[Tagging] Traffic_sign discussion

yo paseopor yopaseopor at gmail.com
Wed Oct 10 19:22:21 UTC 2018


I will explain the things from my point of view.

There was a discussion about direction in traffic signs because a problem
in major online editor iD.

32 messages that starts Fri, Sep 28, 4:52 AM (12 days ago) and finnish Oct
3, 2018, 12:04 AM (7 days ago) . Five days of discussion.

-In the first message Bryan Housel comments a problem "While reviewing a
pull request to add Traffic Sign presets to iD, I came across a tagging
issue with how traffic sign directions are tagged.
The details are here https://github.com/openstreetmap/iD/pull/5333"

-On 4th message Simon Poole comments "I actually mentioned the issue in
Milano. "

-On 9th message Simon Poole says "There are a total of 37'000 forward /
backward variants that would have to be migrated to  traffic_sign + a
suitable sub tag, not an awful lot in the grand scheme of things, but needs
to be done."

So the need for a big change in existing traffic signs was written in
tagging list, and nobody's says "No, it is not a good idea". Well, I did
not agree with that.

I have talk for first time in message number 21. Instead I manage 40
presets ,3 styles and more than 10 configuration files for Kendzi3D plugin
in JOSM, and 3 projects in taginfo with more than 24000 pairs (key=value) I
don't talk. Until I want to justify the nowadays proposal because if you
read it is assumed the change will be done in iD - major online editor of
OSM. Also I was thinking: I don't like changes but what can I do? Fight
against iD solution(the major editor) (Remember the Mapsme subway solution,
isn't it) ?
In 28th message I have said Ok, make you the changes -ironically- (there
was a way to say Hey! There are a lot of changes to do, are you sure?
Changing 30000+ nodes, are you nuts?)

Weekend arrives . I have a little free time so...Really Do I have to fight
against iD proposal? (I see the pull request merged in their github). Or
say the opposite to Simon Poole? Ok, I will give up. Discussion is ended
(there were no more messages). iD pull request was applied so it is
imminent the edition of traffic signs...with two different schemes.
It is better to help, so I have edited all the presets (I'm not a
programmer so for me it is a difficult thing, a pain in the ass if you want
to tell it) , the styles, taginfo (goodbye to the forward backward subkey ,
but hey! traffic signs now will be edited via iD so a lot of people would
edit them .

And after the tool work...why I can't help more? There is more than 24000
nodes (reading taginfo). Ok, but for not having problems the changeset
message will be very clear: "#fastag #traffic_signs Apply
traffic_sign:direction tag to avoid problems with new iD editor as an
agreement on tagging list". The way to do is simple: I would have only made
a simple translation: traffic_sign:forward=* to traffic_sign=* and
traffic_sign:direction=forward, and the same for backward.Traffic signals
have also this solution so It can not be so bad after all. And now traffic
signs will be edited by iD. It is a win-win thing.

I have made the first changes in nodes and then check presets, styles -the
day after I have checked taginfo- and it works. Also I have checked my
email and OSM profile and there was not any message. So I think I was in
the good way - I was helping to do this big change at all.
And going zone by zone making specific overpass queries to make the things
the best as I could with a little computer and low programming knowledge.
When I put the new tags style were working and it shows every traffic sign
different in every country. It was a hard task : 55 changesets with about
16000+ traffic signs modified to the new scheme. Heavy work done.

Then "shit happens" . Mknight says "Wäre es nicht irgendwie sinnvoller, ein
issue für iD zu schreiben, statt etabliertes Tagging zu ändern?" Well, I'm
not German so I have used Google Translator to guess the idea was not of
his agreement. Well, in tagging there is not more messages at all and
people are agree with the change proposed by iD people. In a big thing like
OSM not every one can be agree with it and Mknight does not participate on
the discussion. I hope some people of their community explain to him the
possibility to edit with iD thanks to this change.

But then Mueschel says something similar...D'Oh! "The discussion ended with
your question about the change, not a single answer approving it. Mass
edits should be announced and agreed on in a broader community, and not in
the depth of a thread without anybody noticing."

What? The discussion was ended without nobody against it, Simon Poole
saying there is big change to do, people congrats and making petitions to
iD people...and I assuming before or later I would have to change all the
JOSM stuff. Ok. In +16000 and 55 changesets there were some errors surely,
but what percentatge? How many nodes do I have modified by
error...because...ways does not have a traffic_sign key, isn't it ?

Well. I want to publish the messages people says to me on the changesets:

Mknight "Wäre es nicht irgendwie sinnvoller, ein issue für iD zu schreiben,
statt etabliertes Tagging zu ändern?"
Mueschel "Please stop this undiscussed mechanical edit until further
discussion.It breaks many cases of traffic sign tagging."
Mueschel "The discussion ended with your question about the change, not a
single answer approving it. Mass edits should be announced and agreed on in
a broader community, and not in the depth of a thread without anybody
noticing. You also edited ways with traffic_sign:* tags, where this scheme
will not work at all."
Peilscheibe "I reverted this changeset because diligent tagging (e.g. about
different traffic signs per direction) was removed and replaced by wrong
information. This is not acceptable at all."
Peilscheibe "Just a few examples:
http://osmlab.github.io/osm-deep-history/#/way/25717685
http://osmlab.github.io/osm-deep-history/#/way/629964077
http://osmlab.github.io/osm-deep-history/#/way/629964082
On this ways (and a lot of others) there was a different traffic sign
tagging per direction.
Why? Because in reality the ways are differently signposted per direction
which has de facto and de jure implications on traffic members.
Your edit removed this diligent tagging and wrong values were contributed.
This is a blatant information loss and it is wrong."
geri-oc "Du hast hier Änderungen vor genommen die nicht zutreffend sin. Die
Richtung in Sinne des Weges ist backwart oder forwart. direction gibt eine
Richtung in Grad oder Himmelsrichtung an.

Ich erwarte eine Revert in angemessener Zeit von dir - zu mindest für
diesen Änderungssatz. Über andere unberechtigte Änderungen kannst du gern
im Forum weiter diskutieren, wo ein Revert deiner Änderungssätze ebenfalls
diskutiert wird:
https://forum.openstreetmap.org/viewtopic.php?id=64028

Gruß Gerd - geri-oc"
geri-oc "direction ist ein verkehrter Schlüssel in Bezug auf die
Wegrichtung von Verkehrszeichen. Die Änderungen sind alle falsch, da
direction in Grad oder Himmelsrichtung angegeben wird."
geri-oc "Also, the specification direction is correct only with degrees or
compass. forward or backward is related to the direction of the path in
OSM.I am for a revert of these changes (worldwide)."
geri-oc "57,7274424, 16,5246249 ist nach den richtigen Angaben laut WIKI
und JOSM-Plugin gemappt:
57,7273945, 16,5241181 hast du geändert.
Es existieren zwei verschieden Schlüssel in unmittelbarer Nachbarschaft und
nach der alten Auswertung erscheint der zweite Node nicht mehr. Ich werden
den CS reverten, falls du es nicht selbst machst. Deine Quelle Mapillary
kann auch nicht stimmen, da einige Schilder gar nicht in Mapillary sind,
sondern vor Ort erfasst wurden."

geow "I asked the data working group to revert all changesets of user
https://www.openstreetmap.org/user/yopaseopor/history where the cS comment
is: "#fastag #traffic_signs Apply traffic_sign:direction tag to avoid
problems with new iD editor as an agreement on tagging list“
Thanks
geow"

and then Streckenkundler "It sucks to adjust the data to the Editor. I have
chosen this worde deliberately. This is the best way to scare mappers from
OSM. In my opinion, all your changes must be reset immediately and
promptly. Stinky Greetings..."

Hey! I can be wrong but don't insult me!

Streckenkundler "Whether iD is bad or not, I can't say. Your way of pushing
your ideas was Bad. First a huge proposal, where hardly any one looks
through and it is not yet Decided. Then quickly change data to match the
programming of iD to create Facts.
That's not how it works. Thanks to woodpeck for the Revert."

Well, it was a disaster for me, people claiming "my head", etc.but...
Crisis also means opportunity. So first of all: what is the people wants
about traffic signs? It is the moment to have a big consensus and avoid
these kind of episodes. So I send this message to the tag list. And I hope
some day before or later we have a complete traffic sign scheme.

But I have to say I'm sorry for the misunderstanding of what a consensus is
in a tagging list... but What is a consensus in this list?

yopaseopor

On Tue, Oct 9, 2018 at 10:28 PM Frederik Ramm <frederik at remote.org> wrote:

> Hi,
>
> On 10/09/2018 05:42 PM, yo paseopor wrote:
> > It is not the first attempt to do that. Last days, with iD
> > implementation and my message I have think it was the solution. Also I
> > have waited some days and communicate to this list my intentions to
> > adopt the proposed iD scheme. But when I start to do the
> > modifications... People complains about it (I am sorry if there was some
> > errors "translating" to the new scheme).
>
> Yes, DWG has also received complaints about these edits, and I am in the
> process of reverting them.
>
> At the very least you should have established a consensus on this list
> for the precise edit you want to perform. You should have said: I am
> going to load objects tagged so-and-so, and I am going to apply these
> modifications using these tools.  (consensus doesn't mean everyone has
> to be in favour, but you simply went ahead and changed things and wrote
> in your changeset comment "#fastag #traffic_signs Apply
> traffic_sign:direction tag to avoid problems with new iD editor as an
> agreement on tagging list" which almost sounds like you were fixing a
> bug and there was consensus, neither of which is strictly true.
>
> I'm not saying these edits cannot ever be made, but they can certainly
> not be made in a buggy fashion after a half-hearted attempt at
> discussion with no discernible outcome.
>
> Bye
> Frederik
>
> --
> Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"
>
> _______________________________________________
> Tagging mailing list
> Tagging at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20181010/488a2adc/attachment-0001.html>


More information about the Tagging mailing list