Followup of https://github.com/westnordost/StreetComplete/issues/844 (respectively of https://github.com/westnordost/StreetComplete/issues/844#issuecomment-364666534, which I consider a good idea).
The problem was, in short: Users may add notes for things they don't see on the map, because StreetComplete displays few things.
I would suggest a warning, at least, to make users aware that this is counterproductive:
Maybe just use the same mechanism as when you try to solve quests farer away. Display a message with a "Do not show again" checkbox before the user can enter a note.
This message should then explicitly state that not all things from OSM are shown on the map and one should re-check that before opening a note…
(already got two upvotes, so think this is a good idea)
Well, what kind of warning? What'd be the text?
Also, how big of a problem is this?
Text could be:
You are about to add a note to OpenStreetMap.
Please consider that StreetComplete does not display everything, which is added in OpenStreetMap. Before writing about something, you see is missing, please consult another OpenStreetMap app to check whether it is already displayed.
Also, how big of a problem is this?
Good question. We'll likely see…
/cc @RubenKelevra
I think that this is pretty good! I encountered some users who were surprised that SC creates a new note if they can't answer the question... Perhaps the wording there should be clearer too...
Regarding your text: I think there should not be a comma before "which" and before "you see is missing"...
Actually I only thought of this text when manually adding a note, not when canceling a quest ("cannot answer").
Maybe we don't need an extra alert dialog there, but there were some users surprised that they left a new note if they clicked on "Can't Answer". Actually, the wording is pretty clear to me, so I have no suggestion on how to to it better...
Yet again: I am not talking about the "Can't answer" dialog. That may be a different issue… I am talking about manually leaving a note with the bottom left button in the 4.0 version.
Was that enough feedback?
@rugk Can you give examples of notes that should be discouraged by this warning?
Ask @RubenKelevra, who originally had the origin idea.
@RubenKelevra Can you give examples of notes created using StreetComplete that should be discouraged by this warning? I think that it is problem big enough to justify time to create and maintain something that will be just irritating for 99% of users (and probably be ignored anyway be people creating notes like that).
Well, users usually want to add missing features to the map, so if you see something and you don't find this feature rendered they might add a note like "here's a shop called Rumpelkiste". Which might already mapped, but is not rendered in SC.
they might add a note like "here's a shop called Rumpelkiste"
But is it really happening? I think that adding highly obnoxious warning, just because MAY add useless note is a poor idea.
Given that no POIs are rendered I expect that only extreme minority would make this kind of mistake and that anyway vast majority of such notes would be added by people who spotted something new in their neighborhood (so such note would be anyway useful).
Such warnings should be added only if where it is really useful, not just because it in theory maybe may be useful.
At least a couple of notes have been created due to ENTR8/streetcomplete-mapstyle#55. Like 1345405 and 1345413 (both by the same person). The notes created this way are only detectable by searching for lut.im (if they have a photo) because StreetComplete is not mentioned.
In that case warning also would not be useful, as roads are generally displayed...
because StreetComplete is not mentioned.
Oh, no. Indeed there is not even an edit or so, where one could look it up, as these notes do not have these. (in contrast to changesets). And all other notes by StreetComplete also mention it correctly…
So here is a new issue for that: https://github.com/westnordost/StreetComplete/issues/1012
@matkoniecz yes, this is really happening.
A different example:
https://github.com/westnordost/StreetComplete/issues/844#issuecomment-364649539
Maybe we should display it on the first creation of a note freely on the map, and a "dismiss for this session" button. But it's really important to add this information.
SC have a huge potential to be the first editor by newbies. I've showed it to 3 people and they started mapping with them, by solving tasks, because they love the idea of having a free world map like Wikipedia.
So we have to make some compromises here and there in SC to accommodate the uses by users who haven't ever seen OSM raw data or know what a simplified rendering is. All they probably know is Google Maps, which shows POI, our rendering not. 😉
It'd also be possible to have a checkbox within this warning dialog to _never_ show this message again. And, it would also be possible to i.e. only show this checkbox starting on the Xth time the dialog was shown.
I like this second approach, maybe 10 times is a good start? So in 10 sessions the dialog need to be closed before the user is able to hide it permanently. In this case we're sure, he probably read it more than one time.
Another thing is, notes just created don't appear on the map. And refreshing don't load a demo note I've just created either. There seems to be a somewhat substantial database delay, so writing them to the local cache after creation would be nice. :)
Another thing is, notes just created don't appear on the map. And refreshing don't load a demo note I've just created either.
This is issue https://github.com/westnordost/StreetComplete/issues/962. Please comment/vote/discuss that feature there.
StreetComplete is mentioned in notes since v5.0. So, have StreetComplete-initiated user notes become a problem?
not in my area, I have spotted no issues when I sometimes checked recently added notes worldwide
I also resolved some worldwide notes and only one or two users said in a note that a feature would be missing even though it was simply not displayed on the map of StreetComplete. So I would say that this is not that kind of a big issue...
Ok I'll close it then