Chris,
my problems were always in the other
direction -- too much imagery, not too little.
How much of that too much data was relevant and not obsolete?
And how much of that data was suitable for processing without doctoring ?
OAM was started exactly to be one monolithic mosaic, as an
'open' alternative to Google Maps satellite imagery
I believe that with some creative thinking it is possible to put together a
monolitic mosaic and still keep individual tiled layers separate. But that
should be an issue left for later discussions.
In my opinion the main issue is to define what are the minumum requirements
to bring the OAM project to the level of "an
'open' alternative to Google Maps satellite imagery".
Let's make a list and see how individual issues could be addressed.
I believe tile serving is the smallest of the challenges: if data is
available for serving a server will miracously become available "somewhere"
and if that is not enough one more server will follow.
Iin my mind the main challenge of the project is data in all its aspects:
- how is data made available to the project?
- how it is screened in terms of licencing?
- who will preprocess/verify data?
- how/who will be responsible for keeping some sort of metadata records?
- how/who will tile the imagery?
Chris, you more than anyone have the experience to define all the issues of
the project.
Please spell them out and let's see what the feedback is.
Everyone else: please try to answer the big question WHAT FOR?
To create "an 'open' alternative to Google Maps satellite imagery" ?
Are you convinced it could be done?
I am positive it could ... as long as there is a continued committment by
individuals with skills.
Ciao
Maurizio
----- Original Message -----
From: "Christopher Schmidt" <[email protected]>
To: "Charles R. Schmidt" <[email protected]>
Cc: <[email protected]>; <[email protected]>
Sent: Wednesday, August 12, 2009 1:30 AM
Subject: Re: [OAM-talk] Open Aerial Map Restart Meetup
On Tue, Aug 11, 2009 at 04:20:10PM -0700, Charles R. Schmidt wrote:
Hi All,
While I agree that hardware is a significant challenge I do not think
it's
the only challenge. A distributed approach can greatly alleviate many of
the burdens of central hosting. This would also be more in-line with OGC
style web-services where OAM can act more as a registry, processing
service,
and/or distributed cache.
Another thing to keep in mind that I feel will be a much greater
challenge
is interface
design. OAM was never meant to serve one monolithic mosaic. In the
initial inception OAM was to be much more OSM/Wiki like, allowing
users to rank and tag imagery as it was added to the archive.
Not really. OAM was started exactly to be one monolithic mosaic, as an
'open' alternative to Google Maps satellite imagery, nothing more.
(The fact that the project grew to have so many people thinking that it
was 'meant to' do so many things is evidence of the people in terested
havd a wide variety of varied interests, not that the project itself,
insofar as it exists as a seperate entity, had those goals at its
inception.)
Something like a dynamic mosaic were the imagery severed is a function
of user preferences (acquisition date, spatial
resolution, clouds, event tags, etc, etc). There is little reason these
dynamic mosaics cannot be compiled on the fly with some well designed
data-structures and a maybe a little client side (javascript)
processing.
This statement speaks of a complete lack of understanding of the actual
effort required to do such a thing. Distributing a global mosaic -- though
in and of itself a formidible task -- pales in comparison to creating on
the fly mosaics catered to user preferences. The difference is similar
to the difference in serving up a WMS with reasonable speed to serving
up a cached tileset with reasonable speed -- and for web mapping, you can
see which has had more popularity as web mapping has become more popular.
The goal is admirable, but is not a 'first step' towards making aerial
imagery more widely and openly available.
-
Another important design goal was to maintain the original
imagery. As this project was born in an academic setting we'd like to
maintain its usefulness to academia. Both in terms of an outlet for
institutions to share their imagery, but also as a source of data for
research projects.
A topic raised at the recent experiments was that of licensing.
There are many cases were US Government
agencies (and probably others) are able to share their imagery with the
Disaster Management Community, NGOs, etc, but not everyone.
Are there? I've not run into anyone claiming that is the case for their
imagery: as I've said repeatedly, my problems were always in the other
direction -- too much imagery, not too little.
An OAM
framework could potentially serve as a bridge between the Govt. data
silos
and these communities. However, the framework needs to be flexible enough
to
support limited distribution imagery without mucking up the licensing
considerations for others.
In general, I think that this problem is not one that the OAM community
needs to have as its primary problem to solve. Then again, I stil can't
see that OAM is going to be much more than a collection of like-minded
people
maintaining a registry, if it's going to succeed, so perhaps my vision is
simply too narrow.
Sincerely,
"Charles" Schmidt.
On Tue, Aug 11, 2009 at 10:10 AM, Richard Weait <[email protected]>
wrote:
> On Tue, Aug 11, 2009 at 11:44 AM, David William
> Bitner<[email protected]> wrote:
> > Greetings all,
>
> [ ... ]
> > It appears as though there may be
> > an opportunity to provide travel and accommodations support if we
> > were to
> > have a desire/need to have a face-to-face meetup to see what we can
> > do to
> > bring OAM back online in a sustainable (human and technology)
> > fashion.
> If
> > we were to seek this funding, I would want to be able to go to them
> > with
> a
> > very clear plan of what we were trying to accomplish.
>
> The discussion at State of the Map was interesting and important. The
> in-person discussions you had at NDU appear to have been inspirational
> as well. Have we enough interest here and now to just do it? Or do we
> need to put more heads together and get further inspired?
>
> So what items would we put on the agenda for this meeting? Can we
> tackle these items in email-space?
>
> Agenda
> =====
> 1. discover current hosting and hardware details.
> 2. establish future hosting and hardware requirements and wish list.
> ...
> n. meet in person, raise toasts.
>
> Please add to the agenda. The agenda should tell us if we need to
> meet in person.
>
> > In addition, if there
> > were any "players" in the world that we might want to get involved,
> > this
> > could be a good opportunity to bring them in.
>
> I wonder if this is easier if we can point them to the working,
> relaunched
> site?
>
> Best regards,
> Richard
>
> _______________________________________________
> talk mailing list
> [email protected]
> http://openaerialmap.org/mailman/listinfo/talk_openaerialmap.org
>
_______________________________________________
talk mailing list
[email protected]
http://openaerialmap.org/mailman/listinfo/talk_openaerialmap.org
--
Christopher Schmidt
MetaCarta
_______________________________________________
talk mailing list
[email protected]
http://openaerialmap.org/mailman/listinfo/talk_openaerialmap.org
_______________________________________________
talk mailing list
[email protected]
http://openaerialmap.org/mailman/listinfo/talk_openaerialmap.org
