Random Note tool

Out of curiosity do you track some stats? I wonder how long it takes until all notes in a country have been seen at least once?

Not directly, just observing seq_number (see the source code) per country for those notes which weren’t removed from the queue by someone closing them.
Really, only the smallest countries had all their notes visited.

I also found a bug in PostgreSQL which caused random values to be quite biased when used as a default column value. This was quite visible as the tool returned too much of new notes. This is fixed now, I call random() explicitly in the insert query.

1 Like

huh

is it WONTFIXed anyway somewhere or have you tried filing a bug report? Sounds like a nasty trap.

Well, it’s not a bug in PostgreSQL after some extensive testing. NorthCrab found a bug in my algorithm. If you sort by rand and retire lower values to the higher sequence number, and insert new notes with the current sequence number they have the whole 0-1 range for rand. I will see how to fix this, the good thing is I can always reshuffle notes just by bulk updating the rand column without affecting the state.

1 Like

The point is that now 2-3 people tried to close old Notes around here just because they are old. But:

“There is a connection between this road and the road to the south through the woods”

is still valid after 8 years, and nobody bothered to go there up to now. Its not a reason to close this.

So i am a little annoyed these tools motivating armchair mappers on the other side of the globe to mass-close Notes which may be valid just to tidy up the map.

This is definitly not a sustainable way of handling Notes.

Flo

9 Likes

The issue of hastily closing old Notes on OpenStreetMap is indeed a concern. However, this seems more like a matter of educating new users rather than a limitation of the tool itself. Similar to OSM’s official Notes layer, this tool offers an alternative way to find unresolved notes. It’s important to remember that these tools are only as effective as the users who use them.

Just because a small number of users do not understand how to properly do armchair note mapping, does not mean that we should not provide new and fun ways for the rest of us to do so. And for the same reason, just because a small number of new users do not understand how to properly use the map editors (iD, JOSM, etc.), does not mean that we should not provide them freely. :slightly_smiling_face:

5 Likes

Its definitly not a matter of new users. I have users which are with OSM for more than 10 years which tend to simply close old notes to “tidy up the map”.

And asking them it often shows that they

a) Havent had a look at the data
b) Didnt understand the issue i was describing in the notes but didnt care to ask
c) Did not care about local community to decide on the matter

For me a Note is always a matter for the local community.

I open notes when i do armchair mapping and find stuff i cant decide on Aerials or other information. So my notes always say “Someone on site needs to verify and find out” and then some armchair mapper on the other side of the globe closes that Note. And after more than 15 years looking at aerials i think i can see a LOT more than the casual mapper and find disambiguities in data to the aerial.

Its a PITA to every morning process 5-10 notes. Most are even to lazy to write whether they changed something and why and how they got the information.

So you open up the editor and verify something has changed, reopen the note and ask why they closed or if they misunderstood the issue.

Flo

1 Like

If a person is not properly educated on how to contribute early (new user), there is a high chance they will continue to make the same mistakes (old user). There is a direct correlation between those 2 stages.

Its always the same people - and i tell them over and over - and they dont care.

And this is why i dont want more people be motivated that closing notes is a good thing.

its not

Its a good thing to close done Notes. But they need to be done. The motivation should not be the simple closing of the Note. And all these tools just reward closing notes. Not fixing them.

1 Like

I really don’t see how this tool specifically rewards anybody for closing notes. It simply opens a random note and does nothing else. Do you see a problem with showing people random notes? If so, what’s your opinion on the OSM official notes layer, as it also displays notes at pretty much random.

I designed the tool with that in mind:

  1. it’s fast so that one can skip through a lot of notes they don’t feel like solving
  2. it won’t show the same Notes repeatedly more than absolutely needed

I haven’t seen anyone in Poland use my tool to mass close old Notes without actual resolution of the problem.

2 Likes

Not always. For example I closed several hundred of notes scattered across Ukraine at power=tower nodes, all with full note content “?” and created from anonymous accounts.

(though I have seen that local community is also mass-closing them as a vandalism)

have you tried contacting DWG?

1 Like

I wonder if this could be ported to MapRoulette to get even more eyes on it.

The requirement is to have near-instant updates so that closed notes won’t be shown. I don’t know whether MapRoulette can achieve that.

I think the source of the problem is that some people think notes are a ToDo list, but for many mappers, notes are reports from people in the field that identify an issue between the map and reality.

Creating a note from the computer has no value; at least it is strictly related to the tagging (incorrect tags, names). But notes should not be used as Tasks: “Please visit X place,” or “Please confirm,” because all the OSM data should be confirmed somehow, so there is no value in those notes that forces armchair mappers to go outside. For these things, other tools are more appropriate for specific themes, like MapRoulette, Tasking Manager, or even the new FMTM. To report incorrect mapping, there are other QA tools, like KeepRight.

That’s why closing an 8-year-old note in the middle of a dessert that says: “Please validate if this bin is still there” has no value. No one will drive hundreds of kilometers to verify if the bin is still there; there are many things to map in these 100 kilometers, and the most probably is that the note was incorrectly located.

Let’s remember that notes come from OpenStreetBugs, and this tool was used to report errors in OSM, not to create a to-do list or list places to visit.

Finally, a tool like Random note is great to close those notes that no one will ever visit.

1 Like

It really depends where in the world you are. For some areas that I am familiar with, I could not disagree more.

For example, in a city where almost all POIs have been mapped, the note “I heard this restaurant has changed name, need to have a look to confirm” is very useful.

It would really limit the functionality of notes if we weren’t allowed to use them for things that need surveying.

I often create, visit and resolve these kinds of notes, and sometimes I take detours on walks just to visit them. When you want to armchair-map, you might be annoyed by notes that need a survey; when I am planning my cycle ride home, I might only want to see the notes that actually require a survey.

What would be very useful is if we could distinguish between “armchairable” notes and those that require a survey… and those that don’t really require any action right now, just another visit in a few months’ time.

Maybe we could agree on a set of hashtags, so we could better filter out the notes we don’t want to see…

4 Likes

I personally use this set of hashtags sometimes: Select hashtag for closing notes · Issue #26 · kmpoppe/noteSolver · GitHub

I have seen use of #surveyme for marking notes requiring survey (and which can be skipped when looking through notes fixable remotely).

1 Like

I hope we can get more support for hashtags/keyword-based filtering in tools, or tags on notes. Only if there is better support in tools will more people use it… (though NotesReview already lets you search by keyword).

Has anyone invented a hashtag that means “this can/should be done from an armchair, and doesn’t require a survey”? A tool like the one that this thread is about could prioritise those notes, and a survey mapping app could filter to exclude them.

2 Likes

Here are some hashtag proposals:

Interestingly there’s no hashtag for “this can/should be done from an armchair, and doesn’t require a survey”. I guess because this is the default and people would just solve the note? The hashtags seem to be needed to filter out notes which need a survey.