Re: Call for verification (Was: Re: VANDALISM !)

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
28 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

pangose
Here is yet another example of bad data in our database:

-------- Originalmeddelande --------
Från: Martijn van Exel <[hidden email]>
Skickat: 22 augusti 2020 00:33:24 CEST
Till: [hidden email]
Ämne: Re: [OSM-talk] Use of OSM data without attribution

Curious anecdote: some AllTrails user apparently looked up a phone
number for OSM US and called up Maggie. Turns out the complaint was
about a trail that I originally mapped *blush*. In my defense, that was
9 years ago, I haven't been to that part of town much since I moved, and
nobody else updated the trail, which has since disappeared.

Here is the changeset in case you're interested:
https://www.openstreetmap.org/changeset/89419938

Martijn
--

Maybe we should have some kind of system flagging objects that has not been edited for x number of years and rate all objects in the database according to this?

This would mean that a data consumer can decide based on the score if they want to include the information or not.

E.g. a high quality map should perhaps not contain objects with a revision older than 3 years (and no references or sources)

Or even better: we could implement a verification system with a log that can be queried easily.

IMPLEMENTATION SUGGESTION:

GET Openstreetmap.org/api/verifications/
Lists latest added verifications (outputs 10 entries, &offset can be used to get more, &size can be used to output up to 300 entries)

GET Openstreetmap.org/api/verifications/1234
Outputs verifications for osmid 1234 with the newest first (outputs 10 entries, &offset can be used to get more, &size can be used to output up to 300 entries)

POST Openstreetmap.org/api/verifications/1234
Add a new verification for osmid 1234

On openstreetmap.org we have a new button for every object "Verify this object exists and is correct" which stores the date and userid in the database.

In JOSM we could add the possibility to download verification data for all selected objects or from a new option in the download dialog.

The latest verification date and count of verifications could be made available in a separate dump.

If we had such a system I believe the map data quality could increase considerably by making it dead simple to hide hide old unverified data from e.g. openstreetmap.org. A high-quality map we can be proud of could also give an impetus to local mappers to revisit trails and verify them.

WDYT?

Cheers
pangoSE


pangoSE <[hidden email]> skrev: (22 augusti 2020 09:32:09 CEST)
Hi

80hnhtv4agou--- via talk <[hidden email]> skrev: (22 augusti 2020 03:06:37 CEST)

 
Also there is no wiki on unverified edits.
 

In OSM we don't yet have an established system for verification or accurate machine readable references for the data to my knowledge.

This means the whole database is basically just a mess of biased data that one of our millions of editors thought should be included. Most objects have very few revisions and we have no idea about the overall quality or correctness. It a playground with half-ass quality more than an authoritative and verified source of information (like e.g. Wikipedia). Building upon it can lead to strange things. E.g. https://www.nyteknik.se/popularteknik/mystisk-jatteskrapa-dok-upp-i-flygsimulator-6999771 (building:levels=212 was entered erroneously and committed to the database without any kind of QA follow-up. If someone knows the osmid I would like to know how long this error was present in OSM)

We should really fix this and start a verification effort after implementing a sane verification model.
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

General Discussion mailing list
Nobody claims OpenStreetMap data contains no mistakes.

Are you really expecting that we will be shocked by proof that
some data somewhere is wrong?

I would be able to post one mail per minute with examples of serious
mistakes, forever - even after my death, as it would be fairly easy to automate.

If something is wrong then fix it or create note or switch to other data source
(that WILL contain wrong/inaccurate data - or will be very expensive and cover
some limited area). Or ignore it.

Please stop posting about every single inaccurate data in OSM that you managed to notice.

https://wiki.openstreetmap.org/wiki/Notes are designed as tool allowing to
describe incorrect data that someone is unable or unwilling to fix (and
yes, we have thousands of reports of mistakes)

Aug 22, 2020, 10:12 by [hidden email]:
Here is yet another example of bad data in our database:

-------- Originalmeddelande --------
Från: Martijn van Exel <[hidden email]>
Skickat: 22 augusti 2020 00:33:24 CEST
Ämne: Re: [OSM-talk] Use of OSM data without attribution

Curious anecdote: some AllTrails user apparently looked up a phone
number for OSM US and called up Maggie. Turns out the complaint was
about a trail that I originally mapped *blush*. In my defense, that was
9 years ago, I haven't been to that part of town much since I moved, and
nobody else updated the trail, which has since disappeared.

Here is the changeset in case you're interested:

Martijn
--

Maybe we should have some kind of system flagging objects that has not been edited for x number of years and rate all objects in the database according to this?

This would mean that a data consumer can decide based on the score if they want to include the information or not.

E.g. a high quality map should perhaps not contain objects with a revision older than 3 years (and no references or sources)

Or even better: we could implement a verification system with a log that can be queried easily.

IMPLEMENTATION SUGGESTION:

GET Openstreetmap.org/api/verifications/
Lists latest added verifications (outputs 10 entries, &offset can be used to get more, &size can be used to output up to 300 entries)

GET Openstreetmap.org/api/verifications/1234
Outputs verifications for osmid 1234 with the newest first (outputs 10 entries, &offset can be used to get more, &size can be used to output up to 300 entries)

POST Openstreetmap.org/api/verifications/1234
Add a new verification for osmid 1234

On openstreetmap.org we have a new button for every object "Verify this object exists and is correct" which stores the date and userid in the database.

In JOSM we could add the possibility to download verification data for all selected objects or from a new option in the download dialog.

The latest verification date and count of verifications could be made available in a separate dump.

If we had such a system I believe the map data quality could increase considerably by making it dead simple to hide hide old unverified data from e.g. openstreetmap.org. A high-quality map we can be proud of could also give an impetus to local mappers to revisit trails and verify them.

WDYT?

Cheers
pangoSE


pangoSE <[hidden email]> skrev: (22 augusti 2020 09:32:09 CEST)
Hi

80hnhtv4agou--- via talk <[hidden email]> skrev: (22 augusti 2020 03:06:37 CEST)

 
Also there is no wiki on unverified edits.
 

In OSM we don't yet have an established system for verification or accurate machine readable references for the data to my knowledge.

This means the whole database is basically just a mess of biased data that one of our millions of editors thought should be included. Most objects have very few revisions and we have no idea about the overall quality or correctness. It a playground with half-ass quality more than an authoritative and verified source of information (like e.g. Wikipedia). Building upon it can lead to strange things. E.g. https://www.nyteknik.se/popularteknik/mystisk-jatteskrapa-dok-upp-i-flygsimulator-6999771 (building:levels=212 was entered erroneously and committed to the database without any kind of QA follow-up. If someone knows the osmid I would like to know how long this error was present in OSM)

We should really fix this and start a verification effort after implementing a sane verification model.
talk mailing list


_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

pangose
In reply to this post by pangose
Hi 😀

Mateusz Konieczny <[hidden email]> skrev: (22 augusti 2020 09:55:10 CEST)
>"It a playground with half-ass quality more than an authoritative and
>verified source of information (like e.g. Wikipedia)"
>
>I am not sure whatever you claim that
>Wikipedia is
>"playground with half-ass quality" or
>"authoritative and verified source of information".

I meant that a verification system does exist in Wikipedia and they now require references on all statements to keep up the quality of the articles which is sane IMO. We have no such system.

>OSM would benefit from better verification
>tools and so on but insult-laden post
>filed with misunderstandings will not
>lead towards them.

Sorry if it came across as harsh, I get your point and will try to moderate my criticism a little more.

I love OSM and have contributed a lot over the years and recommend it to everyone I meet who uses maps.

I just sent a follow up email with a suggestion for implementing such a verification system.

I still believe we have data with bad quality in many places (in Sweden). I have to fix stuff often when I visit new places apart from all the stuff we are missing. We are basically trying to keep up with an ever changing surrounding without a good way to indicate our data quality.

We can do better, but we need a new system that make it easy for contributors to verify our precious data (see my previous email).

Cheers
pangoSE

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

pangose
In reply to this post by General Discussion mailing list
Hi again

Mateusz Konieczny via talk <[hidden email]> skrev: (22 augusti 2020 10:20:51 CEST)
>Nobody claims OpenStreetMap data contains no mistakes.
>
>Are you really expecting that we will be shocked by proof that
>some data somewhere is wrong?

No. Are you shocked by my constructive criticism and constructive suggestions for improvement of what I perceive as a problem?


>Please stop posting about every single inaccurate data in OSM that you
>managed to notice.

You seem to have the impression that I'm going to spam the list with examples of errors. Where did you get that idea from? I only included Marjins example because it showed 2 things related to my call for verification:

1) people capable of mapping correctly can produce map errors because our subject changes over time. This means that what I map today correctly might be incorrect next month or next year.

2) Martijn had the intention of producing high quality data but we as a community did not help him archive that over time because we have no system e.g. that could alert him if an object he created has not been verified/changed for the last x years. I would love to have a telegram bot notify me when one of my objects has passed a certain threshold of staleness.

Cheers
pangoSE

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

General Discussion mailing list
In reply to this post by pangose
(1) Wikipedia may strongly encourage or mandate it in theory, but there are
still edits being made without any citations

(2) Wikipedia is explicitly forbidding original research, OSM is explicitly encouraging it
The best edits are where people map things not mapped anywhere else,
or at least not mapped in any other open data source.

It makes impossible to require citations for everything and requiring people
to contribute to Mapillary or equivalent would be an unreasonable burden.

(3) Yes, better verification tools would be likely better.

(4) Have you seen
(BTW, I really need to finish my resurvey opening hours quest for StreetComplete).

(5) Anyone with any real knowledge of OSM is already aware that it has mistakes
"Hey, here is proof that OSM have mistakes" is not interesting in any way.

Aug 22, 2020, 10:26 by [hidden email]:
Hi 😀

Mateusz Konieczny <[hidden email]> skrev: (22 augusti 2020 09:55:10 CEST)
>"It a playground with half-ass quality more than an authoritative and
>verified source of information (like e.g. Wikipedia)"

>I am not sure whatever you claim that
>Wikipedia is
>"playground with half-ass quality" or
>"authoritative and verified source of information".

I meant that a verification system does exist in Wikipedia and they now require references on all statements to keep up the quality of the articles which is sane IMO. We have no such system.

>OSM would benefit from better verification
>tools and so on but insult-laden post
>filed with misunderstandings will not
>lead towards them.

Sorry if it came across as harsh, I get your point and will try to moderate my criticism a little more.

I love OSM and have contributed a lot over the years and recommend it to everyone I meet who uses maps.

I just sent a follow up email with a suggestion for implementing such a verification system.

I still believe we have data with bad quality in many places (in Sweden). I have to fix stuff often when I visit new places apart from all the stuff we are missing. We are basically trying to keep up with an ever changing surrounding without a good way to indicate our data quality.

We can do better, but we need a new system that make it easy for contributors to verify our precious data (see my previous email).

Cheers
pangoSE


_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

Alan Mackie
In reply to this post by pangose


On Sat, 22 Aug 2020, 09:28 pangoSE, <[hidden email]> wrote:
Hi 😀

Mateusz Konieczny <[hidden email]> skrev: (22 augusti 2020 09:55:10 CEST)
>"It a playground with half-ass quality more than an authoritative and
>verified source of information (like e.g. Wikipedia)"
>
>I am not sure whatever you claim that
>Wikipedia is
>"playground with half-ass quality" or
>"authoritative and verified source of information".

I meant that a verification system does exist in Wikipedia and they now require references on all statements to keep up the quality of the articles which is sane IMO. We have no such system.
We have a method of referencing sources on the changeset. Although I do think it lacks granularity sometimes.

I am not keen on repeated references to "authoritative and verified sources of information" as this starts to walk back the primary principle of on the ground verifiability that OSM relies on.

Additional supporting evidence via Mapillary, OpenStreetCam etc. should always be welcome, but in OSM the highest grade of source is "=survey" and I think that is as it should be. OSM has a scope that goes beyond areas with well funded GIS Departments and friendly licenses.

>OSM would benefit from better verification
>tools and so on but insult-laden post
>filed with misunderstandings will not
>lead towards them.

Sorry if it came across as harsh, I get your point and will try to moderate my criticism a little more.

I love OSM and have contributed a lot over the years and recommend it to everyone I meet who uses maps.

I just sent a follow up email with a suggestion for implementing such a verification system.

I still believe we have data with bad quality in many places (in Sweden). I have to fix stuff often when I visit new places apart from all the stuff we are missing. We are basically trying to keep up with an ever changing surrounding without a good way to indicate our data quality.

Data going out if date is just the nature of the beast. It's a perennial problem with official maps too.


We can do better, but we need a new system that make it easy for contributors to verify our precious data (see my previous email).

I think StreetComplete is currently working on a project that helps with this.


Cheers
pangoSE

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Let's think about how we use mailing lists (was: Re: Call for verification (Was: Re: VANDALISM !))

Andy Townsend
In reply to this post by pangose
On 22/08/2020 09:12, pangoSE wrote:
> Here is yet another example of bad data in our database:
>
(to "pangoSE", via the list):

Sometimes it helps to think about how we react to things in the spur of
the moment.  Martijn's anecdote was about something that he mapped 9
years ago that was _correct at the time_ .

People occasionally claim that open mailing lists are "toxic" or
similar, because they allow people to say things quickly without
thinking, and anything posted can't later be deleted because it has been
emailed to everyone.  Posts like the above don't help people who might
be reserved about posting here from doing so.

Some of your recent posts (e.g.
https://lists.openstreetmap.org/pipermail/talk/2020-August/085281.html )
look indistinguishable from those that a troll might make to me.  You
probably don't think of yourself as a troll - you just thing that you
are asking "serious and pertinent questions". You're right that there's
a discussion to be had about exactly how OSM and wikipedia et al are
linked, but taking unrelated comments out of context doesn't help your
argument - it'll just make it more likely that people will file all of
your posts in the "round filing cabinet in the corner of the room"
rather than read them.

Best Regards,

Andy (writing in a personal capacity, and not - thankfully - any sort of
list moderator here)



_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

pangose
In reply to this post by General Discussion mailing list
Hi 😀

Mateusz Konieczny <[hidden email]> skrev: (22 augusti 2020 10:51:49 CEST)
>(1) Wikipedia may strongly encourage or mandate it in theory, but there
>are
>still edits being made without any citations

Yeah I know, but the point is its really hard to create a new article in WP without references without it being flagged for deletion. So by "threatening" with deletion they raise the bar for inclusion and hence hopefully raise the quality too. We have no system to flag for deletion, nor to verify an object.

>
>(2) Wikipedia is explicitly forbidding original research, OSM is
>explicitly encouraging it
>The best edits are where people map things not mapped anywhere else,
>or at least not mapped in any other open data source.

Is this relevant to the discussion?  I proposed a button that makes it easy for a user to state
1) I attest this is correct (no proof or anything required)

NOTE: a malicious user could of course mark all objects in the database as verified, so we probably need a way to handle vandalism, but my implementation is a first draft so feel free to suggest improvements. 😀

>
>It makes impossible to require citations for everything and requiring
>people
>to contribute to Mapillary or equivalent would be an unreasonable
>burden.

I'm not suggesting requiring that, but we should motivate the user to reference a source and make it dead simple to do so. But this is off topic for this thread IMO and we probably need a new system for that too because our current changeset references does not add much value IMO.

>
>(3) Yes, better verification tools would be likely better.

So what do you think about the proposed system?

>
>(4) Have you seen
>https://wiki.openstreetmap.org/wiki/Microgrants/Microgrants_2020/Proposal/Map_Maintenance_with_StreetComplete
>(BTW, I really need to finish my resurvey opening hours quest for
>StreetComplete).

No. Thanks for the link 👍

Cheers
pangoSE

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

Andrew Harvey-3
In reply to this post by pangose


On Sat, 22 Aug 2020 at 18:28, pangoSE <[hidden email]> wrote:
Hi 😀

Mateusz Konieczny <[hidden email]> skrev: (22 augusti 2020 09:55:10 CEST)
>"It a playground with half-ass quality more than an authoritative and
>verified source of information (like e.g. Wikipedia)"
>
>I am not sure whatever you claim that
>Wikipedia is
>"playground with half-ass quality" or
>"authoritative and verified source of information".

I meant that a verification system does exist in Wikipedia and they now require references on all statements to keep up the quality of the articles which is sane IMO. We have no such system.

I do think OSM is slowly moving to such a system, at least in areas that have an active community and are well mapped. I try to collect Mapillary imagery and one of the reasons is it provides a reference for my change, other's who do the same make the data quality a bit higher because others can verify remotely from the imagery.

If I see a suspicious change, I'll post a changeset comment asking if they are sure and if it changed recently especially when I have visited there recently and seems unlikely it would have changed since.

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Let's think about how we use mailing lists (was: Re: Call for verification (Was: Re: VANDALISM !))

pangose
In reply to this post by Andy Townsend
Thanks for the heads up. Words are really powerful so its wise to weigh them carefully before disclosure.

I have no intentions of highlighting the errors of others just for the sake of it. Individual errors are not interesting, but they can give input to a healthy discussion.

I have ideas for improving OSM and raise the quality. I can see that it helps to be friendly and diplomatic so I'll try being that.

I will try moderate my criticism and sleep on it when I have something to suggest instead of posting right away.

I'm really passionate sometimes and I guess it spills over into my emails.

No harm intended.

Have a nice day.

Cheers
pangoSE

Andy Townsend <[hidden email]> skrev: (22 augusti 2020 11:23:11 CEST)
On 22/08/2020 09:12, pangoSE wrote:
Here is yet another example of bad data in our database:

(to "pangoSE", via the list):

Sometimes it helps to think about how we react to things in the spur of
the moment.  Martijn's anecdote was about something that he mapped 9
years ago that was _correct at the time_ .

People occasionally claim that open mailing lists are "toxic" or
similar, because they allow people to say things quickly without
thinking, and anything posted can't later be deleted because it has been
emailed to everyone.  Posts like the above don't help people who might
be reserved about posting here from doing so.

Some of your recent posts (e.g.
https://lists.openstreetmap.org/pipermail/talk/2020-August/085281.html )
look indistinguishable from those that a troll might make to me.  You
probably don't think of yourself as a troll - you just thing that you
are asking "serious and pertinent questions". You're right that there's
a discussion to be had about exactly how OSM and wikipedia et al are
linked, but taking unrelated comments out of context doesn't help your
argument - it'll just make it more likely that people will file all of
your posts in the "round filing cabinet in the corner of the room"
rather than read them.

Best Regards,

Andy (writing in a personal capacity, and not - thankfully - any sort of
list moderator here)
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

Jo-2
In reply to this post by pangose


On Sat, Aug 22, 2020, 11:30 pangoSE <[hidden email]> wrote:
Hi 😀

Mateusz Konieczny <[hidden email]> skrev: (22 augusti 2020 10:51:49 CEST)
>(1) Wikipedia may strongly encourage or mandate it in theory, but there
>are
>still edits being made without any citations

Yeah I know, but the point is its really hard to create a new article in WP without references without it being flagged for deletion. So by "threatening" with deletion they raise the bar for inclusion and hence hopefully raise the quality too. We have no system to flag for deletion, nor to verify an object.

I find this highly annoying on Wikipedia and it is the reason I don't contribute there anymore.

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

General Discussion mailing list
In reply to this post by pangose



Aug 22, 2020, 11:28 by [hidden email]:
Hi 😀

Mateusz Konieczny <[hidden email]> skrev: (22 augusti 2020 10:51:49 CEST)
>(1) Wikipedia may strongly encourage or mandate it in theory, but there
>are
>still edits being made without any citations

Yeah I know, but the point is its really hard to create a new article in WP without references without it being flagged for deletion. So by "threatening" with deletion they raise the bar for inclusion and hence hopefully raise the quality too. We have no system to flag for deletion, nor to verify an object.

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

pangose
In reply to this post by Jo-2
Hi

Jo <[hidden email]> skrev: (22 augusti 2020 11:44:49 CEST)

>On Sat, Aug 22, 2020, 11:30 pangoSE <[hidden email]> wrote:
>
>> Hi 😀
>>
>> Mateusz Konieczny <[hidden email]> skrev: (22 augusti 2020
>> 10:51:49 CEST)
>> >(1) Wikipedia may strongly encourage or mandate it in theory, but
>there
>> >are
>> >still edits being made without any citations
>>
>> Yeah I know, but the point is its really hard to create a new article
>in
>> WP without references without it being flagged for deletion. So by
>> "threatening" with deletion they raise the bar for inclusion and
>hence
>> hopefully raise the quality too. We have no system to flag for
>deletion,
>> nor to verify an object.
>>
>
>I find this highly annoying on Wikipedia and it is the reason I don't
>contribute there anymore.

Interesting. I guess you are not the only because  https://en.m.wikipedia.org/wiki/Deletionpedia exist.
I don't propose we annoy our users the same way, because the downside is fewer editors.

I guess its a choice on an continuum between general low quality edits and many editors and generally higher quality edits and fewer editors.

Right now OSM accepts almost any crap edit you can throw at it with a big thank you and we have no really good way of measuring the quality of what remains after our sometimes spotty QA.

I would like to help change that by providing better tools for verification and follow up of things you added/edited in the past.

I would very much love a telegram bot flagging a new user making an edit to an object I help curate, but no such tool exist to my knowledge today.

WDYT? Would such a tool be nice to have?

Cheers
pangoSE

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

Alan Mackie


On Sat, 22 Aug 2020 at 11:02, pangoSE <[hidden email]> wrote:
Hi

Jo <[hidden email]> skrev: (22 augusti 2020 11:44:49 CEST)
>On Sat, Aug 22, 2020, 11:30 pangoSE <[hidden email]> wrote:
>
>> Hi 😀
>>
>> Mateusz Konieczny <[hidden email]> skrev: (22 augusti 2020
>> 10:51:49 CEST)
>> >(1) Wikipedia may strongly encourage or mandate it in theory, but
>there
>> >are
>> >still edits being made without any citations
>>
>> Yeah I know, but the point is its really hard to create a new article
>in
>> WP without references without it being flagged for deletion. So by
>> "threatening" with deletion they raise the bar for inclusion and
>hence
>> hopefully raise the quality too. We have no system to flag for
>deletion,
>> nor to verify an object.
>>
>
>I find this highly annoying on Wikipedia and it is the reason I don't
>contribute there anymore.

Interesting. I guess you are not the only because  https://en.m.wikipedia.org/wiki/Deletionpedia exist.
I don't propose we annoy our users the same way, because the downside is fewer editors.

I guess its a choice on an continuum between general low quality edits and many editors and generally higher quality edits and fewer editors.

Right now OSM accepts almost any crap edit you can throw at it with a big thank you and we have no really good way of measuring the quality of what remains after our sometimes spotty QA.

I would like to help change that by providing better tools for verification and follow up of things you added/edited in the past.

I would very much love a telegram bot flagging a new user making an edit to an object I help curate, but no such tool exist to my knowledge today.
OSMCha tags new users and offers RSS feeds for saved filters. I'm not aware of  a way to do this for "ways once touched by [username]" though.

WDYT? Would such a tool be nice to have?

Cheers
pangoSE

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

dieterdreist
In reply to this post by pangose


sent from a phone

> On 22. Aug 2020, at 10:15, pangoSE <[hidden email]> wrote:
>
> Here is yet another example of bad data in our database:


fix it ;-)

Of course OpenStreetMap contains errors, just like any other source, and probably more, given that most contributors are laymen and have very few experience (few total edits, often just 1).

On the other hand, we may be very fast when something changes, very flexible in emergencies (think Haiti), and have interesting niche data that commercial and public data providers don’t care for.

It all depends on the local community in the end. If you have reached a critical mass to have locals everywhere, it will work great and bugs will wash out. Otherwise the data might get stale just like any other data. Also using the data is essential to find the problems, for example the 212 story garage is likely fixed now ;-)

I tend to agree with Steve A.

Cheers Martin
_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

Shawn K. Quinn
In reply to this post by General Discussion mailing list
On 8/22/20 03:20, Mateusz Konieczny via talk wrote:
> Nobody claims OpenStreetMap data contains no mistakes.

There are a lot of cases where OSM data is better than that in Google
Maps, Mapquest, Bing Maps, etc. Unfortunately there are also a lot of
cases where the converse is true; in particular, we have almost no
addressing data save for the few places where dedicated mappers have
added it via exhausting on-foot surveys (not to be confused with
exhaust*ive* surveys, speaking from experience here) or gotten lucky
enough to score a compatible import.

To its credit, Vespucci at least tells mappers "object may be out of
date" when it has sat unedited for over a year. I have missed
out-of-date data sitting right under my nose, the best example of this
being the Whitehall Hotel in downtown Houston (finally noticed and fixed
a while back).

> https://wiki.openstreetmap.org/wiki/Notes are designed as tool allowing to
> describe incorrect data that someone is unable or unwilling to fix (and
> yes, we have thousands of reports of mistakes)

I have also used notes (and seen the notes feature used by others) to
quickly note business information that I can't add in Vespucci or
another app right then and there. Yes, I do close a lot of my own notes,
and I suspect I'm not the only one.

--
Shawn K. Quinn <[hidden email]>
http://www.rantroulette.com
http://www.skqrecordquest.com

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

Shawn K. Quinn
In reply to this post by pangose
On 8/22/20 03:26, pangoSE wrote:
> I meant that a verification system does exist in Wikipedia and they
> now require references on all statements to keep up the quality of
> the articles which is sane IMO. We have no such system.

The big, huge difference between Wikipedia and OSM is that Wikipedia
does not allow original research at all, whereas OSM thrives on the
original research of everyone who contributes and in fact it is the
stuff that comes from third parties that has to be vetted more closely
for license compliance and copyright issues.

I agree we could do better in the quality control department but a lot
of things added to OSM will be added there first before any third
parties pick them up. That makes references a bit problematic, IMO.

--
Shawn K. Quinn <[hidden email]>
http://www.rantroulette.com
http://www.skqrecordquest.com

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

pangose
Hi Shawn

"Shawn K. Quinn" <[hidden email]> skrev: (23 augusti 2020 00:31:28 CEST)

>On 8/22/20 03:26, pangoSE wrote:
>> I meant that a verification system does exist in Wikipedia and they
>> now require references on all statements to keep up the quality of
>> the articles which is sane IMO. We have no such system.
>
>The big, huge difference between Wikipedia and OSM is that Wikipedia
>does not allow original research at all, whereas OSM thrives on the
>original research of everyone who contributes and in fact it is the
>stuff that comes from third parties that has to be vetted more closely
>for license compliance and copyright issues.
>
>I agree we could do better in the quality control department but a lot
>of things added to OSM will be added there first before any third
>parties pick them up. That makes references a bit problematic, IMO.

All edits in OSM must be verifyable on the ground if I understood this correctly: https://wiki.openstreetmap.org/wiki/Verifiability

Problem is to really make this easy to review without visiting the same spot we would in many cases need a good photo or perhaps multiple photos from different angles.
Unfortunately we neither encourage nor support image uploading anywhere hosted by ourselves or others (we could probably easily integrate mapillary uploading in the website and in our mobile tools. I take photos with osmtracker sometimes but cannot upload them to mapillary from inside JOSM). I'm not saying it should be a demand, but I think we would gain a lot in many changeset discussions if adding images to the chat and changesets is made possible or if images in mapillary in the area were visible and referencable on the changeset discussion page.

Alternatively we could cook our own image storage service if we want. We got the money for it now and commercial persistent object storage solutions are available from multiple providers releasing the burdon of infrastructure maintenance on our operations working group. WDYT?

This and my proposal to mark features as verified at this point in time could potentially make it much easier to judge the overall quality of our data and map.

We would still be lacking a REAL granular referencing system where every statement (tag) is references individually with a date, author and optionally a photo. That would be really awesome, but it would require additions to the main database model and ruby website to support (this is perhaps a perfect GSoC project). Being able to browse to a specific tag on an object and discuss that would be a crucial addition to the website because now we are forced to comment on the changeset (or sending pms) and I think its really cumbersome to manually reference which one of the sometimes hundreds of objects I'm talking about.

Andy Allen (he runs  http://www.thunderforest.com/ which has a nice vector map service by the way on a free limited tier) a former member of the operations working group and current co-maintainer of the rails website posted this a year ago:
https://gravitystorm.github.io/osmf-infra-plans/ and this july the OSMF and the operations working group announced hiring of a Senior Site Reliability Engineer: https://mobile.twitter.com/OSM_Tech/status/1287395222847139846

This seems like a good move. We would benefit a lot from being able to easily load balance and adjust VMs on our own or someone elses openstack infrastructure where we can easily provision new servers for development or testing when needed instead of having dedicated physical hardware servers that causes availability issues if they break because of single point of failures.

See also https://operations.osmfoundation.org/ 

BTW osm-fr already made this move and is mostly running VMs now and has moved some of their VMs (heavy tile rendering) into the OVH cloud to manage their hardware more efficiently. See https://wiki.openstreetmap.org/wiki/FR:Serveurs_OpenStreetMap_France

Cheers
PangoSE

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

pangose

--
Skickat från min Android-enhet med k9.
_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Call for verification (Was: Re: VANDALISM !)

pangose
In reply to this post by dieterdreist
Hi Martin

Den Sat, 22 Aug 2020 19:30:23 +0200 Martin
skrev Re: [OSM-talk] Call for verification (Was: Re:  VANDALISM !):

> sent from a phone
>
> > On 22. Aug 2020, at 10:15, pangoSE <[hidden email]> wrote:
> >
> > Here is yet another example of bad data in our database:  
>
>
> fix it ;-)

Yeah! But to fix "it" (it being the overall low or unknown quality of
the map) we need good tools that encourage reviewing and fixing.

We
have a discoverability and usability problem IMO in this area. I fixed
loads of errors during my time and I like it, but I have a poor grasp
of the overall quality of the map in the area and OSM is not making it
easy for me to find the less good quality spots with
stale/old/non-reviewed data.

>
> Of course OpenStreetMap contains errors, just like any other source,
> and probably more, given that most contributors are laymen and have
> very few experience (few total edits, often just 1).
>
> On the other hand, we may be very fast when something changes, very
> flexible in emergencies (think Haiti), and have interesting niche
> data that commercial and public data providers don’t care for.

Yes, that really nice. I would like to find a middle ground between fast
and poor/unknown and slow and high degree of verification.

>
> It all depends on the local community in the end. If you have reached
> a critical mass to have locals everywhere, it will work great and
> bugs will wash out. Otherwise the data might get stale just like any
> other data. Also using the data is essential to find the problems,
> for example the 212 story garage is likely fixed now ;-)

Yeah, I agree. Lets make it easy for a local community to keep the map
verified and up to date. We could e.g. set a verification-needed
flag on objects edited in a changeset with "please review". That would make it easy create an
overview of all things todo in your local area based on the objects -
not the changesets that touched them (they can easily be found in
todays interface from the object).

/pangoSE

_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
12