[OSM-talk] Software configuration | Re: iD influencing tagging

[Rory McCann]

On 07.04.19 14:43, John Whelan wrote:
Tagging is not always easy, but I do have concerns when iD is so 
commonly used but the recommended tags do not align with OpenStreetMap 
I'll say normals.


Specifically one of my concerns is a semi-detached house is not 
recognised in iD only the more general tag house.


In JOSM, people, or groups, can make their own tagging presets. AFAIK iD 
unfortunately doesn't have this feature. If it did, the iD version on 
openstreetmap.org could be configured to something special, people could 
have their personal tagging presets "saved" somehow, maybe one could 
"load other presets from a remote address" via a URL parameter, allowing 
one to "load a preset with a link". Then each iD user wouldn't be able 
to complain to the iD developers. But that's not possible now.


For now, the OSM wiki voting isn't binding, the people who make editors 
have a lot of control. I'm sure patches & feature requests would be 
welcome



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

Re: [OSM-talk] Software configuration | Re: iD influencing tagging

[Jóhannes Birgir Jensson]

iD seems to have this feature via MapRules: 
https://github.com/openstreetmap/iD/pull/5617 


10. apríl 2019 kl. 21:08, skrifaði "Rory McCann" :

> On 07.04.19 14:43, John Whelan wrote:
> 
>> Tagging is not always easy, but I do have concerns when iD is so
>> commonly used but the recommended tags do not align with OpenStreetMap
>> I'll say normals.
>> 
>> Specifically one of my concerns is a semi-detached house is not
>> recognised in iD only the more general tag house.
> 
> In JOSM, people, or groups, can make their own tagging presets. AFAIK iD
> unfortunately doesn't have this feature. If it did, the iD version on
> openstreetmap.org could be configured to something special, people could
> have their personal tagging presets "saved" somehow, maybe one could
> "load other presets from a remote address" via a URL parameter, allowing
> one to "load a preset with a link". Then each iD user wouldn't be able
> to complain to the iD developers. But that's not possible now.
> 
> For now, the OSM wiki voting isn't binding, the people who make editors
> have a lot of control. I'm sure patches & feature requests would be
> welcome
> 
> ___
> 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] Software configuration | Re: iD influencing tagging

[Yuri Astrakhan]

On Wed, Apr 10, 2019 at 5:04 PM Rory McCann  wrote:

> In JOSM, people, or groups, can make their own tagging presets. AFAIK iD
> unfortunately doesn't have this feature. If it did, the iD version on
> openstreetmap.org could be configured to something special, people could
> have their personal tagging presets "saved" somehow, maybe one could
> "load other presets from a remote address" via a URL parameter, allowing
> one to "load a preset with a link".


Rory, I am actually hopping data items could become a general preset
storage for all editors. Each preset would be stored as a data item, and
editors would use a script to convert preset into a JSON file, or
eventually use it directly.  This approach has a number of advantages:

* easy to edit by community, track changes, and fix/revert in case of an
error
* easy to translate - one click description editing for every possible
language
* easy to verify / validate / cross-check - there are many ways to query
presets and to run validations on them, so an invalid preset can be quickly
fixed/reverted
* part of wiki - presets can be viewed as part of the regular wiki pages,
e.g. on a Tag:... page
* easy to add pictures and icons - part of wiki, or from commons - just
another property
* easy to categorize/partition presets - just another property to add
preset into a category, or have a whole category tree.
* easy to extend schema - just add more properties to target a new editor
feature - other editors will simply ignore it.
* ties with all other keys / tags / relations / relation roles -- if a
preset requires a tag, it is not a string, it is actually a reference to
that tag, with its own properties.
* easy to build custom UI -- structured data allows developers to have
custom editors for those presets

Obviously this has to be done in a non-disruptive, gradual way, and having
good quality control. I'm still researching on the best way to achieve this.

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

Re: [OSM-talk] Software configuration | Re: iD influencing tagging

[Mateusz Konieczny]

11 Apr 2019, 01:14 by yuriastrak...@gmail.com:
> * easy to edit by community
>
I am dubious whatever "anybody canedit any preset stored as wikidata
items" will be considered as benefit
> , track changes, and fix/revert in case of an error
>
All of that is easier with current 
method of keeping them in
git repositories.___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Software configuration | Re: iD influencing tagging

[Yuri Astrakhan]

On Thu, Apr 11, 2019 at 2:10 AM Mateusz Konieczny 
wrote:

> * easy to edit by community
>
> I am dubious whatever "anybody can
> edit any preset stored as wikidata
> items" will be considered as benefit
>

One could also doubt that allowing direct OSM and Wikipedia edits by anyone
would be considered as a benefit... But it does, doesn't it?  Worst
case scenario: someone breaks a preset - with so many eyes on them (exposed
via wiki pages, used by all editors, monitored via numerous tools,
cross-checked by validation queries, etc etc etc), it will be fixed within
minutes.

But we are talking more distant future. The initial idea is to generate
JSON / XML files from data items. So someone edits a data item, a script
will create a Pull Request for preset files. Devs can validate them all
before merging -- you get all the benefits I listed, plus more thorough
validation.

, track changes, and fix/revert in case of an error
>
> All of that is easier with current
> method of keeping them in
> git repositories.
>

Except there are several of these repositories, right? And some are
actually stored as wiki files for JOSM, without an easy diffing? Plus
another place to do translations. Plus there is no way you can see images
as part of those JSONs or XMLs? And plus you have to be a developer to
understand JSON.  But yes, pure iD presets have a good tracking feature in
of itself using github. Just doesn't offer all the other benefits.
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Software configuration | Re: iD influencing tagging

[Mateusz Konieczny]

Apr 11, 2019, 9:17 AM by yuriastrak...@gmail.com:

>  Worst case scenario: someone breaks a preset - with so many eyes on them 
> (exposed via wiki pages, used by all editors, monitored via numerous tools, 
> cross-checked by validation queries, etc etc etc), it will be fixed within 
> minutes.
>
In case of presets I see problem not with vandalism but with divergent opinions 
where both sides
have good intentions.

Also, presets are closer to infrastructure like
https://en.wikipedia.org/w/index.php?title=Template:Citation_needed&action=edit 

that on attempt to edit gets it is a "cascade-protected page, therefore only 
administrators can edit it."
rather than article pages.
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Software configuration | Re: iD influencing tagging

[Mark Wagner]

On Thu, 11 Apr 2019 03:17:27 -0400
Yuri Astrakhan  wrote:

> On Thu, Apr 11, 2019 at 2:10 AM Mateusz Konieczny
>  wrote:
> 
> > * easy to edit by community
> >
> > I am dubious whatever "anybody can
> > edit any preset stored as wikidata
> > items" will be considered as benefit
> >  
> 
> One could also doubt that allowing direct OSM and Wikipedia edits by
> anyone would be considered as a benefit... But it does, doesn't it?
> Worst case scenario: someone breaks a preset - with so many eyes on
> them (exposed via wiki pages, used by all editors, monitored via
> numerous tools, cross-checked by validation queries, etc etc etc), it
> will be fixed within minutes.

Wikipedia is a good comparison, but not in the way you intended it.

Wikipedia has special permissions for changing widely-used
templates or the website interface itself: the "template editor" and
"interface administrator" permissions.  An ordinary editor can only
mess up one article at a time, while a template editor can mess up a
half-million articles in a single go, and an interface administrator
can vandalize every page on Wikipedia at once.

If someone breaks the preset for something like "building=yes", then
sure, it'll be spotted and fixed in a matter of minutes.  But in the
meantime, there'll be hundreds of mis-tagged buildings, many of them in
places that nobody will review for years.

-- 
Mark

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

Re: [OSM-talk] Software configuration | Re: iD influencing tagging

[Tigerfell]

I am pretty sure you can protect data items.

The cited template is transcluded into more than 380,000 pages. The most often 
transcluded template in the OSM wiki is used on about 78,000 pages. I do not 
think that a full protection was considered necessary, because we wanted to 
keep the wiki open and I can not recall an incident yet.

Of course we can discuss about locking everything down that could cause 
problems, but that would oppose the idea of an openly editable OpenStreetMap 
and it would be hard to explain that the docs are protected while everyone can 
just delete some area in the map.

I think the general move of Yuri is right (generalising and opening the machine 
readable docs for the community) and I appreciate that. Doing so also prevents 
accusations against developers as discussed initially.

Tigerfell


Apr. 11, 2019, 11:30 a.m. by mark+...@carnildo.com:

> On Thu, 11 Apr 2019 03:17:27 -0400
> Yuri Astrakhan <> yuriastrak...@gmail.com > > 
> wrote:
>
>> On Thu, Apr 11, 2019 at 2:10 AM Mateusz Konieczny
>> <>> matkoni...@tutanota.com >> > wrote:
>>
>> > * easy to edit by community
>> >
>> > I am dubious whatever "anybody can
>> > edit any preset stored as wikidata
>> > items" will be considered as benefit
>> > 
>>
>> One could also doubt that allowing direct OSM and Wikipedia edits by
>> anyone would be considered as a benefit... But it does, doesn't it?
>> Worst case scenario: someone breaks a preset - with so many eyes on
>> them (exposed via wiki pages, used by all editors, monitored via
>> numerous tools, cross-checked by validation queries, etc etc etc), it
>> will be fixed within minutes.
>>
>
> Wikipedia is a good comparison, but not in the way you intended it.
>
> Wikipedia has special permissions for changing widely-used
> templates or the website interface itself: the "template editor" and
> "interface administrator" permissions.  An ordinary editor can only
> mess up one article at a time, while a template editor can mess up a
> half-million articles in a single go, and an interface administrator
> can vandalize every page on Wikipedia at once.
>
> If someone breaks the preset for something like "building=yes", then
> sure, it'll be spotted and fixed in a matter of minutes.  But in the
> meantime, there'll be hundreds of mis-tagged buildings, many of them in
> places that nobody will review for years.
>
> -- 
> Mark
>
> ___
> 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] Software configuration | Re: iD influencing tagging

[Yuri Astrakhan]

Thx Mark, makes sense. Data items are wiki pages too, and we can protect
them the same way. Still, I think it's a moot point -- at least initially
preset data will go via GitHub as regular pull requests.

On Thu, Apr 11, 2019, 05:34 Mark Wagner  wrote:

> On Thu, 11 Apr 2019 03:17:27 -0400
> Yuri Astrakhan  wrote:
>
> > On Thu, Apr 11, 2019 at 2:10 AM Mateusz Konieczny
> >  wrote:
> >
> > > * easy to edit by community
> > >
> > > I am dubious whatever "anybody can
> > > edit any preset stored as wikidata
> > > items" will be considered as benefit
> > >
> >
> > One could also doubt that allowing direct OSM and Wikipedia edits by
> > anyone would be considered as a benefit... But it does, doesn't it?
> > Worst case scenario: someone breaks a preset - with so many eyes on
> > them (exposed via wiki pages, used by all editors, monitored via
> > numerous tools, cross-checked by validation queries, etc etc etc), it
> > will be fixed within minutes.
>
> Wikipedia is a good comparison, but not in the way you intended it.
>
> Wikipedia has special permissions for changing widely-used
> templates or the website interface itself: the "template editor" and
> "interface administrator" permissions.  An ordinary editor can only
> mess up one article at a time, while a template editor can mess up a
> half-million articles in a single go, and an interface administrator
> can vandalize every page on Wikipedia at once.
>
> If someone breaks the preset for something like "building=yes", then
> sure, it'll be spotted and fixed in a matter of minutes.  But in the
> meantime, there'll be hundreds of mis-tagged buildings, many of them in
> places that nobody will review for years.
>
> --
> Mark
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk