Biztos, hogy ez a „Családi menüs hely” nevű étkező a mező közepén van? Biztos, hogy ez a neve?

Thanks for editing Hungary from... Brasil? :-)

Nevet nem kell nekik adni; amúgy nagyon jó, hogy ha felveszed ezeket, mert minden turista vágyálma, hogy megtalálja nyáron őket.

I have reverted that. You should not translate hungarian to "hungarian", and definitely have to check your spelling before editing.

<insert cake here>

Erre, de itt nem látom a <number>-<number> formát. Én vagyok vak?

Ez a direction=* forma nincs a wikin dokumentálva. Honnan van?

:-)

Wo? Deutschland oder near Nawababad?

This is an interesting discussion which desires a more verbose medium than this comment section. As I have replied in the issue: it is a matter of taste whether one avoids multipolys or not and definitely not a policy or a technical requirement. I'm sure there has been a discussion of multiple overlapping ways vs. non-closed multipoly members but I'm also pretty sure there wasn't any conclusion, so it's still a matter of taste. (Similarly I haven't been able to find how to map multiple large POIs inside a larger building, apart from indoor mapping schemes.) When a construct is valid it should be handled by OSM processing systems no matter how unexperienced editors may be confused by it (they can be confused by lots of things in the db indeed).

Technically this object now represents the "avoid overlaps" believers: the one physical building is shared between two entities: one half is school, other half is governmental, and there is no connection inbetween but a shared inner wall. I cannot even change the outer edge to _one_ poly since half of it belongs to school and half to the gov't multipoly. (And no reason to use building:part since the physical building have the same attributes.)

All in all I understand your points, I agree with their subjective reasoning and I opt not to use them here, since the current construct is just as valid and should be handled, and it's a good test case that way. :-)
(As you may have noticed just 20m north of the school I have tagged POIs as nodes inside the simple poly building, and I usually do it that way.)

Picit még javítgattam rajta, benne maradt egy gazdátlan vonal meg a benzinkút, viszont kimaradt egy szántó. Gondolom potlatch-ben nehezebb látni, mint nekem JOSM-ban. De jó lett, köszönjük!

Just a note: when you put access=no on a node on a way it may affect a route planner for car traffic to avoid the node. You probably should use foot=no on a closed pedestrian crossing.

Nincs boundary=small_region key. És igen, végignéztem majdnem az összes boundary értéket hogy látok-e jobbat. Az admin_level=7 sem random: megyénél kisebb, járásnál nagyobb.
De átteszem akkor valami láthatatlanra, meglátjuk a renderer meg a nominatim mit kezd vele.

Teljes mértékben nyitott vagyok a javaslatokra. Mit javasolsz?

Átnéztem kb. 3-4 féle lehetséges megoldást, de egyik sem tetszett igazán. Ezek földrajzi régiók, de a földrajzi régiókat admin aboundaryval jelöltük, mint az Alföld, Észak-Magyarország, stb. Ezen analógia alapján lett boundary, valamint a definíció szerint, vagyis hogy ezek a megyéknél kisebb földrajzi területek, melyeknek nincs fizikailag látható határa.

If you see problems or have advices feel free to act positive and supportive! :-)

Ennek mi volt a forrása? Felmérés volt?

Thanks!

Two is still not a lot. You probably cannot find a dozen since 2008. :-)

Two sidenotes: 1) History tab is completely useless, exactly for that reason. Use WhoDidIt or achavi or other tools which draw bounding boxes around real edits and not the maxed out area.

2) you probably see that it was a semi-automated edit through Level0 and you may have guessed (since you didn't look) that there were a few similar tag repairs on the _worldwide_ map. Writing separate, extremely detailed descriptions for changes containing just a few objects is just not realistic. As you say: checking a changeset with only four object "really wouldn't have been that much extra work". :-)
But next time I'll try to put even more description in it, okay?
(If you wonder: these fixes were based on taginfo and visibly mistyped tags. Worldwide. And I usually don't touch it if there's more than a dozen or isn't absolutely visible that it wasn't intentional. Part of the QA efforts. Try it sometimes. :-))
Ps: I only wish that those 90%+ of the edits would possess ANY description at all! :-/

Hi! Since you have used plural would you be so kind to point out my "changesets" which cover most of the planet? I'm sure you have dozens to point to. :-)

As for this _one_ changeset please notice that it contains _four_ individual objects. You are expected to be able to check manually what's it about (if you cannot please tell me and I'll gladly help). But even if you don't, the description really tells it all: Fix, Place, Typos. It contains a "fix" for "typos" in the "place" key's values. And really exactly that, and only that. You surely don't expect me to include the changes itself in the changeset?

Also examining the change you may realise that it's worldwide since the typos were worldwide, and few in numbers.

So as much as I would like to help you I see no way to do it differently next time. If you do, please tell me and I'll try to do it better next time.

Thanks!

/thanks :-)