@xamanu while this is not really the right place for a tech dicsussion. gitea uses https://github.com/markbates/goth for OAuth support and while most of the providers are for OAuth 2.0 support at least the Xero one is 1.0a. So I’m not quite sure what would be “quite difficult” about adding OSM, work yes.

In any case does your answer imply that gitlab supports (mis-)using 1.0a?

Did the group consider recommending gitea over the (incredibly clumsy imho) gitlab?

gitea while not perfect would seem to be nearer to what we would expect from FOSS and it would seem to be substantially more realistic to implement something like authentication via osm.org than with gitlab.

osmand supposedly supports conditional access for some combinations.

But nothing is likely to support what you tagged: - maxweightrating is essentially unused compared to maxweight - conditional restriction tagging in turn restrictions isn’t supported by anybody afaik (and the comment wrt maxweightrating applies even more here) - in general the conditional tagging grammar is massively underspecified and mainly serves to humour people that think stringing long combinations of tags together is a good idea and will actually do something.

The question is, why tag it in such a complicated and likely non-functional fashion in the first place? Simply add maxweight:hgv=10 to the road segment going straight on (and if necessary any thing going off to the right) and that should force router that support maxweight:hgv to not plan the route straight on and plan a turn to the left (if you feel better about it you can add maxweightrating:hgv tags too).

There are some known issues with 15.1 that will be fixed in the January maintenance up date https://github.com/MarcusWolschon/osmeditor4android/issues (or if you are building the app yourself as soon as you re-build).

@SK53 there are some underlying architectural issues, that, if changed, would make use on large screen devices easier. Essentially it is re-factoring to allow use of the property editor in parallel with the map display. This isn’t such an issue on phones as there is no space to display it any way (though there are some other reasons why this would still be helpful).

@apm-wa to clarify this again: I’m not accusing the board of dark motives, or anything else than the best intentions. I’m just pointing out that things seem to be confused with respect to how the working groups were operating up to now and the impact of the proposed change.

Just a comment on “ I originally proposed it as a working group but was informed by old-timers that working groups may ONLY be created by the community, not by the Board,” the old timers in question must of have suffered a lapse of memory because if anything the exact opposite was true, see the board minutes I linked to.

Naturally creating a working group without community buy in the sense that there are actually people prepared to work on it is difficult, so obviously creating the working group and having people willing to serve on it tend to go hand in hand which is what might have caused the confusion.

There is nothing stopping the board from creating an additional agreement outside of the NDAs that would define the space in which volunteers can act. But in any case if the board is concerned that it can’t, legally squeaky clean, delegate responsibility to non-board members, then it doesn’t just have an issue with volunteers and the working groups, it is going to have one with employees too. So best completely revamp this in the AoAs instead of doing something that in practical terms is just going to cause issues.

_ But if they’d create a working group or a different informal structure for that, they’d run into the problem that such entity could not make any legally binding statements towards personnel (like instructing them what to do, handle requests for leave, dealing with payment matters, approval of deliverables/invoices from independent contractors etc.)_

What do you think the working groups have been doing the last decade+? Naturally within an agreed remit and budget, and with the purse strings held by the treasurer (in a larger organisation that would typically be delegated to staff too), but de facto this has been done by the WG, who do you think signed off on all the trademark related bills, or who orders and checks hardware? In the real world and not lala land, boards of companies do not actually do all the work themselves.

The solution is to simply add addresses (with postcodes), naturally only if Malaysia uses something that fits in to the Schema we currently use osm.wiki/Key:addr If it doesn’t that is something that would have to be discussed.

And this is the app https://www.refill.bz.it/?mz=9&mc=46.640008243515915,11.351623535156252 looks harmless, probably just buggy.

@SomeoneElse see for example osm.org/user/danielbir with > 6000 empty changesets.

@SomeoneElse I was actually prompted to have a quick look at the empty changeset numbers, and they have doubled from the usual ~5’000/month since August and the “Refill Südtirol” app is the major culprit (for the increase, the rest are the usual suspects JOSM, maps.me and id).

Generating a list of user ids as I write.

There are regularly updated statistics on editor usage avaliable here osm.wiki/Editor_usage_stats

It seems as if you are actually using changeset counts and not edits as you claim, this is not only extremely mislieading and distorts actual use of the various apps, changeset counts are simply not a good metric for anything.

@CjMalone this has been tried numerous times and the result has been the non-commital statements we’ve receive to date (not an outright no, but not an outright and clear, official “yes” either).

Please note that when you are using this Strava layer that there is no actual permission to use the data and you are running the risk of having your edits removed if there is trouble down the road (and that is why you need to use hackish workarounds to access the layer in the 1st place).

You don’t mention how the models you intend to employ were trained. Regardless of the source of the training data, you need to consider that there may be residual intellectual property from the source of the training data in the output of the models.

This looks like a known artefact of the rendering on OpenCycleMap, that is due to the ways being merged and ending up pointing in the wrong direction. The creator of OCM is working on the issue, but this make some time to fix.

@tyr_asd “if” is the key, it just isn’t clear right now if it includes the version or not. Regardless of ToU compliance or none, version specific blocking currently can only happen if the user agent includes the version (and that it is running on iOS I suppose), as there is no API internal way of blocking apps based on a version string or similar.

@stephan75 If the version isn’t reported in the user agent on upload there is no easy way to do that.