[OSM-talk] Things People Say
tangararama at gmail.com
Tue Jan 3 08:02:40 GMT 2012
Thanks for your response and all the other responses from everyone.
I'm a bit delayed in getting back (bad time of year). My message
sparked a bit of debate and many people raised a lot of interesting
points which parallel my own thinking. I'm not going to respond to
each message individually because a) I don't have time, and b) I don't
want to create another snowball, but it was all really appreciated and
I think some people really hit some core issues and it shows we have a
lot of very smart people in the community. There are some differences
in vision but a common passion for the project which is great.
Frederik - I appreciate your insightful reply and I really do get what
you're saying and why you're saying it. I haven't been a participant
on the lists but I've been lurking the past few months. I think the
community benefits hugely from your involvement. I've trawled through
all the various debate topics and come up with what seem to represent
where my thought is coming from - and hopefully this might explain why
some have a difference in our vision of what OSM can be.
SENSE OF COMMUNITY / IDENTITY
A couple of other people brought this up too and I think it's
important. I didn't realise how important this was until reading about
the idea to make maps less prominent on OSM. The map is the
representation of the fruits of our hard work - I'll be sad to see it
relegated to a dark corner. I enthusiastically contribute because I
feel like I am a part of something. I see the maps, and I make them
better where I can. It's a thrill to see my updates appear. That's the
"gamification" aspect right there - seeing contributions appear on the
map right alongside the work of other contributors.
If we take this "we're just a database" to the extreme, and leave all
rendering to external parties. No maps on OSM. "Come and contribute to
our database".... well... I'd stop contributing today if it were like
that. Why? Because that's when the project feels more like a labour-
harvesting exercise for the benefit of others. There's just not much
thrill putting something in a database that might appear on the maps
used by A, B, C, X, Y, and Z companies at some stage. Do you believe
the end users of those maps give a stuff about OSM? When you use
Google Maps do you think about Tele Atlas?
Don't get me wrong - I love the idea our data will be shared and
rendered externally, but I think maps need to remain prominent on the
OSM website because it's the visual representation of our work and
cements a sense of community.
"THE WORLD'S MAP" - MY VISION
This is getting into dream stuff, but I think it's important to share
our dreams... The fact you can't represent the needs of everyone on
one map is absolutely correct. I didn't literally mean a one-size-fits-
all approach. I mean we should provide some interface that can
interpret and present the relevant parts of our fantastic database of
information in flexible new ways. I was deliberately vague on this
because it's a fuzzy concept. There were some very interesting points
raised about cartography, and also quite valid arguments about the
demand to render and serve all that graphical information being cost
prohibitive - totally agree. I didn't mean to trivialize by drawing
comparisons with Wikipedia.
I think of the places where our data will get to... from tiles
externally rendered and shared through MapQuest, OpenCycleMap, etc..
or custom-rendered maps produced on a range of personal client-side
packages - e.g. used by journalists, government, academic, industry,
commercial, and private use - all producing stunning customised maps,
mashed up with statistics, wikipedia information, etc... I really
hope all this flourishes.
And then I think of OSM rendering all the Mapnik tiles and I get
disheartened and it's easy to go down the path of lessening the
relevance of maps at OSM and leaving these issues for others to
tackle. Frederik is right when he asks why should we decide what's on
the map? Is there a better way to do this? Can we do away with
tiles? Tiles work for graphical maps - but we have a database. Are we
trying to make a database fit into a traditional graphical map by
thinking in terms of tiles? I think a lot of this also ties back with
Frederik's valid concerns raised just before Christmas about the
volume and scope of data going into the system and the need to be able
to drill up or down and separate data into layers of relevance.
My vision is this - for the visitor to OSM (or wikipedia-type sister
project) to become immersed in some kind of interactive map viewing
experience. Let them visually mine our database. Do away with rendered
tiles. Use client-side in-browser rendering. Choose from 1000's of
different map style sheets / templates / cartographic style sheets -
whatever you want to call them... use these to display their world in
they way that's important to them. And let them contribute their own
templates to the project (almost like WordPress templates). View the
world from a "Chinese or Tibetan" viewpoint, or any other point of
view they can possibly imagine. The template is a starting point. Let
them filter, drill, cross reference, look at different layers.
These templates are a whole new layer of data - they define data
subsets and ways to present it visually. I think the view of our data
is as important as the geographic data itself. As the geographic
database becomes more complete, contributors to the project will move
away from contributing new geographic data, and focus on contributing
and perfecting map style templates.
The template may even have a role in editing the map data and help
with one of Frederik's other points he raised before Christmas
relating to the overhead needed to edit (i.e. that you must download
the entire scope of data for the geographic area you're working on). A
template will only download the relevant data (e.g. public transport
routes). Thus, an edit mode invoked from that template view could take
us into an editing mode where we only work on that data subset.
Just my thoughts and dreams. Lets hope 2012 is a great year for OSM.
Happy New Year,
On 30/12/2011, at 11:20 AM, Frederik Ramm wrote:
> On 12/29/2011 10:39 PM, Ben Johnson wrote:
>> There seems to be a duality of identity here. On one hand, some are
>> saying lets make it more accessible and friendly to "ordinary
>> On the other hand, some appear embarrassed by the prominence of
>> maps to
>> represent what our community is all about, and they want to retain a
>> geeky "we are not a map, we are a database" ideology.
> This is not a geeky ideology, this is the heart of our project.
> Otherwise we'd all be using the Gimp (or maybe Inkscape), and not
> JOSM or Potlatch.
>> The two goals are completely incompatible because "ordinary people"
>> expect OSM to be all about maps.
> When in fact we're about providing the ingredients so that great
> maps become possible. This is the big thing about OSM: We don't
> decide how the map looks like. We don't decide what is shown on the
> map. We don't make all these decisions for you. If you want these
> decisions made for you, use Google Maps. If you want to be part of a
> project where one guy can make a cycle map and another guy can make
> a transport map and someone else prints posters and someone else
> still makes nautical maps, all from the same data, and all precisely
> because the core of the project doesn't make the map but just the
> raw data, then OSM is for you.
>> Again, what is embarrassing about a map?
> It is not embarassing, but misleading. If we were to convey the idea
> that we are about "a map", then people would come and say "uh, how
> can I change this motorway to become orange instead of blue, and
> also I would like the name of that pub displaced a bit so it doesn't
> obscure the name of the nearby church."
> Even today we - occasionally, thank god - have people who actually
> delete things from OSM because they don't want them on "their" map.
> We have to educate them about how the data is the same for all, but
> the map need not be.
>> I really do hope OSM finds its way through this quagmire of
>> identity and
>> eventually becomes the world's map, widely used, integrated, and
>> in all kinds of spheres.
> No. There is no "the world's map". Everyone wants a different map.
> And that's why we must *not* fall into the trap of trying to provide
> the right map for everyone. The great thing about us is that, given
> the right tools, people can *make* the right map for themselves.
> This is too difficult now, but it can become easier. Take our data,
> take your ideas, and make your map - not "take our map".
> Frederik Ramm ## eMail frederik at remote.org ## N49°00'09"
> talk mailing list
> talk at openstreetmap.org
More information about the talk