Idiotic entries in OpenStreetBugs

Gibuld · 2009-11-13T09:11:13.000Z

When regarding the OpenStreetBugs I´m meanwhile somehow disappointed. Many of the entries seem to origine from people who obviously misunderstand this tool.
Due to my understanding OSM is a project still in development. I myself have mapped this year various villages in my environment, first beginning with mapping the residentials and the streets connecting the villages.
When I now find OpenStreetBugs telling (Not in my very region):
Track is missing
This street doesn’t end here
Traffic sign missing
Shop is missing
Building is missing (he means : the shape of the building!)
Still incomplete
Somewhere here is a little park with benches
Could be a wet walk in this region
Ski lift seems to have more columns
amenity=fountain or man_made=water_well? (where the map still hasn’t any according entry)

Then I`ll confess: Yes, you fool these things still are missing, like many other things which still are missing. And I’m just waiting until all missing objects have been entered in the Bugs-Tool until I will continue mapping. BTW: Why didn’t you map yourself what is offending your eyes?

Short summary to OSM: Please Announce to the users of OpenStreetBugs clearly, only to enter BUGS !

I deny to manage such entries in OpenStreetBugs, but I’m glad about any report of real bugs, which I really do manage. And: I’ll continue my way of mapping in the order that I find convenient.
All these dummy reports lead to the situation, that OpenStreetBugs seems to be treated somehow irrelevant by the OSM mappers, since many real bugs have not been treated for months.
If the reader cannot follow my statement: I’m speaking from Southern Germany and the Alps. Perhaps in other regions OpenStreetBugs has more meaningful entries.

Alex_McKee · 2009-11-13T11:19:26.000Z

I’m afraid I’m guilty of making such entries. It can be useful to have a reminder. Also sometimes the map data looks complete, but isn’t because of new development/redevelopment. In these cases, which is the type of reports I’ve made in my own municipality, it really is a problem with the existing map data.

Vclaw · 2009-11-13T13:18:18.000Z

I don’t see the problem with marking missing / incomplete streets etc in Openstreetbugs. I’ve done it a few times myself. Its useful if I notice something is missing, but I’m not able to survey it properly then, so I have a reminder to return and map it. Or another mapper in the area can fix it.

Though I suppose it might be useful to have different categories of bugs. Then you could choose to only display “missing street” bugs, or wrong name bugs, or whatever type you want to see.

Alex_McKee · 2009-11-13T14:23:21.000Z

Concur, it would be very useful to be able to categorise bugs.

Gibuld · 2009-11-13T17:20:14.000Z

Well, may be it would be helpful to categorise bugs in OSM.
Unfortunately this will not answer, who is responsible to care about the reports. I feel there is an increasing amount of reports “missing something”. And it seems to me most such reports originate from “sniffers” who are redundantly running around, finding single missing objects, but have no idea about the state of the mapping progress, neither are willing to map themselves.
At least I can say I feel bothered when I’m the only person to map a village in situ and then comes a guy and reports “here is a missing fountain” – anything else - while I’m about to trace house numbers.
Who will care about such message ? Since I’m the sole real mapper out there, only I could do it. But I will not – I have to map, since I feel responsible for villages and hamlets of about 10k people and their environmental streets. One fine day, perhaps, the fountain will interest me, but for navigation the fountain is irrelevant, as irrelevant as the shapes of buildings.
I’ve found regions, where you can wait long time for potlatch because of the huge amount of buildings to be loaded, while nearby many countryside roads are missing.

Using OSM bugs as a personal reminder is somewhat out of my imagination.

Lambertus · 2009-11-14T16:46:08.000Z

Whey are people adding poi, landuse, buildings, powerlines, marine attributes, pubs etc? It’s OpenStreetMap you ignorant mappers!

Well, we’re not only mapping streets these days anymore do we, so why not only bugs in OpenStreetBugs? Perhaps because the other tools are too complicated for a lot of people who just want to report that there is still a bench missing in a corner of that park…

Btw, in a map environment missing data could be called a bug.

Gibuld · 2009-11-14T19:39:59.000Z

Yes, it couldbe called a bug. Perhaps its your definition of a bug. My definition of a bug is: Something is wrong and prevents correct operation – in our case the correct navigation with OSM maps - you know, also your nice map would be incorrect. As long as mapping of a region is still in progress, there is a lot of missing objects. What do you think, the OpenStreetBugs would look like, if all the people living out there would report those missing objects they find worth being mapped, but only one mapper really is willing to map – and additionally he should look after the “bug” messages. Months ago I’ve reported in OpenStreetBug seven real bugs. I just reviewed their state: No report has been closed or commented, but 1 is solved, two others have been reworked as “fixme”. What does this tell us: The OSM mappers don't seem to work on the reports of OpenStreetBugs. Instead, occasionally they find themselves their mistakes. I´ll now close myself the solved reports. In the future Ill save myself this work (always looking: has someone solved – and learned from his failure?) and do the task myself silently, since unconnected streets are really disturbing when you use MS to enter a route with waypoints.

Btw: Unconnected streets seem to be a very common failure in the OSM data. Many of those failures may originate from the time where potlatch was unable to visualize connections. Sometimes it is very clear, where a connection is missing, in other cases only on site knowledge could give a correct answer.
So I would propose:
There exists a tool which is able to scan the database for unconnected streets (don’t remember it’s name). This tool should be run on the whole database. In case that problems were found, the last member who touched a problematic object (at least one month ago) should be informed by email, he might review that object.
I’m writing this, because I use OSM maps for traffic navigation and this is not always a pleasure.

Yes, meanwhile I also see the advantages of prioritizing the OpenStreetBugs. If then there also exists a filter to only display real bugs, perhaps the OpenStreetBugs would find more attention amongst the members.

amm · 2009-11-14T21:09:27.000Z

I suspect the tool you are referring to is keepright ( [http://keepright.ipax.at/]) and is indeed a very useful tool to find various different “bugs” in the data.

However automatically correcting things via a bot, or spamming people with error messages they haven’t asked for is imho a bad idea. I would rather try and promote the various Quality Assurance tools ( [http://wiki.openstreetmap.org/wiki/Quality_Assurance] ) more prominently.

Gibuld · 2009-11-15T09:22:33.000Z

Don’t know whether it is keepright. Curiously there is only in the Netherlands discussion forum an entry about keepright - which I don’t understand.

You are right, automatically correcting something is dross, so nobody has proposed this. I don´t think I would consider it as spam if OSM would send me one or several mails with the following content:
Dear OSM fellow. OSM just has run keepright on its database. It also reported possible problems with objects that you have worked on as the last person at least one month ago.
Could you please verify the reports on the following objects.
Following now a list of links into the OSM-Map, each link containing an arrow to the object in question (same as the arrow when doing a search in the OSM map).

To give you an impression of what I mean comprehensively with my statements:
This Wrong Crossing I corrected yesterday together with three other failures of same quality. If you have an OSM map of that region, youll still be able to proof the failure. It is only necessary to try to create a route in MS and every few minutes you'll find such failures. I say : Each such failure is more serious for the correctness of an OSM map, than if all the surrounding kennels were missing. And the obviously missing residentials (visible within the buildings) also would be more important than the biuldings shapes.
And since those failures are very common, I can say: the OSM data contains thousands and thousands of such failures. The above failure was clear and easy to correct, but in many cases only persons knowing the precise situation can solve a problem.