Re: [OSM-talk] Untagged Nodes and Ways

[Craig Wallace]

On 2017-01-31 18:02, Andreas Vilén wrote:

I always have all checks activated so it seems it just misses stuff
sometimes... Could the issue be that I never reinstall Josm when I
update but just overwrite the josm-tested.jar file with the new one?


Sounds like you have told JOSM to ignore an error from the validator. 
Then it seems it will remember that, and ignores all errors of that type.
This is stored in your user preferences, so isn't affected if you update 
the jar file.


Not sure if there's an easy way to reset the ignore list. You could find 
it in your user profile, then delete it. In Windows, probably in your 
AppData folder, then JOSM\validator


Or in JOSM you can go to Preferences, then Advanced, then click More, 
then reset preferences. Though that will reset any other options you 
have set it in JOSM.



___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[Clifford Snow]

On Tue, Jan 31, 2017 at 10:02 AM, Andreas Vilén 
wrote:

> I always have all checks activated so it seems it just misses stuff
> sometimes... Could the issue be that I never reinstall Josm when I update
> but just overwrite the josm-tested.jar file with the new one?


That's what I do. The actual command for running JOSM is something like
java -jar josm-test.jar so overwriting your jar file would be an issue so
you should be okay.


-- 
@osm_seattle
osm_seattle.snowandsnow.us
OpenStreetMap: Maps with a human touch
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[Andreas Vilén]

I always have all checks activated so it seems it just misses stuff
sometimes... Could the issue be that I never reinstall Josm when I update
but just overwrite the josm-tested.jar file with the new one?

/Andreas

On Mon, Jan 30, 2017 at 6:59 PM, john whelan  wrote:

> I seem to recall that JOSM allows you to pick and choose what to validate
> both for what is in JOSM and what should be validated on uploading.
>
> It's quite useful in that you can just validate for duplicate ways or
> select highway=whatever then check for crossing ways.
>
> If you're mapping in Africa the highway wiki recommends tagging path
> rather than footway.  Apparently the locals use small motorcycles on them.
>
> Cheerio John
>
> On 30 Jan 2017 12:29 pm, "Andreas Vilén"  wrote:
>
>> The Josm validator doesn't always work. I have noticed untagged ways
>> uploaded by myself long after I created them, and I wouldn't miss a message
>> like that.
>>
>> Often it's footways where I've drawn a lot, tagged them in bulk and
>> missed a few.
>>
>> /Andreas
>>
>> Skickat från min iPhone
>>
>> 30 jan. 2017 kl. 16:23 skrev Clifford Snow :
>>
>>
>> On Mon, Jan 30, 2017 at 2:16 AM, Rory McCann 
>> wrote:
>>
>>> Well, to be pedantically correct*, many untagged ways are members of
>>> relations which have tags.
>>>
>>
>> untagged ways as member of a relation were not included in the count.
>> There are over 7M untagged ways that are part of a relation.
>>
>> My guess is that people started to add in a feature but somehow forgot to
>> tag them before uploading. JOSM gives a nice warning, which I've received.
>> iD gives a warning as well but not as a new dialog box that you have to say
>> yes go ahead and upload.
>>
>> Clifford
>>
>>
>> --
>> @osm_seattle
>> osm_seattle.snowandsnow.us
>> OpenStreetMap: Maps with a human touch
>>
>> ___
>> talk mailing list
>> talk@openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/talk
>>
>>
>> ___
>> talk mailing list
>> talk@openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/talk
>>
>>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[john whelan]

I seem to recall that JOSM allows you to pick and choose what to validate
both for what is in JOSM and what should be validated on uploading.

It's quite useful in that you can just validate for duplicate ways or
select highway=whatever then check for crossing ways.

If you're mapping in Africa the highway wiki recommends tagging path rather
than footway.  Apparently the locals use small motorcycles on them.

Cheerio John

On 30 Jan 2017 12:29 pm, "Andreas Vilén"  wrote:

> The Josm validator doesn't always work. I have noticed untagged ways
> uploaded by myself long after I created them, and I wouldn't miss a message
> like that.
>
> Often it's footways where I've drawn a lot, tagged them in bulk and missed
> a few.
>
> /Andreas
>
> Skickat från min iPhone
>
> 30 jan. 2017 kl. 16:23 skrev Clifford Snow :
>
>
> On Mon, Jan 30, 2017 at 2:16 AM, Rory McCann  wrote:
>
>> Well, to be pedantically correct*, many untagged ways are members of
>> relations which have tags.
>>
>
> untagged ways as member of a relation were not included in the count.
> There are over 7M untagged ways that are part of a relation.
>
> My guess is that people started to add in a feature but somehow forgot to
> tag them before uploading. JOSM gives a nice warning, which I've received.
> iD gives a warning as well but not as a new dialog box that you have to say
> yes go ahead and upload.
>
> Clifford
>
>
> --
> @osm_seattle
> osm_seattle.snowandsnow.us
> OpenStreetMap: Maps with a human touch
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[Andreas Vilén]

The Josm validator doesn't always work. I have noticed untagged ways uploaded 
by myself long after I created them, and I wouldn't miss a message like that.

Often it's footways where I've drawn a lot, tagged them in bulk and missed a 
few.

/Andreas

Skickat från min iPhone

> 30 jan. 2017 kl. 16:23 skrev Clifford Snow :
> 
> 
>> On Mon, Jan 30, 2017 at 2:16 AM, Rory McCann  wrote:
>> Well, to be pedantically correct*, many untagged ways are members of
>> relations which have tags.
> 
> untagged ways as member of a relation were not included in the count. There 
> are over 7M untagged ways that are part of a relation.
> 
> My guess is that people started to add in a feature but somehow forgot to tag 
> them before uploading. JOSM gives a nice warning, which I've received. iD 
> gives a warning as well but not as a new dialog box that you have to say yes 
> go ahead and upload.
> 
> Clifford
> 
> 
> -- 
> @osm_seattle
> osm_seattle.snowandsnow.us
> OpenStreetMap: Maps with a human touch
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[Clifford Snow]

On Mon, Jan 30, 2017 at 2:16 AM, Rory McCann  wrote:

> Well, to be pedantically correct*, many untagged ways are members of
> relations which have tags.
>

untagged ways as member of a relation were not included in the count. There
are over 7M untagged ways that are part of a relation.

My guess is that people started to add in a feature but somehow forgot to
tag them before uploading. JOSM gives a nice warning, which I've received.
iD gives a warning as well but not as a new dialog box that you have to say
yes go ahead and upload.

Clifford


-- 
@osm_seattle
osm_seattle.snowandsnow.us
OpenStreetMap: Maps with a human touch
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[Martin Koppenhoefer]

2017-01-30 14:51 GMT+01:00 john whelan :

> >
> I think a hint like JOSM presents you when you try to upload an untagged
> way or node (that aren't part of a relation either) should be sufficient. I
> wouldn't reject those ways by the api. There might also be some sense in
> this, e.g. when working on big things a mapper might want to upload part of
> his work before it is finished, to continue work on it later.
>
> Well yes but how long do you wait?
>


some days, max some weeks. If you can make some sense by adding a pertinent
tag (in an area you know) you should do it right away of course, but if you
find some long untagged way which seems to have been a lot of work to add
it, I'd rather be conservative before deleting it, e.g. contact the
original author or add some tags to it.

Cheers,
Martin
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[john whelan]

>
I think a hint like JOSM presents you when you try to upload an untagged
way or node (that aren't part of a relation either) should be sufficient. I
wouldn't reject those ways by the api. There might also be some sense in
this, e.g. when working on big things a mapper might want to upload part of
his work before it is finished, to continue work on it later.

Well yes but how long do you wait?

Cheerio John

On 30 January 2017 at 07:19, Martin Koppenhoefer 
wrote:

>
> 2017-01-30 11:16 GMT+01:00 Rory McCann :
>
>> > I would be interested in hearing from you why we should ever allow
>> > untagged ways to be uploaded.
>>
>> Well, to be pedantically correct*, many untagged ways are members of
>> relations which have tags.
>
>
>
> I think a hint like JOSM presents you when you try to upload an untagged
> way or node (that aren't part of a relation either) should be sufficient. I
> wouldn't reject those ways by the api. There might also be some sense in
> this, e.g. when working on big things a mapper might want to upload part of
> his work before it is finished, to continue work on it later.
>
> Cheers,
> Martin
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[Martin Koppenhoefer]

2017-01-30 11:16 GMT+01:00 Rory McCann :

> > I would be interested in hearing from you why we should ever allow
> > untagged ways to be uploaded.
>
> Well, to be pedantically correct*, many untagged ways are members of
> relations which have tags.



I think a hint like JOSM presents you when you try to upload an untagged
way or node (that aren't part of a relation either) should be sufficient. I
wouldn't reject those ways by the api. There might also be some sense in
this, e.g. when working on big things a mapper might want to upload part of
his work before it is finished, to continue work on it later.

Cheers,
Martin
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[Rory McCann]

On 28/01/17 23:55, Clifford Snow wrote:
> I would be interested in hearing from you why we should ever allow
> untagged ways to be uploaded. 

Well, to be pedantically correct*, many untagged ways are members of
relations which have tags.

* The best kind of correct



signature.asc
Description: OpenPGP digital signature
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[john whelan]

> It's true untagged ways don't render, trivial amount of disk space, and
they don't route. What they do is create work to fixed or remove them. At
one time in life I was a quality person with a goal of creating systems to
do it right the first time. My sense is that it isn't a new user problem,
but a systems problem. I've spend time now and them cleaning up stuff using
Frederik's OSMI tool. I'd really rather spend my time adding and improving
the map than fixing errors.

True enough but I've added a lot more buildings and highways in Africa for
a lot less effort by just adding tags to untagged ways.  Load up a country
dump into JOSM maybe cut it down first and just let JOSM validation and the
todo list plugin do their thing.

Cheerio John

On 28 January 2017 at 19:32, Clifford Snow  wrote:

>
> On Sat, Jan 28, 2017 at 3:08 PM, Andy Townsend  wrote:
>
>>
>> Also, let's be honest - untagged ways aren't a "problem" in any other
>> sense other than disk space.  They won't confuse renderers or routers.  If
>> there are too many in a given area they might confuse other mappers, but
>> these other mappers are perfectly free to delete them if they can't figure
>> out what they were supposed to be.
>
>
> It's true untagged ways don't render, trivial amount of disk space, and
> they don't route. What they do is create work to fixed or remove them. At
> one time in life I was a quality person with a goal of creating systems to
> do it right the first time. My sense is that it isn't a new user problem,
> but a systems problem. I've spend time now and them cleaning up stuff using
> Frederik's OSMI tool. I'd really rather spend my time adding and improving
> the map than fixing errors.
>
>
> --
> @osm_seattle
> osm_seattle.snowandsnow.us
> OpenStreetMap: Maps with a human touch
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[Clifford Snow]

On Sat, Jan 28, 2017 at 3:08 PM, Andy Townsend  wrote:

>
> Also, let's be honest - untagged ways aren't a "problem" in any other
> sense other than disk space.  They won't confuse renderers or routers.  If
> there are too many in a given area they might confuse other mappers, but
> these other mappers are perfectly free to delete them if they can't figure
> out what they were supposed to be.


It's true untagged ways don't render, trivial amount of disk space, and
they don't route. What they do is create work to fixed or remove them. At
one time in life I was a quality person with a goal of creating systems to
do it right the first time. My sense is that it isn't a new user problem,
but a systems problem. I've spend time now and them cleaning up stuff using
Frederik's OSMI tool. I'd really rather spend my time adding and improving
the map than fixing errors.


-- 
@osm_seattle
osm_seattle.snowandsnow.us
OpenStreetMap: Maps with a human touch
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[john whelan]

The interesting thing is looking at Bing there seems to be no difference
between the areas marked off and the surrounding area.  I'm not sure they
are serving any useful purpose as we get updated imagery.

Thanks for the input.

Cheerio John

On 28 January 2017 at 19:28, Andy Townsend  wrote:

> On 29/01/2017 00:17, john whelan wrote:
>
> I'm wandering through Africa at the moment basically tagging untagged ways
> and area=yes.  I think I've corrected several thousand but something that
> I've just hit are boxes tagged hires=yes or no.
>
>
> Those I think are referring to where there was Bing high resolution
> imagery and where there wasn't.  I've never been convinced that things like
> that belong in OSM in the first place (though others may disagree) - you'll
> probably be able to find some previous list discussion if you go back far
> enough.
>
> Cheers,
>
> Andy
>
>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[Andy Townsend]

On 29/01/2017 00:17, john whelan wrote:
I'm wandering through Africa at the moment basically tagging untagged 
ways and area=yes.  I think I've corrected several thousand but 
something that I've just hit are boxes tagged hires=yes or no.


Those I think are referring to where there was Bing high resolution 
imagery and where there wasn't.  I've never been convinced that things 
like that belong in OSM in the first place (though others may disagree) 
- you'll probably be able to find some previous list discussion if you 
go back far enough.


Cheers,

Andy

___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[john whelan]

I'm wandering through Africa at the moment basically tagging untagged ways
and area=yes.  I think I've corrected several thousand but something that
I've just hit are boxes tagged hires=yes or no.

Most are more than a year old.  These I'm unsure what to do with them if
anything should they simply be deleted if more than a year old?

Thoughts?

Thanks John

On 28 January 2017 at 18:08, Andy Townsend  wrote:

> On 28/01/2017 22:55, Clifford Snow wrote:
>
>>
>> iD issue 3806 [2] requested that untagged ways would prevent the user
>> from uploading a changeset. Bryan thought it is fine for user to add an
>> untagged way for someone else to fix. I see his point that they may get
>> discouraged and not continue to contribute to OSM.
>>
>
> Turning that around, can you suggest a way of communicating the fact that
> this is a problem to the user using terms that they'd understand?  I'd find
> it difficult to do that.
>
> I fairly regularly talk to new mappers about untagged ways and "area=yes"
> ways (which are the more common example) and have the "if you want
> renderers and routers to know what something is, you need to do more than
> give it a name" conversation.  It'd be impossible to do that if they
> weren't allowed to upload in the first place.
>
> Also, let's be honest - untagged ways aren't a "problem" in any other
> sense other than disk space.  They won't confuse renderers or routers.  If
> there are too many in a given area they might confuse other mappers, but
> these other mappers are perfectly free to delete them if they can't figure
> out what they were supposed to be.
>
> Best Regards,
>
> Andy
>
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Untagged Nodes and Ways

[Andy Townsend]

On 28/01/2017 22:55, Clifford Snow wrote:


iD issue 3806 [2] requested that untagged ways would prevent the user 
from uploading a changeset. Bryan thought it is fine for user to add 
an untagged way for someone else to fix. I see his point that they may 
get discouraged and not continue to contribute to OSM.


Turning that around, can you suggest a way of communicating the fact 
that this is a problem to the user using terms that they'd understand?  
I'd find it difficult to do that.


I fairly regularly talk to new mappers about untagged ways and 
"area=yes" ways (which are the more common example) and have the "if you 
want renderers and routers to know what something is, you need to do 
more than give it a name" conversation.  It'd be impossible to do that 
if they weren't allowed to upload in the first place.


Also, let's be honest - untagged ways aren't a "problem" in any other 
sense other than disk space.  They won't confuse renderers or routers.  
If there are too many in a given area they might confuse other mappers, 
but these other mappers are perfectly free to delete them if they can't 
figure out what they were supposed to be.


Best Regards,

Andy


___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk