 280 BC according to this website (and seems to have been rebuild as Roman):

https://vulciturismo.com/il-parco/

There also seem to be multiple opportunities for hiking, horse riding etc in the park.

[StyXman]

I wanted to also display the name and maybe an icon, and I found that we're already doing it, but in the amenity-points layer, and only starting from ZL 17 as part of the code that displays @culture names. I chose to make a new block and point to the places it has to be in synch with. Renders will come soon.

[StyXman]

Vulci:

Roma:

Olympía:

Athína:

Messini:

[polarbearing]

historic=archaeological_site does not in any way imply any sort of touristic importance

To reinforce that, there are archaeologic sites that are the exact opposite. I remember archaeologic probing along a planned motorway, and in one spot relevant objects were found. It created some local news about this heritage, the motorway was changed to bypass the spot. The site is listed as archaeologic, and mapped in OSM as such. Public access is forbidden, thus not touristic at all.

[polarbearing]

BTW, what is that big bright-red flower icon in the first post's examples?

[StyXman]

what is that big bright-red flower icon in the first post's examples?

My cursor, the screenshot app is including it :(

[jeisenbe]

I don't think it makes sense to only render the combination of tourism=attraction + historic=archaeological_site in a particular style. As mentioned, this is only 4.5% of historic=archaeological_site features, and it's an even lower percentage of tourism=attraction.

In the past, this style used to render tourism=attraction with a pink color fill. However, this tag has never been used in a very consistent way: as noted here, it can be anything from ancient ruins to carnival rides to a historic railway; anything that a mapper considers of interest for tourists. So in PR #1063 the polygon fill color was removed, with this comment:

"Rendering of tourism=attraction areas is not really useful, as it may be exactly anything and anyway this pink area rarely gets displayed."

Then in issue #1257 there was a request to "Add area render for tourism=attraction" and this was rejected with these comments #1257 (comment) and #1257 (comment)

There was also #1824 (comment) which declined rendering tourism=attraction lines.

More recently - in PR #3603 - we stopped rendering the tourism=attraction text label at lower zoom levels, so it now is rendered from z17 whether mapped as an area or point, if there are no other tags.

[jeisenbe]

It will highlight places like https://www.openstreetmap.org/way/476382208 which look more like actual archeological excavations than tourist attractions.

The motorway was changed to bypass the spot. The site is listed as archaeologic, and mapped in OSM as such. Public access is forbidden

I believe a large part, perhaps the majority, of historic=archaeological_site features may be closer to these. If we look at the key site_type, 37,000 features are a site_type=tumulus defined as "Tumulus - a mound of earth and stones raised over a grave or graves".

The next most common are "fortification" (stone walls and earth walls), "megalith" and "settlement".

Most of these are not at all similar to a theme park or zoo-like feature. The historic European city centres, with their world-famous historic sites, are the exception.

Because historic=archaeological_site does not appear to be similar to theme parks and zoos in function for most general map users, it would be better to render it differently.

A more reasonable comparison might be museums or other cultural features. Currently there is no polygon fill rendering for tourism=museum either, however, the @societal-amenities features are the most similar.

There was also some discussion about using #fbecd7 (formerly the farmland color) or #eecfb3 (the former allotments color) for urban cultural amenities (museums, theaters, etc). However, this needs to be discussed further.

[StyXman]

I don't think it makes sense to only render the combination of tourism=attraction + historic=archaeological_site in a particular style.

I did this to not render sites that are not open to the public. There is not much combination with access tags. Maybe I shouldn't worry about this at all.

[jeisenbe]

@StyXman would you be interested in trying a rendering with #eecfb3 as a solid polygon fill color?

[jeisenbe]

This would need to rebased for v5.0.0 - @StyXman, are you considering working on this further?

[StyXman]

I wish I had the time and energy. Family life has been intense this winter. I have no ETAs.

[jeisenbe]

Closing as stale. If anyone is interested in rendering these features and other cultural amenitys with #eecfb3 fill color, please try it out in a new PR

[soshial]

I also think, that all archeological polygons should have some boundary/fill drawn. Currently, only the center with icon is shown.

Carto:

In the editor: