Surprise, one-character notes can be house numbers.
#ReopenedWithoutComment
“Description: This process addresses instances where a contributor reopens a closed note without leaving a comment explaining the reason for reopening… If there is no response, the commented note is closed after approximately 20 days.”
Isn’t it obvious that newbies often accidentally open and close notes?
#OldNote Description: This process comments on old notes that were created more than 10 years ago. … If there is no response after approximately 20 days, the note is automatically closed.
This is absolutely terrible. So we will soon begin to close notes from MAPS.ME indicating that POIs are closed.
And yes, even old notes can be relevant in places where there are few mappers.
I realize we all want to resolve all the notes as soon as possible. But not like this, please.
Thank you for creating this thread regarding the mentioned OSM note processes from my bot.
After the recent changes and test runs, opening a thread here in the forum was still on my to-do list for today.
Currently, the bot’s goal is to regain more attention for already created notes. While it is mentioned in the profile description that notes could be automatically closed at some point, this process is not yet active permanently.
That is true. But just a single number without the information that it actually is a housenumber, has no information in it. And if the author of the original note answers to the question from the bot, the note will not be closed.
Yes. That is why the bot asks “Was this accidental, or does the reported issue still persist?”. If this question is answered, the note will not be closed.
Here I partly agree with you. But a node that is 10 years old is probably not up to date.
By first commenting on the node and then waiting for a reply, this goal is achived. If there is no response in multiple weeks the people who at some point were interested in this node, are obviosly not interested anymore. So I do not see much of a loose closing these nodes. It makes it easyer to focus on relevant nodes that include actuall data that can be mapped.
I wonder if some people would create a note for each individual house along a block. You’d be able to tell pretty easily that they’re house numbers when looking at them on a map, but not just looking at the text in isolation.
Then again, the better solution would be for osm-website to imitate Discourse and impose a minimum character limit, so that notetakers don’t accidentally waste their time.
The original post pointed out something I hadn’t considered before: that if the note is about a POI that has closed, it could remain relevant much longer than a note about a POI that has opened. Then again, I’ve encountered notes about closed POIs that have since been deleted from OSM by mappers who weren’t looking at the notes layer.
On a slightly different tack (and this is just my opinion, so feel free to ignore):
The “ContributionReviewer” name, and some of the language used, gives it a quasi-official appearance - when in fact it’s just some guy like you or me.
For example, it says:
Despite our previous attempts to gather additional information for your note, we haven’t received a response, and it seems there has been inactivity.
Due to this, we’re closing the note now.
Who’s “we”? The new mapper will assume it’s OpenStreetMap corporately.
Could you consider renaming it to “NeisBot” or something less official-sounding, and changing the language to avoid the corporate we?
Yesterday I saw these comments from the bot, and so far I haven’t seen, for Greece, any note closure. Personally, I find those comments useful since they are most likely indeed by mistake rather than actual suggestions, and them being in a dense area with several other notes nearby, helps clearing them out easier before going area by area.
Yes, thank you @Richard
I agree, the name of the user profile as well as the wording in the comments sounded too official.
Both have just been adjusted by me, see NeisBot | OpenStreetMap
Somewhat related, I’ve just commented on a note - perhaps NeisBot’s handler could consider training it to recognise when geometry issues pretty obviously still exist
Hey, first of all thank you for yet another good tool!
But there’s a problem with the pattern recognition. This note among others were incorrectly closed, and the bot also used two undocumented hashtags. The hashtag #StartEnd seems to suggest that the bot searches for “start” and “end” in notes, but this note contained neither; the closest thing would be “endast” wich has nothing to do with start or end (start or slut in Swedish). For Swedish, it is actually preferred that it checks for compound words (such as in this case, where the closing was correct), but then it needs to check the correct languages in the respective countries.
Heyhey, thanks for the hint. I noticed it last night when you reactivated the note. I closed the two notes manually, but unfortunately I didn’t read my translated text correctly. I apologise for inadvertently closing the note. There are currently no plans to implement a process in the bot for the mentioned #hashtag.
