On Thu, Mar 17, 2011 at 05:18:25PM -0500, Mark Korver wrote:
> I've been checking out the progress with the Imagery Index.  It's
> lookin good. So have been thinking about "Access Tools" abit now that
> Chris has the index working.
> 
> Assuming the data is Processed image data.
> I have some area of interest, so I use a bbox and some other filters
> to get back a list of urls, having those urls I download the data to
> my application.
> Once I have the data local, the tool auto-builds index and mapfile (
> MapServer ) allowing me to see it as a wms.
> Once I can view it as a wms ( assuming overviews are in the tifs here
> ) I can check it and run code to TMS it.
> Once I have the TMS ( like on S3 ) I can delete the source tifs.
> 
> I would imagine all of this working off of some app that you just
> started from a public EC2 instance that would start you off with OL
> showing you the result of your query to the Imagery Index  ( Chris
> already showed this to us ), then allow you to run the process kind've
> like TileMill.
> 
> This shouldn't be so hard to do with one application instance, the
> trick would be to do it with 100 to speed things up.
> 
> Is this the kind've thing people have in mind for Access Tools?

Yep, that's exactly one sort of access tool. Of course, another is
the simple demo that I built -- using the API + OpenLayers to display
outlines of the layers, for example. Another potential access tool
is one that allows you to download a set of data, and deploy it on a
machine -- perhaps in response to a humanitarian disaster, for example --
and ship that machine off to some other location and use all the
data.

Basically, 'access tools' is code for "Anything people want to do with
OAM" -- and our goal is to make that as easy as possible. So, for
example, one of the things I wnat to do that I didn't think of was
make a seperate overview of "This is the imagery this server is hosting",
built from the index... but I can't stash additional data about layers
in the index (like a WMS URL for a layer). So that requires extending
the imagery index -- I'm thinking with tags on layers for storage 
of additional key/value data that can be consumed by layers.

You can imagine building other types of tools: Evaluation of cloud
cover, landcover changes using dated images and diffing them, etc. 
These are all a bit more advanced: I think that the thing you described
is exactly the kind of thing we're hoping people will be able to
do relatively easily, and is a solid first target for access tools.

Regards,
-- 
Christopher Schmidt
Web Developer

_______________________________________________
talk mailing list
t...@host134.hostmonster.com
http://host134.hostmonster.com/mailman/listinfo/talk_openaerialmap.org

Reply via email to