Accepted or rejected?

classic Classic list List threaded Threaded
98 messages Options
12345
Reply | Threaded
Open this post in threaded view
|

Accepted or rejected?

Jan van Bekkum
The guideline to determine if a proposal is accepted is

A rule of thumb for "enough support" is 8 unanimous approval votes or 15 total votes with a majority approval, but other factors may also be considered (such as whether a feature is already in use).

This sounds a bit strange to me: a proposal with 8 approval votes and 1 decline would be rejected, while one with 8 approval votes and 7 declines would be accepted.

I suppose that this is what was intended:

"enough support" is 8 approval votes on a total of 14 votes or less and a majority approval otherwise.

Regards,

Jan

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

Re: Accepted or rejected?

Dan S
Hi,

No, I think it means what it says. Or at least, I think we have
treated it that way for a long while.

When there is very low interest (i.e. very few votes) - which is
pretty common - then even one dissenting vote is enough to make us
step back and think again, whereas if there are enough votes to make
"majority approval" a meaningful concept (I admit that 15 is a low
number for quorum) then we accept that there will always be some
disagreement, and so we use majority rather than unanimity.

This is how I interpret it. I'm not saying it's the best rule of thumb
out there. I'd say there's no point changing it in small ways - no-one
likes the tag voting system, and overhaul would be better than slight
tweaks.

Anyway, it is only a rule of thumb!

Best
Dan


2015-03-14 11:24 GMT+00:00 Jan van Bekkum <[hidden email]>:

> The guideline to determine if a proposal is accepted is
>
> A rule of thumb for "enough support" is 8 unanimous approval votes or 15
> total votes with a majority approval, but other factors may also be
> considered (such as whether a feature is already in use).
>
> This sounds a bit strange to me: a proposal with 8 approval votes and 1
> decline would be rejected, while one with 8 approval votes and 7 declines
> would be accepted.
>
> I suppose that this is what was intended:
>
> "enough support" is 8 approval votes on a total of 14 votes or less and a
> majority approval otherwise.
>
> Regards,
>
> Jan
>
> _______________________________________________
> Tagging mailing list
> [hidden email]
> https://lists.openstreetmap.org/listinfo/tagging
>

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

Re: Accepted or rejected?

Friedrich Volkmann
In reply to this post by Jan van Bekkum
On 14.03.2015 12:24, Jan van Bekkum wrote:

> The guideline to determine if a proposal is accepted is
>
> A rule of thumb for "enough support" is /8 unanimous approval votes/ or /15
> total votes with a majority approval/, but other factors may also be
> considered (such as whether a feature is already in use).
>
> This sounds a bit strange to me: a proposal with 8 approval votes and 1
> decline would be rejected, while one with 8 approval votes and 7 declines
> would be accepted.
>
> I suppose that this is what was intended:
>
> "enough support" is 8 approval votes on a total of 14 votes or less and a
> majority approval otherwise.

Yes, this should be reworded as you suggest. The current wording caused
confusion multiple times.

However, we should keep the mention of "other factors ... such as whether a
feature is already in use", especially when it comes to deprecation of
existing tags. I think that this should be even more clearly pointed out. A
majority of 8:7 votes cannot be sufficient for a deprecation of a tag used
by thousands of mappers.

--
Friedrich K. Volkmann       http://www.volki.at/
Adr.: Davidgasse 76-80/14/10, 1100 Wien, Austria

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

Re: Accepted or rejected?

Friedrich Volkmann
In reply to this post by Dan S
On 14.03.2015 12:50, Dan S wrote:
> When there is very low interest (i.e. very few votes) - which is
> pretty common - then even one dissenting vote is enough to make us
> step back and think again, whereas if there are enough votes to make
> "majority approval" a meaningful concept (I admit that 15 is a low
> number for quorum) then we accept that there will always be some
> disagreement, and so we use majority rather than unanimity.

As you are already indicating, 15 is too low a quorum in that case. We
cannot considering 8:7 votes an approval when we cosider 8:1 votes an
approval. That would mean that more negative votes would turn a rejection to
an approval, which is absurd.

--
Friedrich K. Volkmann       http://www.volki.at/
Adr.: Davidgasse 76-80/14/10, 1100 Wien, Austria

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

Re: Accepted or rejected?

brycenesbitt
In reply to this post by Jan van Bekkum
On Sat, Mar 14, 2015 at 4:24 AM, Jan van Bekkum <[hidden email]> wrote:
This sounds a bit strange to me: a proposal with 8 approval votes and 1 decline would be rejected, while one with 8 approval votes and 7 declines would be accepted.

Anything with that level of opposition (7 declines) is probably a flawed proposal.

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

Re: Accepted or rejected?

brycenesbitt
In reply to this post by Friedrich Volkmann
On Sat, Mar 14, 2015 at 5:47 AM, Friedrich Volkmann <[hidden email]> wrote:
As you are already indicating, 15 is too low a quorum in that case. We
cannot considering 8:7 votes an approval when we cosider 8:1 votes an
approval. That would mean that more negative votes would turn a rejection to
an approval, which is absurd.

Exactly that happened.  There was a proposal with 7 votes, some positive some negative.
3 more people voted no, flipping it to approval.


If the purpose of the wiki procedure is to find consensus, a bare 50% majority indicates
a near complete failure.

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

Re: Accepted or rejected?

Kotya Karapetyan
Proposal: let's change it to "8 unanimous approval votes or 10 or more votes with at least 74 % approval ones"?

I agree that the current situation looks funny pretty often.

On Sat, Mar 14, 2015 at 6:46 PM, Bryce Nesbitt <[hidden email]> wrote:
On Sat, Mar 14, 2015 at 5:47 AM, Friedrich Volkmann <[hidden email]> wrote:
As you are already indicating, 15 is too low a quorum in that case. We
cannot considering 8:7 votes an approval when we cosider 8:1 votes an
approval. That would mean that more negative votes would turn a rejection to
an approval, which is absurd.

Exactly that happened.  There was a proposal with 7 votes, some positive some negative.
3 more people voted no, flipping it to approval.


If the purpose of the wiki procedure is to find consensus, a bare 50% majority indicates
a near complete failure.

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



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

Re: Accepted or rejected?

brycenesbitt
On Sat, Mar 14, 2015 at 12:13 PM, Kotya Karapetyan <[hidden email]> wrote:
Proposal: let's change it to "8 unanimous approval votes or 10 or more votes with at least 74 % approval ones"?

+1 on that.  Anything without a super-majority clearly needs more discussion and/or experience.  The proponents
are free to starting using the tags of course.  Often using tags helps refine them: bad ideas become apparent.

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

Re: Accepted or rejected?

Clifford Snow
In reply to this post by Friedrich Volkmann
The reality is that a tag becomes "approved" once it is adopted by developers and is used extensively. Voting has its purpose, mainly to weed out proposals that need more work. As others have said 8 approvals and 7 declines indicate that more work needs to be done. Even if a proposal receives 8 approvals and no declines, it really hasn't been accepted. 

I would suggest adopting " Conditional Approval" approach. If the proposal receives sufficient votes, it becomes "Conditionally Approved." Only after it becomes widespread and adopted by JOSM and iD it becomes an "Approved" tag. Conditional Approval would last for a period of time which may be extended. For example, a 1 year period with 1 or 2 6 months extensions. If it doesn't reach the widespread use threshold, the tag will be declared as "Not Approved." It will be up to the originators to track it's use and request updates to JOSM and iD. Some subjectivity to the term, "widespread" needs to be applied. For instance, a niche shop might be approved with only a small number of tags. However, a fitness_centre tag would required several hundred before approval.


--
@osm_seattle
OpenStreetMap: Maps with a human touch

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

Re: Accepted or rejected?

Warin
In reply to this post by brycenesbitt
On 15/03/2015 4:44 AM, Bryce Nesbitt wrote:
On Sat, Mar 14, 2015 at 4:24 AM, Jan van Bekkum <[hidden email]> wrote:
This sounds a bit strange to me: a proposal with 8 approval votes and 1 decline would be rejected, while one with 8 approval votes and 7 declines would be accepted.

Anything with that level of opposition (7 declines) is probably a flawed proposal.


Depends on the reasons for rejection;

If the majority point to some failure .. then yes there may be a need for revision.

On the other hand if they all point in different directions and at least most of those reasons can be refuted then I'd say not.

-------------------
Rules .. meant for the obedience of fools and the guidance of the wise.



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

Re: Accepted or rejected?

Friedrich Volkmann
In reply to this post by Clifford Snow
On 14.03.2015 21:27, Clifford Snow wrote:
> I would suggest adopting " Conditional Approval" approach. If the proposal
> receives sufficient votes, it becomes "Conditionally Approved." Only after
> it becomes widespread and adopted by JOSM and iD it becomes an "Approved"
> tag.

No. Editor developers aleady have too much power. Editor support often
depends on the mood of one single person. I would rather say that, for a
given number of occurrences, editor support should be considered a counter
indicator for approval. When usage spreads in spite of no editor support,
that means that mappers choose the tag on purpose. When usage remains
intermediate in spite of editor support, that means that mappers use the tag
only because it is imposed by the editor.

--
Friedrich K. Volkmann       http://www.volki.at/
Adr.: Davidgasse 76-80/14/10, 1100 Wien, Austria

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

Re: Accepted or rejected?

Friedrich Volkmann
In reply to this post by brycenesbitt
On 14.03.2015 21:11, Bryce Nesbitt wrote:
> On Sat, Mar 14, 2015 at 12:13 PM, Kotya Karapetyan <[hidden email]
> <mailto:[hidden email]>> wrote:
>
>     Proposal: let's change it to "8 unanimous approval votes or 10 or more
>     votes with at least 74 % approval ones"?
>
>
> +1 on that.  Anything without a super-majority clearly needs more discussion
> and/or experience.

In that case, we shouldn't mark it as "rejected", but rather as something
like "not proven".

--
Friedrich K. Volkmann       http://www.volki.at/
Adr.: Davidgasse 76-80/14/10, 1100 Wien, Austria

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

Re: Accepted or rejected?

Warin
In reply to this post by Friedrich Volkmann
On 16/03/2015 7:11 PM, Friedrich Volkmann wrote:

> On 14.03.2015 21:27, Clifford Snow wrote:
>> I would suggest adopting " Conditional Approval" approach. If the proposal
>> receives sufficient votes, it becomes "Conditionally Approved." Only after
>> it becomes widespread and adopted by JOSM and iD it becomes an "Approved"
>> tag.
> No. Editor developers aleady have too much power. Editor support often
> depends on the mood of one single person. I would rather say that, for a
> given number of occurrences, editor support should be considered a counter
> indicator for approval. When usage spreads in spite of no editor support,
> that means that mappers choose the tag on purpose. When usage remains
> intermediate in spite of editor support, that means that mappers use the tag
> only because it is imposed by the editor.

Mappers don't use a tag .. even ones 'suggested' by an editor unless they 'fit'.
Beginner mappers, like me, use the wiki in searching for a suitable tag, it aids understanding.
They don't rely on the editor to find suitable tags as it does not provide enough information.
If the wiki description is a poor match but no other tag is found you may find that tag is used or the data is not entered.
Few beginner mappers will make a new tag. They may make a node with a note.. but that is about it.

Approval and rejection at the moment are only tagging group indicators.. the best 'indicator' is that it is rendered.
And that is not a function of JOSM nor iD .. but the renderers .. there are a few of them .. if they all render some OSM object then that tag has 'made it'.
I think the 'approval' and 'rejection' should stay where it is .. it is not the be all and end all of a tag.

As for increasing the 'approval' vote to a minimum of 10 with 75% .. ok ..
as long as there is a time limit on the minimum number of 10,
at the end of, say 6 weeks, the number requirement needs to be dropped altogether.
This would encourage people to vote as after 6 weeks their lack of voting does not matter.


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

Re: Accepted or rejected?

Marc Gemis

On Mon, Mar 16, 2015 at 11:31 AM, Warin <[hidden email]> wrote:
Approval and rejection at the moment are only tagging group indicators.. the best 'indicator' is that it is rendered.
And that is not a function of JOSM nor iD .. but the renderers .. there are a few of them .. if they all render some OSM object then that tag has 'made it'.
I think the 'approval' and 'rejection' should stay where it is .. it is not the be all and end all of a tag.

-1,
 Think about the surface, the turn:lanes, destination or 3D buildings keys. They are not rendered on all "few renderers". Still they are important enough to keep (just assume they just past your approval process), as some navigation software will rely on them or specialized maps.
I don't think "being rendered on all renderers" is a proper decision criteria.

regards

m

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

Re: Accepted or rejected?

dieterdreist

2015-03-16 11:55 GMT+01:00 Marc Gemis <[hidden email]>:
I don't think "being rendered on all renderers" is a proper decision criteria


+1, the list of tags mostly not rendered but well established is long:
opening_hours
wikipedia
start_date
operator
(population) (is actually taken into account when rendering)
turn_restrictions
routes (well, some renderers do show them, but osm carto doesn't)
description
inscription
note
website
phone
url
...

plus all other keys that don't even get imported into most of the rendering databases.


cheers,
Martin

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

Re: Accepted or rejected?

Warin
On 16/03/2015 10:05 PM, Martin Koppenhoefer wrote:

2015-03-16 11:55 GMT+01:00 Marc Gemis <[hidden email]>:
I don't think "being rendered on all renderers" is a proper decision criteria


+1, the list of tags mostly not rendered but well established is long:
opening_hours (used by some renderers into GPS 'maps' such as OSMAnd)
wikipedia
start_date
operator
(population) (is actually taken into account when rendering)
turn_restrictions (used by at least some routers on GPS 'maps')

routes (well, some renderers do show them, but osm carto doesn't) (Rendered by some as you say)

description
inscription
note (not rendered .. for use by mappers to make notes to other mappers ? thus not required to be rendered?)
website
phone
url
...

plus all other keys that don't even get imported into most of the rendering databases.


cheers,
Martin

I think most, if not all, of the tags you list .. I'm not using... other than the ones I've made notes on... while they may be long established they may not be used by new mappers and thus be less populated than they could be. Hard to measure, but that is my take on it. I think there are keys that could be rendered that simply miss out because they are not frequently used or are not present as a tag option.

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

Re: Accepted or rejected?

Marc Gemis

On Mon, Mar 16, 2015 at 10:04 PM, Warin <[hidden email]> wrote:
inscription
note (not rendered .. for use by mappers to make notes to other mappers ? thus not required to be rendered?)

Visible in a popup in geschichtskarten for historical items.

But you were talking about all renderers I thought. Now you seem happy that there is 1 renderer showing the feature/data ? 
I'm still convinced that features that people want to map will be mapped, regardless of the state of the tagging proposal.


m.

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

Re: Accepted or rejected?

Kotya Karapetyan
Dear all,

I think we deviated from the original question quite a bit. The point was that the current number of votes proposed in the wiki for accepted/rejected decision was self-contradicting. Even if there may be different opinions on that, the very discussion shows that the situation is not clear.

I propose to clarify it by changing the recommended number of votes in https://wiki.openstreetmap.org/wiki/Proposed_features#Approved_or_rejected
from "...8 unanimous approval votes or 15 total votes with a majority approval..."
to "...8 or more unanimous approval votes or 10 or more total votes with more than 74 % approval...".
This will not change anything in terms of the ongoing discussion of how the approval influences other things. So the discussion can continue. But we'd introduce some mathematical logic in the process.

I don't think there is a procedure to vote on such proposals, so please just give it +1 here if you agree. We change it when we have 8+ "plus ones" if there are no significant objections to this change. Once again, please note: we are not discussing the consequences of approval/rejection, we just change the rule of thumb recommendation to a mathematically more sound one.

Cheers,
Kotya


On Tue, Mar 17, 2015 at 7:35 AM, Marc Gemis <[hidden email]> wrote:

On Mon, Mar 16, 2015 at 10:04 PM, Warin <[hidden email]> wrote:
inscription
note (not rendered .. for use by mappers to make notes to other mappers ? thus not required to be rendered?)

Visible in a popup in geschichtskarten for historical items.

But you were talking about all renderers I thought. Now you seem happy that there is 1 renderer showing the feature/data ? 
I'm still convinced that features that people want to map will be mapped, regardless of the state of the tagging proposal.


m.

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



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

Re: Accepted or rejected?

jonathan-2
+1

Jonathan

http://bigfatfrog67.me

From: [hidden email]
Sent: ‎Tuesday‎, ‎17‎ ‎March‎ ‎2015 ‎14‎:‎04
To: [hidden email]

Dear all,

I think we deviated from the original question quite a bit. The point was that the current number of votes proposed in the wiki for accepted/rejected decision was self-contradicting. Even if there may be different opinions on that, the very discussion shows that the situation is not clear.

I propose to clarify it by changing the recommended number of votes in https://wiki.openstreetmap.org/wiki/Proposed_features#Approved_or_rejected
from "...8 unanimous approval votes or 15 total votes with a majority approval..."
to "...8 or more unanimous approval votes or 10 or more total votes with more than 74 % approval...".
This will not change anything in terms of the ongoing discussion of how the approval influences other things. So the discussion can continue. But we'd introduce some mathematical logic in the process.

I don't think there is a procedure to vote on such proposals, so please just give it +1 here if you agree. We change it when we have 8+ "plus ones" if there are no significant objections to this change. Once again, please note: we are not discussing the consequences of approval/rejection, we just change the rule of thumb recommendation to a mathematically more sound one.

Cheers,
Kotya


On Tue, Mar 17, 2015 at 7:35 AM, Marc Gemis <[hidden email]> wrote:

On Mon, Mar 16, 2015 at 10:04 PM, Warin <[hidden email]> wrote:
inscription
note (not rendered .. for use by mappers to make notes to other mappers ? thus not required to be rendered?)

Visible in a popup in geschichtskarten for historical items.

But you were talking about all renderers I thought. Now you seem happy that there is 1 renderer showing the feature/data ? 
I'm still convinced that features that people want to map will be mapped, regardless of the state of the tagging proposal.


m.

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



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

Re: Accepted or rejected?

Tod Fitch
In reply to this post by Kotya Karapetyan
+1

On Mar 17, 2015, at 7:04 AM, Kotya Karapetyan wrote:

Dear all,

I think we deviated from the original question quite a bit. The point was that the current number of votes proposed in the wiki for accepted/rejected decision was self-contradicting. Even if there may be different opinions on that, the very discussion shows that the situation is not clear.

I propose to clarify it by changing the recommended number of votes in https://wiki.openstreetmap.org/wiki/Proposed_features#Approved_or_rejected
from "...8 unanimous approval votes or 15 total votes with a majority approval..."
to "...8 or more unanimous approval votes or 10 or more total votes with more than 74 % approval...".
This will not change anything in terms of the ongoing discussion of how the approval influences other things. So the discussion can continue. But we'd introduce some mathematical logic in the process.

I don't think there is a procedure to vote on such proposals, so please just give it +1 here if you agree. We change it when we have 8+ "plus ones" if there are no significant objections to this change. Once again, please note: we are not discussing the consequences of approval/rejection, we just change the rule of thumb recommendation to a mathematically more sound one.

Cheers,
Kotya


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

smime.p7s (2K) Download Attachment
12345