[OSM-talk] [english 95%] A process for rethinking map features

Mike Harris mikh43 at googlemail.com
Sat Aug 15 13:09:41 BST 2009


Kai - this makes very good sense. But how do we set up a working group? And
would it - and its findings/recommendations - be acceptable to the majority
of the OSM community?

I am increasingly convinced that the discussion group is not going to cut it
when the subject is so complex. 


Mike Harris

-----Original Message-----
From: Kai at VieleVisels [mailto:kai at vielevisels.de] 
Sent: 15 August 2009 11:13
To: talk at openstreetmap.org
Subject: Re: [OSM-talk] [english 95%] A process for rethinking map features

Hi,
what's our present way: Someone makes a proposal, some other people make
their own suggestions, or refuse the proposal but seldom there are
suggestions to reach a consensus.

But no one collects the requirements for this topic in a structured manner!

I think, establishing a (small) working group is the right thing to do. And
the first job of this WG is to collect all data (via Mailing List, Wiki,
specific forums, ...). Then everyone has the chance, to add his requirement
(eg country specific, mapper or developer of editor, cyclist or hiker, ...).

Out of this, the group can work out a proposal which considers all users. 
Specific questions could be discussed with people not in the WG who
contributed to the requirements.
This proposal should go to the proposal page and be voted on. All people
having contributed to the requirement should be informed of the ongoing vote
to have the possibility to discuss and comment in the unlikely situation of
a not working proposal.
The result should go into the wiki, and also in the editors (since i know
many osm mappers who don't participate in the mailing list and don't want to
search the whole wiki for information and just do it like they think...)

For our path/footway discussion which started this:
The WG has to collect information:
* find influencing factors (laws, routing, local rules, ...)
* laws for accessabbility in the different states and countries (implicit
and explicit)
* which signs are used in reality (in all countries)
* what's the meaning of the signs
* what should be displayed on the map
* affected tags
* ...
* small survey how the tags are actually used and rendered
* structure this information
The WG could work out a proposal. I'm convinced, there is a solution to our
discussion but we won't find it exchanging hundreds of mails in the Mailing
List.

Yours Kai
----- Original Message -----
From: "Tom Chance" <tom at acrewoods.net>
To: <talk at openstreetmap.org>
Sent: Monday, August 10, 2009 4:49 PM
Subject: [english 95%] [OSM-talk] A process for rethinking map features


>
> Dear all,
>
> If the wood/forest and path/footway arguments have taught us one thing,
> it's that the current model doesn't work all the time (100s of emails,
> disorganised wiki discussions, votes with 20 or so random people). We
> develop, over years, one set of tags like
> highway=footway/cycleway/bridleway/etc. and then over time we realise the
> schema isn't quite right. But we're incapable of discussing it in a
> structured manner, and we rarely get a useful consensus.
>
> For simple matters like proposing a completely new, minor tag it's fine.
> Where competing proposals for new features, like house numbers, live side
> by side we generally find a superior solution gaining traction.
>
> Where proposals throw up bigger or more complicated questions about
> existing tags, used on thousands or even millions of nodes and ways, the
> whole thing is falling apart.
>
> So...
>
> I propose that we grow up a little and use something like this process:
>
> - Tags are proposed on the wiki, no change to current practice
> - If the proposal throws into question existing, accepted tags, defer the
> proposal to small working groups
> - These working groups study the wider questions and formulate a complete
> proposal for new tags, deprecation, etc.
> - At SOTM present and discuss their proposals and vote
> - If proposals are accepted, a combination of carrot (rendering
> stylesheets, Potlatch presets, etc.) and sticks (error checking,
> auto-correcting bots) to implement the accepted proposals
>
> So for example Nick Whitelegg and Martin Simon might lead a group to work
> out how best to tag paths of all kinds. If their proposal was accepted at
> SOTM 2010, somebody would create a map highlighting all the ways that
> probably need to be corrected and a massive effort to bring things in line
> with the new schema would kick off.
>
> Does this sound workable?
>
> Regards,
> Tom
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk 








More information about the talk mailing list