On the HR-SRB border the border line should continue along the outer edge of Pocket 3 until it reconnects with the recognized border, because the land in between is disputed/unclaimed territory. Currently the map incorrectly treats this section as if it is a part of Croatia and not as a seperate territory. The map also sometimes glitches as you zoom in and does show the border. I tried every available layer and the only one that shows it correctly is the Tracestrack Topo layer map. In the editor it isn’t exactly marked with the line for the border, but I can see another grey line that doesn’t appear to be a part of OSM.
This is a link to the part of the map I am talking about with the layer Tracestrack Topo toggled on: OpenStreetMap
Source: Croatia–Serbia border dispute - Wikipedia
That’s apparently because @AustinAyers reinstated the Verdis nonsense in a series of changesets. In the process, they broke the international boundaries and e.g. the boundary of Kopački Rit nature park at that location.
TracestrackTopo layer is, as far as I know, either moribund or very rarely updated, so it did not catch that change.
I’m going to restore the status quo ante unless someone beats me to it. It’s probably better to undo the changes rather than fix them manually.
…the boundary way used to be way:1280813047.
Thanks for the quick response. Just to confirm, the area marked as Free Republic of Verdis will be removed/undid and the borders will stay the same as a disputed Croatia-Serbia border region.
Correct.
I could not make a revert myself using revert.monicz.dev, because of Overpass issues.
If anyone has a better tool, the offending changesets, in temporal order, are:
183032057, 183032231, 183032390, 183032494, 183032523, 183032760, 183032942, 183033037, 183033085, 183033328, 183033404, 183033491, 183035316, 183035369, 183035396, 183035626, 183035717, 183035746, 183035949, 183038275
the first two being the worst offenders.
sadly RevertUI - Job 12721 also failed
@SomeoneElse - maybe you can revert this thrash?
For the record, a similar situation at Liberland was (IMO sensibly) solved using a place=region boundary relation 19735507, utilizing existing disputed borders, and was given alt_name=Liberland. However, AustinAyers bullishly deleted the correct boundary way, falsely claiming that “Croatia no longer disputes this land” and then went on to add Verdis as a full-fledged country.
Sigh. Message that they must read sent and reverted.
With regard to micronations, if there genuinely is something there, then it should be in OSM as something (but not an actual country). Molossia is a tourism=attraction, and that seems about right there.
Wikipedia suggests that there is nothing currently at “Verdis”, so there is probably nothing that should be in OSM.
“Liberland” looks more problematic in OSM. It has been added with boundary=administrative, which it obviously isn’t. place=region is a bit of a stretch of “Liberland”, but for “Siga” (which it is an alt_name for) it might be valid. There is apparently no one there currently.
fake administrative boundary removed in Changeset: 183155550 | OpenStreetMap
further fixes may be needed
Yes, it’s probably best to stay away from anything that uses a public Overpass server that will likely be overloaded if you’re going to use the results to immediately update OSM (as well as https://github.com/Zaczero/osm-revert , also https://github.com/Zaczero/osm-relatify and any others).
I’m not familiar with that one, but the Readme has lots of caveats about why it might not work - I’d stay away for that reason.
Good morning folks,
I wanted to apologize for a change made to a pre-existing boundary. I did not think that I moved that relation on that changeset. If I did - I am sorry.
With micronations, I know some are serious, and some, are not serious. Some are just “for show”.
Reading through recent articles published by major news outlets, it appears that Croatian authorities no longer claim Pocket-3 as part of Croatia. And - Serbia has consistently stated it is not theirs either.
Source: NBC News
There are a few other sources, worded differently, that are independent of Verdis’ claims.
My question to the OSM community is at what point, can you them add them as a Country/Nation? They just openly met with the United Nations and had their blessing and support (unlike any other micronation in the region).
Thank you,
Austin Ayers
Thank you for reaching out.
What you’re claiming, however, contradicts what is said in that NBC article, which is pretty sympathetic towards the idea. Few relevant quotes:
“In October 2023 is when we started a settlement, and a few days later Croatian authorities quickly destroyed that settlement,” Jackson said, noting that he was given a permanent ban from Croatia
For now, Verdis only exists on a website where people can apply for citizenship.
he is planning to attend a protest outside the Croatian embassy in London, […]. “We will also hold a general meet-and-greet at a local cafe or pub after the protest,” the website says.
Jackson says he now lives in “exile” in Dover, where he socializes with friends and plays video games.
In other words, the whole “country” is just an elaborate joke.
I’m amenable to adding a place=locality point or relation at that location, so that anyone interested may be able to find it, but no admin_level nonsense. That also has a downside that it well be used as a honeypot for petty semi-vandalism like yours.
The specific portion of the article was:
“Known as “Pocket 3,” the strip of forest roughly the size of the Vatican was among several left unclaimed by both countries after the Croatian war of independence ended in 1995.”
Again - what is needed to be done to get Verdis recognized as a country on OSM?
What piece of document (United Nations, other GIS mapping, etc…) is required for country recognition on the platform?
There was already a locality=place done prior to my edit, I believe.
Have you looked at previous discussions on related topics on this forum? There was a thread on micronations a couple of years ago. You might also want to read about Somaliland and the SADR (the latter at least is linked from the micronations thread).
I think it’s clear that Verdis is nowhere near the level of “countryness” of Somaliland - which in turn has so far not reached a consensus to classify it as a country in OSM.
So looking through a few of the links you provided:
Once other countries begin to recognize them, and, their passports are accepted at other countries, then it can be added.
Thanks!
I think that has since been undone by the same mapper (if I’m looking at the right object).
The relevant relation, originally added by @JeroenHoek in 2025, is 19478817. It has been tweaked by several mappers since, including Badojo.
I don’t have a particular opinion about validity of that. It surely is a terra nulius claimed by a joke microstate that garnered some publicity on slow news days. How do we express that in tagging… meh.
Note that all we do on OpenStreetMap (which I did introduce there) is claiming that there are people who call that piece of land, that location, Verdis. This is a claim which is easy to verify, because Verdis has been in a bunch of media.
Mapping it as a multipolygon is sloppy mapping, and I suggest restoring the boundary type (along with the disputed status). Places can be defined as a boundary type without laying claim to the title of country. You can map the boundaries of any place with type=boundary (including localities and neighbourhoods).
As this is a locality with a very exact and known boundary, omitting that is wrong. @Badojo why did you remove the boundary type? Places are not multipolygons.
Obviously, no admin_level of any type applies here at the current point in time.
I am strongly against as there is no real administration of any kind there. You may as well map boundary of real estate manage by home owner association as boundary=administrative