Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-16 Thread John Smith
On 17 June 2010 14:38, Ben Welsh  wrote:
> Thomas, if I understand you right, you are asking about the mapping tiles,
> correct? All of our mapping tiles are drawn from Google. Though we're using
> OpenLayers, rather than the Google API, most of the time to pull them in. In
> the future, I would love to make custom tiles with Mapnik and Cascadenik,
> but I haven't found the time. Burning our hood boundaries into the map is
> almost too much fun to pass up.

His concern is with the ambiguity of Google T&Cs about deriving things
from their map, however Ed Parsons clarified this a little bit when he
stated you could publicly distribute your favourite hiking trail and
so on, but not vectorise every street, it seems to me boundaries
wouldn't really be the same as vectorising every street.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-16 Thread Ben Welsh
Apollinaris, our boundaries have a classification system we roughed out. How
well they would mesh with OSM is something I'd love to hear a critique on.
Simplifying things a bit, the taxonomy of our database is two tiers:
Neighborhoods and Regions. Neighborhoods come in three types: 1, parts
of a city; 2, complete cities; 3, unincorporated areas. And each
neighborhood (i.e. Santa Monica) belongs to a larger region (i.e. The
Westside).
The regions are, by design, untied from any municipal boundaries, since the
general sense in LA is that many of the commonly understood
regions are broken up into several cities. A great example is the San
Fernando Valley, which has a small island in the middle, namely the city of
San Fernando.
The strangeness of the municipal boundaries is actually one of the main
reasons we wanted to do this.
*
*
Thomas, if I understand you right, you are asking about the mapping tiles,
correct? All of our mapping tiles are drawn from Google. Though we're using
OpenLayers, rather than the Google API, most of the time to pull them in. In
the future, I would love to make custom tiles with Mapnik and Cascadenik,
but I haven't found the time. Burning our hood boundaries into the map is
almost too much fun to pass up.



On Wed, Jun 16, 2010 at 5:13 PM, Thomas Ineichen wrote:

> Hi Ben,
>
> I'm just wondering: are the suggestions/improvements by your readers based
> on anything else than Google Maps?[1]
>
> Regards,
> Thomas
>
> [1] e.g.
> http://projects.latimes.com/mapping-la/debates/westside/#comment-form
>
>
> ___
> talk mailing list
> talk@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk
>



-- 
palewire.com
work: 213-473-2624
cell: 213-254-5570
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-16 Thread Thomas Ineichen
Hi Ben,

I'm just wondering: are the suggestions/improvements by your readers based
on anything else than Google Maps?[1]

Regards,
Thomas

[1] e.g.
http://projects.latimes.com/mapping-la/debates/westside/#comment-form


___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-16 Thread Apollinaris Schoell
I think it's a good idea but needs a good idea for the tagging with these 
different combinations and dividing.
neighborhood names are common in other cities too and well known to locals. So 
it is valuable info for osm and should be rendered too. currently some are 
added as place nodes and also rendered as such. having them as an area is even 
better.
sure there will be debates about exact boundary but over time either osm 
converges to the locally used ones or osm will tell people where they are and 
they may get used to follow osm




On 16 Jun 2010, at 6:13 , Ben Welsh wrote:

> At the risk of over complicating things, let me give a little more info. 
> 
> LA County is a fragmented place with many different cities and unincorporated 
> areas puzzled together. Our "neighborhoods" are in fact three different types 
> of areas consolidated.
> 
> 1. Cities divided into neighborhoods. i.e. 
> http://projects.latimes.com/mapping-la/neighborhoods/city/los-angeles/
> 2. Complete cities, drawn by their formal boundaries. i.e. 
> http://projects.latimes.com/mapping-la/neighborhoods/neighborhood/west-hollywood/
> 3. Unincorporated areas that are "Census Defined Places": 
> http://projects.latimes.com/mapping-la/neighborhoods/neighborhood/east-los-angeles/
> 
> On top of that, there are dozens of small unincorporated areas that are 
> basically islands floating between everything else. We've lumped them in with 
> a bordering neighborhood: 
> http://projects.latimes.com/mapping-la/neighborhoods/unincorporated/list/page/1/
> 
> Why did we throw all these together and call them neighborhoods? Because our 
> goal is to have a single common denominator we can spread across the entire 
> county and use for comparison. That's why we build them out of Census tracts, 
> so we could rack up demographics about them all. i.e.: 
> http://projects.latimes.com/mapping-la/neighborhoods/income/median/neighborhood/list/
> 
> As time goes on, we plan to divide up all of the cities into smaller 
> neighborhoods, not just Los Angeles, we did in a first round last year. In 
> cases where cities have official hood boundaries (LA does not) we'll likely 
> use those. 
> 
> More info about the project and process is here: 
> http://projects.latimes.com/mapping-la/neighborhoods/about/
> 
> On Wed, Jun 16, 2010 at 5:23 AM, John F. Eldredge  wrote:
> This sounds like a good compromise to me, as most people will have a general 
> agreement of where a given neighborhood is located, but differ about where 
> the boundaries are located.
> 
> --
> John F. Eldredge -- j...@jfeldredge.com
> "Reserve your right to think, for even to think wrongly is better than not to 
> think at all." -- Hypatia of Alexandria
> 
> -Original Message-
> From: Ed Avis 
> Date: Wed, 16 Jun 2010 08:46:09
> To: 
> Subject: Re: [OSM-talk]
>Q: Is OSM interested in neighborhood and r
>egional boundaries for L.A.?
> 
> A compromose would be to add the centre of each neighbourhood (as 
> locality=place
> or similar) but not the exact boundaries.
> 
> --
> Ed Avis 
> 
> 
> ___
> talk mailing list
> talk@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk
> 
> ___
> talk mailing list
> talk@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk
> 
> 
> 
> -- 
> palewire.com
> work: 213-473-2624 
> cell: 213-254-5570
> ___
> talk mailing list
> talk@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk


___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-16 Thread Richard Weait
On Wed, Jun 16, 2010 at 9:13 AM, Ben Welsh  wrote:
> At the risk of over complicating things, let me give a little more info.
> LA County is a fragmented place with many different cities and
> unincorporated areas puzzled together. Our "neighborhoods" are in fact three
> different types of areas consolidated.
[ ... ]

Dear Ben,

It must have been great fun to participate in this project.  I see
that you and the Los Angeles Times understand the problems related to
crowd sourcing neighborhood boundaries perfectly.

See "You gotta stop is somewhere"
http://www.latimes.com/includes/projects/img/thumb-westside-300x100.png

Also this neighborhood map for Tarzana is wonderful.
http://projects.latimes.com/mapping-la/neighborhoods/comments/11501/

Your consultation with the community in Los Angeles (650
user-generated maps, 100 revisions) sounds like you have substantial
interest and perhaps even consensus locally.  I think that's
wonderful.  Presuming that the participation in your project is likely
to reduce border disagreements, I think it would be a nice addition to
OSM.

I notice that you publish your data as cc-nc-sa.  To include it in OSM
you would have to agree to allow OSM to publish it as cc-by-sa and
then ODbL after the license upgrade.  Of course you would lose the
explicit Los Angeles Times credit as well since OSM expects a
simplified "Maps and Data CCBYSA OpenStreetMap (and Contributors)"

And again, I think it is important to get feedback from others in the
Los Angeles OSM community.  Have a look over at talk-us.  They might
have something similar in the works.  I'm sure you find the conjecture
by all of us "seagulls" interesting but we all know that one active
local mapper on the ground is better than a self-important expert from
Toronto.  ;-)

Best regards,
Richard

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-16 Thread Ben Welsh
At the risk of over complicating things, let me give a little more info.

LA County is a fragmented place with many different cities and
unincorporated areas puzzled together. Our "neighborhoods" are in fact three
different types of areas consolidated.

1. Cities divided into neighborhoods. i.e.
http://projects.latimes.com/mapping-la/neighborhoods/city/los-angeles/
2. Complete cities, drawn by their formal boundaries. i.e.
http://projects.latimes.com/mapping-la/neighborhoods/neighborhood/west-hollywood/
3. Unincorporated areas that are "Census Defined Places":
http://projects.latimes.com/mapping-la/neighborhoods/neighborhood/east-los-angeles/

On top of that, there are dozens of small unincorporated areas that are
basically islands floating between everything else. We've lumped them in
with a bordering neighborhood:
http://projects.latimes.com/mapping-la/neighborhoods/unincorporated/list/page/1/

Why did we throw all these together and call them neighborhoods? Because our
goal is to have a single common denominator we can spread across the entire
county and use for comparison. That's why we build them out of Census
tracts, so we could rack up demographics about them all. i.e.:
http://projects.latimes.com/mapping-la/neighborhoods/income/median/neighborhood/list/

As time goes on, we plan to divide up all of the cities into smaller
neighborhoods, not just Los Angeles, we did in a first round last year. In
cases where cities have official hood boundaries (LA does not) we'll likely
use those.

More info about the project and process is here:
http://projects.latimes.com/mapping-la/neighborhoods/about/

On Wed, Jun 16, 2010 at 5:23 AM, John F. Eldredge wrote:

> This sounds like a good compromise to me, as most people will have a
> general agreement of where a given neighborhood is located, but differ about
> where the boundaries are located.
>
> --
> John F. Eldredge -- j...@jfeldredge.com
> "Reserve your right to think, for even to think wrongly is better than not
> to think at all." -- Hypatia of Alexandria
>
> -Original Message-
> From: Ed Avis 
> Date: Wed, 16 Jun 2010 08:46:09
> To: 
> Subject: Re: [OSM-talk]
>Q: Is OSM interested in neighborhood and r
>egional boundaries for L.A.?
>
> A compromose would be to add the centre of each neighbourhood (as
> locality=place
> or similar) but not the exact boundaries.
>
> --
> Ed Avis 
>
>
> ___
> talk mailing list
> talk@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk
>
> ___
> talk mailing list
> talk@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk
>



-- 
palewire.com
work: 213-473-2624
cell: 213-254-5570
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-16 Thread John F. Eldredge
This sounds like a good compromise to me, as most people will have a general 
agreement of where a given neighborhood is located, but differ about where the 
boundaries are located.

-- 
John F. Eldredge -- j...@jfeldredge.com
"Reserve your right to think, for even to think wrongly is better than not to 
think at all." -- Hypatia of Alexandria

-Original Message-
From: Ed Avis 
Date: Wed, 16 Jun 2010 08:46:09 
To: 
Subject: Re: [OSM-talk]
Q: Is OSM interested in neighborhood and r
egional boundaries for L.A.?

A compromose would be to add the centre of each neighbourhood (as locality=place
or similar) but not the exact boundaries.

--
Ed Avis 


___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-16 Thread Pieren
We already had a discussion about something smaller than suburbs last year:
http://lists.openstreetmap.org/pipermail/talk/2009-September/041903.html

But I don't know if you consider "quarters" or "districts" differently as
"neighborhoods".

Pieren
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-16 Thread John Smith
On 16 June 2010 19:17, Tom Hughes  wrote:
> Does that matter if the boundaries are essentially guesswork inventions
> anyway?

If we used that logic we would only ever map from very hi-res very
high accurate aerial imagery then because anything less is mostly
guess work...

> It sounds like these aren't any sort of officially defined areas, but more
> the kind of fluid local names for approximate areas.

Actually it's worst than that, at least here, because when you are
near a suburb border different databases can place you in different
suburbs.

I think most databases are generated from extrapolations, but the
original boundaries would have been drawn up on paper, and some times
they do shift but older suburbs tend to be pretty static. Also suburb
boundaries here sometimes have signs up on major roads when you move
between them. That said, suburbs are somewhat different in Australia
to similarly named places in the US, there is more of them and they
cover smaller areas.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-16 Thread Tom Hughes
On 16/06/10 10:04, John Smith wrote:
> On 16 June 2010 18:46, Ed Avis  wrote:
>> A compromose would be to add the centre of each neighbourhood (as 
>> locality=place
>> or similar) but not the exact boundaries.
>
> That doesn't tell you what objects exist inside those boundaries...

Does that matter if the boundaries are essentially guesswork inventions 
anyway?

It sounds like these aren't any sort of officially defined areas, but 
more the kind of fluid local names for approximate areas.

Tom

-- 
Tom Hughes (t...@compton.nu)
http://compton.nu/

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-16 Thread M∡rtin Koppenhoefer
2010/6/16 Steve Bennett :
> OSM does need to think more carefully about what exactly is in and out
> of scope.


The scope is IMHO the worldmap drawn with the knowledge of locals,
that's why I'd consider these informal neighbourhoods precious to our
data, even more as they are not "official" boundaries so OSM could
maybe become the main source for them.

I'm actually against too much discussion about relevancy of things to
be put into the db. It might be problematic to insert ephemeral stuff,
especially if it does not get maintained, but besides this I'd
personally like to see as much information as possible inserted.
Things (e.g. those areas, alternative names, ...) that only the locals
know of (but to them is commonly known), and that is not written in
other publications or even on the ground I'd consider the most
precious data to collect.

cheers,
Martin

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-16 Thread John Smith
On 16 June 2010 18:46, Ed Avis  wrote:
> A compromose would be to add the centre of each neighbourhood (as 
> locality=place
> or similar) but not the exact boundaries.

That doesn't tell you what objects exist inside those boundaries...

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-15 Thread Steve Bennett
On Wed, Jun 16, 2010 at 1:28 PM, Ben Welsh  wrote:
> Long story short: I'm curious whether our boundaries might have a home in
> the OSM database. I don't know a ton about the project, but I've always

IMHO they might be useful, on the basis that they're not just any old
informal boundaries, they have the credibility of a major newspaper
behind them. On the other hand, if they really do just change
arbitrarily, that's less valuable.

OSM does need to think more carefully about what exactly is in and out
of scope. Some people want everything if it's verifiable. Others want
to draw limits. The exact same debate goes on constantly at Wikipedia,
but is much more sophisticated.

At the very least, the names would be useful in nominatim or whatever,
to help find stuff.

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-15 Thread John Smith
On 16 June 2010 13:44, Richard Weait  wrote:
> As a visitor to LA, or viewer of LA on OSM, I think it would be
> interesting to see the neighbourhood names.  It sounds useful to
> visitors and locals.

Not only that, but such boundaries are useful to save adding is_in=*
tags to everything...

> As an OSM contributor, I'd hate to see the contribution of informal
> areas become contentious or a focus for an edit war.

These shouldn't really become the focus of an edit war, but they can
be painful to deal with when they follow similar paths to roads and
other similar man mad features, but this is a bigger issue and as long
as they are tagged properly at least they can be hidden in JOSM these
days...

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-15 Thread Richard Weait
On Tue, Jun 15, 2010 at 11:28 PM, Ben Welsh  wrote:
> Hello listers,
> I'm a developer at the Los Angeles Times. We just put out a set of
> boundaries for 272 neighborhoods and 16 regions that cover Los Angeles
> County. http://projects.latimes.com/mapping-la/neighborhoods/
> The idea is to draw formal lines that try to capture informal areas commonly
> used by locals. It's an art, not a science, but we're trying to have fun
> with it. And have invited users into some OSM type debates along the way.
> See: http://projects.latimes.com/mapping-la/debates/westside/
> Long story short: I'm curious whether our boundaries might have a home in
> the OSM database. I don't know a ton about the project, but I've always
> admired it from a distance, and I would love to get our small development
> team, which does a fair amount of mapping, somehow involved with the
> community.
> I hope this isn't interpreted as spam. I don't mean any disrespect. I'm just
> honestly curious what y'all think and this seemed like the place to drop a
> line.

Hi Ben,

Welcome to OpenStreetMap.  Thanks for the cool article about OSM
today.  Sure, it was under Mike Swift's name, but you showed up here
so you get the thanks.

Your question is interesting.  I'm not going to answer it.  ;-)

As a visitor to LA, or viewer of LA on OSM, I think it would be
interesting to see the neighbourhood names.  It sounds useful to
visitors and locals.

As an OSM contributor, I'd hate to see the contribution of informal
areas become contentious or a focus for an edit war.

Perhaps some of our LA locals will join the discussion.

Best regards,
Richard

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


[OSM-talk] Q: Is OSM interested in neighborhood and regional boundaries for L.A.?

2010-06-15 Thread Ben Welsh
Hello listers,

I'm a developer at the Los Angeles Times. We just put out a set of
boundaries for 272 neighborhoods and 16 regions that cover Los Angeles
County. http://projects.latimes.com/mapping-la/neighborhoods/

The idea is to draw formal lines that try to capture informal areas commonly
used by locals. It's an art, not a science, but we're trying to have fun
with it. And have invited users into some OSM type debates along the way.
See: http://projects.latimes.com/mapping-la/debates/westside/

Long story short: I'm curious whether our boundaries might have a home in
the OSM database. I don't know a ton about the project, but I've always
admired it from a distance, and I would love to get our small development
team, which does a fair amount of mapping, somehow involved with the
community.

I hope this isn't interpreted as spam. I don't mean any disrespect. I'm just
honestly curious what y'all think and this seemed like the place to drop a
line.

Thank you,

Sincerely,

Ben Welsh.

P.S. We already distribute all of the shapes in KML and GeoJSON. See:
http://projects.latimes.com/mapping-la/api/
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk