Debugging help - Mourne, Gullion & Strangford UNESCO Global Geopark boundary

An unclosed (multi)polygon will not render, nor should it.

If it is a de facto or especially a de jure (in GB) “national park,” then I think OSM is OK with adding a protect_class=2 tag. If not, but it IS a “UNESCO Global Geopark” (and it certainly appears to be such a thing), then add protect_class=98 and know two things: this will not render in Carto, and it will be correctly entered into OSM. You don’t like that it doesn’t render? Tough. You are “doing the right thing” by tagging it correctly.

“An unclosed (multi)polygon will not render, nor should it” - yes I know, hence me posting looking for help in debugging it :slight_smile: - I suspected I’d missed a bit somewhere, but I couldn’t find it!

As regards ‘not tagging for the renderer’ - I know that… (although tbf, that’s not quite the same as taking a tagging scheme you read on OSM Wiki as gospel - there’s loads of out-of-date stuff on there, where people dreampt up some tags a dozen years ago but hardly anyone used or supported them)

If you’d like me to fix it (by adding 11.2 meters of way), I can do that in about five seconds.

Yes please :pray:

It’s not a National park because the status is awarded by UNESCO, not a national government, although it sounds like this one consists of 3 AONBs the boundaries of which are set by government. From reading the wiki page it sounds more like a protect_class=98, but to be honest the use of protect_class=5 for National Parks in GB has always struck me as a bit odd :slight_smile:

There are more gaps than the original 11.2 meters I found, several in fact. Hang on, I’m working on it.

Furthermore, I’ve been rather involved with many people (both in my country, USA, as well as around the world both in wiki and in one-on-one email conversations) about this very topic, including this very same debate (and related ones). It is in “slow flux,” and conversations like this are actually helpful. It’s slow going, but it is going.

1 Like

It’s a new UNESCO Global Geopark (UGGp).
OSM Wiki’s blurb on the ‘protect_class’ tag says UGGp’s should be “protect_class=98”, but I’m wondering how many of the 195 UGGp’s on planet Earth are currently tagged in that way? :wink:

It’s not yet a ‘National Park’ in the UK legal sense, (although it has been put forward to be designated as one, it may yet come to pass)

Changeset/136765902 now makes this a properly constructed type=boundary relation: The “gaps” were closed, so this is now a fully “closed polygon” as OSM defines this. The area=yes tag was deleted (as incorrect). The protect_class=5 tag was changed (as incorrect) to a protect_class=98 tag, as Andy’s and @ramthelinefeed 's comments make it clear this is correct tagging. This means it will not render in Carto. However, I believe this is now much more properly entered into OSM. If and when GB declares this as a de jure National Park, feel free to change the protect_class=98 tag to a protect_class=2 tag, and then Carto WILL render it.

Yet to be done: * ‘wikipedia:en’ tag is set, but no ‘wikipedia’ tag. Make sure to set ‘wikipedia=language:value’ for the main article and optional ‘wikipedia:language=value’ only for additional articles that are not just other language variants of the main article.

Happy to help.

1 Like

Thank you, sir :smile:

1 Like

For completeness, now that that is closed that does load into the database behind

gis=> select osm_id,name from planet_osm_polygon where osm_id = -15904145;
  osm_id   |                      name
 -15904145 | Mourne Gullion Strangford UNESCO Global Geopark
(1 row)

(although it currently isn’t rendered there for the reasons mentioned above)

1 Like

That’s good! :slight_smile:

Out of interest, I ran an overpass turbo query for 'protect_class=98 and it pulled about 30 or 40 items.

There’s nearly 200 UNESCO Global Geoparks (UGGp’s) in the real world.
But I don’t know how many UGGp’s simply aren’t on OSM in the first place, never mind tagged as protect_class 98. (Also, class 98 includes other things besides UGGp’s).

Ahhh, the “the more (oddities, problems, things that might be fixed or improved…) I LOOK for in OSM, the more I FIND!” issue.

I’m smiling, I’m nodding. Go! There are icebergs, there are “tips of icebergs!”

As you can see from the results, that’s a real hodge-podge of things, varying from big marine reserves like this to small sites, some of which are already in OSM as something else, and some of which don’t really have a defined boundary.

I suspect that most aren’t in OSM and most aren’t really verifiable on the ground - the only way that I know about Fforest Fawr Geopark is because a pub landlord there told me about it - it’s just the western side of this national park (that is in OSM) and takes its name from, but is otherwise unrelated to, this feature.

I’ll have a think about rendering protect_class-98 features sensibly but I don’t hold out much hope - just a label in the middle of Derwent Valley Mills would not help much, and as mentioned above it doesn’t really have a solid boundary.

The things is, UNESCO Global Geoparks get some fairly strict scrutiny by the UN before they get awarded their status, with (I imagine) fairly tightly defined definitions of the geographical extent of the site - and they are periodically reviewed, so can lose their status if they don’t stick to the rules.
So they ought to lend themselves quite well to being mapped as clear cut areas.

(In the case of this new Mourne, Gullion & Strangford one, I have not yet seen UNESCO’s own definition document, just the fluffy press releases that they and the council put out)

I don’t know how strict the scrutiny is as Lochaber Geopark ceased to be a UNESCO one in 2011 and is trying to rejoin. These Geoparks don’t seem to have the same status as World Heritage Sites, which do have to fulfil and sustain commitment to certain criteria.

I know the Villuercas-Ibores-Jara Geopark in Extremadura, Spain is highly regarded locally in terms of tourism potential. So I do appreciate there may be a desire to see them on maps.

Hi again folks.
I had to do quite a big revision of the boundary for the Mourne, Gullion & Strangord UGGp (as it turned out the boundary UNESCO went with in the end was different from the proposed one).

Would anyone who is a whizz with a validator tool like to check I haven’t screwed it up, please?

Also - I have a horrible feeling that I have mucked up the Strangford & Lecale Area of Outstanding Natural Beauty boundary in the process - Relation: ‪Strangford & Lecale AONB‬ (‪1428604‬) | OpenStreetMap
(as it seems to have stopped rendering) - but I can’t find the bug. Can anyone spot the problem?

Yes, I fixed that yesterday, and it reappeared today. It was doing that that led to my question at :slight_smile:

I get an email when national parks “appear” and “disappear” in a rendering database - here it is reappearing:

>    -1428604 | Strangford & Lecale AONB
< (83 rows)
> (84 rows)

Thanks! :slight_smile:

Also, to be very strict, an AONB is not a ‘National Park’ in the UK legal defintion … although it is a similar thing (more or less a restriction on planning permission, in practice)

True, and they are tagged differently in OSM to reflect that. However, I do store them both as “national_park” in the database I’m counting them in.

1 Like