[OSM-talk] Organizational mapping policy, the role of the DWG

Johan C osmned at gmail.com
Sat May 24 04:14:21 UTC 2014


Hi Frederik,

> That's because you have been involved in a data import which didn't
follow the import guidelines properly, and now you're biased against DWG
because we called you out on it and terrified you. But you shouldn't mix
those bad feelings with the issue at hand here as they are completely
separate.

[Don't take the following paragraph personally, it's about your role and
not you as a person:]
Ha, Freud speaking :-) To be honest, the SWOT of OSM (
http://wiki.openstreetmap.org/wiki/Talk:Future) already has several items
about decision making. But as you bring it up, you can compare the current
level of decision making on imports in OSM by the following example. Person
A is having a discussion with person B. Person A knows that person B has a
gun which could be used to kill person A, but person A tries to forget
about that in the discussion. Than  person B has two wishes. Person A
decides after serious consultation with a group of other persons to
incorporate wish 1 in the decision, but not to incorporate wish 2 for
reasons which are valid for the group. Five months after the decision is
made person B does not like that wish 2 was not seen as an order. So,
person B shows his gun. What should person A do now if he wants to stay
alive?
It's not that 'we' called out. You as a list member (since you did not
identify in a role speaking on behalf of the DWG) posted something which is
to date still unclear to me: you didn't answer three questions relevant to
me. So nope, I don't have bad feelings against DWG. In fact, I think they
are doing an important job in keeping the database healthy. And they could
even do a better job when roles are divided. Since it's likely that imports
will increase in the near future because of the massive release of open
geodata in 2015 in Europe the role of the DWG in taking decisions based on
clear guidelines and some clear requirements will even get more important.

> The DWG is occasionally (by those whom we admonish or block for bad
behaviour) accused of acting in an arbitrary way - people would prefer
to have a rule that says: Don't do this or we will intervene, instead of
just having a general hand-wavy "as long as nobody complains you might
be fine".

And people are right - for the DWG, too, it would be nice to be able to
say "these are the rules that the community has agreed on", instead of
making things up as we go along.

I completely agree with you. It wouldn't be difficult to have a right
balance of powers.
https://wiki.openstreetmap.org/wiki/Open_Data_License/Community_Guidelines/How_We_Create_Community_Guidelines
already
shows a way that the community is involved in a good manner (transparancy)
and that the balance of powers is correct (OSMF in the legislative role by
endorsement). We can learn from Wikipedia about the judicial role: they
have got mediation:
http://en.wikipedia.org/wiki/Wikipedia:Mediation_Committee/Policy.
Balancing of powers in which the DWG only has an executive role can make
decisions by the DWG more accepted by the community.

> Aren't you active in the "Future Group" - should you not have some
understanding for the idea of looking ahead and thinking about
tomorrow's problems? Instead you essentially say, "the house is not on
fire so why establish safety rules". That's not very forward-looking is it?

OSM can't do without rules. That's why I think the line 'supporting, but
not controlling' is ridiculous. Indeed, central command without governance
is dangerous: the community should be in charge. But no, a motorway has not
be be tagged as a footway just because an indivual member likes that. I'm
one of the people behind the SWOT so I'm forward looking. It wouldn't be
bad to have a SWOT for the DWG. That would lead to risk assessments of
current and future risks and enable a workplan made in conjunction between
DWG and the community.

> I don't think this is "DWG picking up the legislative role" - it's more
"DWG consulting the community" because, as I said, as long as there's no
guideline we'll decide things on a case-by-case basis.

Also when there are clear guidelines, policies, requirements en so on
decisions will be made on a case-by-case basis. But within the boundaries
of these clear guidelines, policies, requirements and so on. That's the
fate of an executive body. Clear guidelines, policies, requirements etc.
should always have a context, a goal: what are they made for, what's the
spirit behind them? And with mediation in place, the community will have a
place to go to when they don't agree with a decision by the DWG. And that
may result in an improvement of guidelines, policies, requirements and so
on.

> I don't know what the problem is here. Most of DWG discussions happen on
the DWG mailing list. We discussed an issue, we decided to ask what the
community was thinking - where's your problem?

In the fact that it fell out of the sky, in the fact that the LWG approach
was not followed and in the fact that questions (like the question of
Clifford) were not answered.

> What? Do we have a language problem here? Let me quote from Paul's
message:...
> In fact, this message from Paul constitutes the very transparency that
you were asking for earlier. What can be more transparent than emailing
the list and saying: "Look, people, we think we might need some policy
here, what do you think?"

It would have helped when the process had been sketched (this is step 1,
step 2 will be a wiki page which can be edited by the community in the
coming x months, step 3 will be an endorsement process by OSMF in month
...). One thing was very good: Paul clearly identified himself as speaking
on behalf of the DWG

> I will answer these questions to the best of my ability

Thanks for answering my four questions.

It's easy for me to have comments on the DWG from a sideline. The Dutch do
have a expression "De beste stuurlui staan aan wal'. I actually don't know
the right translation for it in English, it might be something like 'It's
easy to shout from the sideline'. Therefore I want to help the DWG. Which
also enables me to work closely together with you. I admire (to make sure:
this is absolutely not meant cynical) the things you are doing for OSM like
the clearly visible amount of time you are spending on OSM, like having the
Geofabrik tools and much more.

What I can offer: on average about five hours per week. Over 25 years of
professional expertise in finance and control (yep, I'm not an ICT guy).
Over 4,5 years OSM experience including + 6000 changesets. A crazy mapper
(in the words of the wonderful http://www.hdyc.neis-one.org/?). Experience
in dealing with vandalism. Experience in a big import with all it's ups and
downs. Some idears I would like to work on as posted be me yesterday and in
this posting. A bit of humour. Quite frequently positive energy. Knowledge
of the difference between process and content. Some grey hairs.
Transparancy, integrity, accountability. Open for feedback. The promise to
answer to questions within a week. Communication with as much respect and
positive energy as possible, which also means identifying myself in which
role I communicate. And maybe most important: love for OpenStreetMap, which
I hope will have the best mappingcommunity and map(data) in the world
within a decade.

Cheers, Johan
Dutch community member


Johan,
>
> On 05/23/2014 12:43 AM, Johan C wrote:
> > So, if there’s no fire, no epic problem and a constant need to be
> integer and delicate, why is the DWG picking up the legislative role?
>
> I'd perhaps not word it in such a manner.
>
> The DWG is occasionally (by those whom we admonish or block for bad
> behaviour) accused of acting in an arbitrary way - people would prefer
> to have a rule that says: Don't do this or we will intervene, instead of
> just having a general hand-wavy "as long as nobody complains you might
> be fine".
>
> And people are right - for the DWG, too, it would be nice to be able to
> say "these are the rules that the community has agreed on", instead of
> making things up as we go along.
>
> These things take a lot of time and discussion. It can easily be a year
> from suggesting that the community think about something, to the
> development of something that might be called a policy. The DWG cannot
> and will not usually poll the community on a concrete issue ("we have
> received this complaint where a company did this and that without
> consulting anyone, now what should we do"); and by looking at Wikipedia
> I think it would be naive to assume that OSM won't see more and more
> corporate or organised editing.
>
> Aren't you active in the "Future Group" - should you not have some
> understanding for the idea of looking ahead and thinking about
> tomorrow's problems? Instead you essentially say, "the house is not on
> fire so why establish safety rules". That's not very forward-looking is it?
>
> My take is: We don't need a policy now but we will some day need one;
> even today, while not yet strictly necessary, having a policy would
> create a fair playing ground for everyone.
>
> I don't think this is "DWG picking up the legislative role" - it's more
> "DWG consulting the community" because, as I said, as long as there's no
> guideline we'll decide things on a case-by-case basis.
>
> > To answer that question for myself, I wanted to view the DWG minutes.
> They can be found on this page:
> http://www.osmfoundation.org/wiki/Data_Working_Group
> > That is, the DWG only seems to meet once a year. The last meeting
> minutes are from 2014-01-05. The previous one is from 2013-02-07. Too bad,
> nothing in this minutes resembles discussions about an organizational
> mapping policy.
>
> I don't know what the problem is here. Most of DWG discussions happen on
> the DWG mailing list. We discussed an issue, we decided to ask what the
> community was thinking - where's your problem?
>
> > This lack of transparancy, along with a luckily failed attempt to issue
> guideline requirements
>
> What? Do we have a language problem here? Let me quote from Paul's message:
>
> "For this reason the DWG is considering if it is necessary to issue
> guidelines for organizational editing."
>
> So, "considering if it is necessary". He doesn't even say that we have
> made up our mind yet, nor that we were "attempting to issue" something.
>
> Then further:
>
> "Some possible guideline requirements could involve ..."
>
> All this was meant as an (early) consultation of the community on this
> list with the aim of hearing what people think. I don't know how in the
> world you got the impression that we were "attempting to issue
> guidelines" and that this attempt had "luckily failed"!
>
> In fact, this message from Paul constitutes the very transparency that
> you were asking for earlier. What can be more transparent than emailing
> the list and saying: "Look, people, we think we might need some policy
> here, what do you think?"
>
> > makes me confused, worried and even a bit terrified.
>
> That's because you have been involved in a data import which didn't
> follow the import guidelines properly, and now you're biased against DWG
> because we called you out on it and terrified you. But you shouldn't mix
> those bad feelings with the issue at hand here as they are completely
> separate.
>
> > Some questions for the DWG, as a call for transparancy:
>
> I will answer these questions to the best of my ability but I think the
> questions are at odds with the way the DWG, or indeed almost any other
> OSMF working group, functions. I am very much interested in separating
> the questions "should we have an organisational mapping policy and what
> should it say" from your other point that I'd paraphrase as "are the DWG
> really a bunch of secretive power-grabbing bastards that should stop
> meddling with stuff outside of their mandate" ;)
>
> > 1)    Being an executive body, why do you think you have the mandate to
> being a legislative body as well?
>
> I think you're misunderstanding the situation. If you want to stick with
> political analogies, DWG is an executive and partly also judicative
> body, but the laws on which we base our work are extremely general -
> with a few exceptions, essentially it's "don't be a dick". We then have
> to decide who is a dick and why. You can get blocked and your edits
> reverted today for letting 20 of your pupils add imaginary data to OSM -
> organizational edit policy or no organizational edit policy.
>
> Having a few more rules that codify, for everyone to read, when you are
> considered to be a dick and when not, would likely make things better
> for everyone - provided of course these rules are not made up from thin
> air but agreed by a broader community. Hence the message that started
> this thread.
>
> > 2)    Are you considering other guidelines, requirements, policies or
> standards at the moment? If yes, which?
>
> We have, on and off, discussed whether we should consolidate the various
> "how to be good mapper" documents on the wiki (there are 3 or 4
> different pages) into one document but that hasn't had much traction.
> The last guideline/requirement/policy/standard that we made up and that
> was finally signed off by OSMF was the document created in response to
> the NE2 situation where we wrote down that you'll be blocked if you're a
> bad guy - not much new but it helped.
>
> I think you might have a wrong picture of how DWG works. It's just a
> small mailing list where one person floats an idea and another person
> says "sounds good, let's work on this a bit more" or maybe "nah, I don't
> like this" and then if something gets enough traction we do something
> about it. It is not that we have an agenda with issues and deadlines and
> whatnot, or a "process" that leads a document from draft to finalization
> in 36 months or so. We're not the European Council ;)
>
> Just because we're not considering a guideline about, say, mapping
> disputed territories at the moment, this could still mean that we send
> an email to the talk list in 5 days that reads just like Paul's - "hey
> people, we think we might need a policy here, it could encompass these
> points, what do you think". This is not a process that needs ripening
> through countless meetings, and consulting with the community doesn't
> mean there will be a policy five days later!
>
> > 3)    What process did you have before May 13 to start considering
> issuing guidelines for organizational editing?
>
> The generic OSMF process for issuing guidelines is that a working group
> comes up with something, then it goes to Management Team for approval,
> and depending on what the nature of the thing is, Management Team might
> also ask the OSMF board of directors for approval.
>
> Whether or not the working group in question involves the community in
> coming up with a policy is the decision of the working group; it might
> be more appropriate for some working groups (like DWG) or less for
> others (like OWG).
>
> For this particular issue (organizational mapping policy), we haven't
> got a roadmap - we just wanted to hear opinions from people.
>
> As far as DWG is concerned, there's of course also something that I
> would call "best practice". For example, after telling someone to stop
> reverting someone else's edits, how many times will we tell him before
> blocking him for longer? Things like these are not codified in policies
> that go through MT and board and are published, but they're just "the
> way we usually do it", and you won't find these written down.
>
> > 4)    Did you have meetings in the past 12 months which were not
> published? And if so, could you please publish these minutes on the DWG
> site?
>
> No. Frankly I am very happy that we are able to do most work without
> having to actually agree on a date and time on IRC.
>
> Bye
> Frederik
>
> --
> Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20140524/9d418b5f/attachment-0001.html>


More information about the talk mailing list