The core requirements of this process

[Jeff Yutzler]

(Take 2 of starting a thread with a different subject)

 On Mon, Mar 19, 2012 at 5:24 PM, Paul Ramsey <pra...@opengeo.org> wrote:

- Is a core requirement of this process that the final solution is

signed off on by a standards body? Or just that the final solution
works?

 

Good question.  Bear with me, but there are actually six parts of the process from the Army's perspective (not necessarily in chronological or priority order).

1. Choose a technology solution

2. Develop standard(s) 

3. Develop conformance tests

4. Develop a reference implementation

5. Produce guidance (best practices)

6. Perform an interoperability experiment

The standards piece is annoying, but it is vital to the Army for governance reasons.  Saying "just use product XXX" isn't going to cut it when the Lockheed Martins and Northrop Grummans of the world are trying to land 8, 9, or 10 figure contracts to build Army Programs of Record.  It is too easy to do the wrong thing.  Having a standard makes it possible to mandate doing the right thing and having conformance tests allows verification that the right thing was done.  Paul D. is funded by AGC to develop the standard, hopefully with assistance or at least buy-in from the other participants.  

-Jeff

--
Jeff Yutzler
Image Matters LLC
Mobile: (703) 981-8753
Office: (703) 428-6731

[Scott C]

As for the vision and ultimate goals of the project, we want both: 

1) A final solution that works for developers and users

2) A standard in OGC

The strategy is to start with a willing coalition of developers to create the solution and build up community support so that it's vetted well before going before a standards body.  I'm going to go ahead and beat this drum again:  The Army wants this to be an industry standard because it won't work otherwise.  The need has been voiced by NGA and Army because of the data distribution and operational challenges they face.  The stakeholders are every single entity that would provide data to, receive data from, or implement solutions for them.  In my simplistic way of looking at this, I would even go so far as to say we want something as ubiquitous as a shapefile (in terms of software support), that's more robust, simpler to exchange because it's self-contained, and targeted for supporting smartphones and tablets running iOS and Android.

We have to generate community and vendor buy-in.  We have to do this in such a way it's relatively easy to pick up and implement as a new format in existing software.  There will be a session on this at FOSS4G-NA to brief this effort to the community and try to generate more community interest. 

So there is urgency felt because the Army and NGA are facing this problem now.  However, the understanding is that this has to be done right.

[Brad Hards]

On Tuesday 20 March 2012 13:55:08 Scott C wrote:
> We have to generate community and vendor buy-in. We have to do this in
> such a way it's relatively easy to pick up and implement as a new format
> in existing software.

"Here is a standard schema, here is the library that supports it, no charge to
develop with, no charge to use, pretty much already ported to any platform you
are about." I'm not sure how developing another product (with no community
experiences, no support in existing tools, etc) will be better unless you're a
standards developer or big player trying for a monopoly position.

You could choose to document sqlite and spatialite, but what would that really
achieve?

> There will be a session on this at FOSS4G-NA to
> brief this effort to the community and try to generate more community
> interest

Not being North American, I won't be there. I will (probably) be in D.C. this
weekend and early next week though, and may have some time if anyone wants to
face-to-face. Contact me off list if you want to pursue this option.

Note that I am (and will be) doing all of this on my own time - not
representing my employer or its client(s). There isn't anything in this for me
personally other than the satisfaction of a good outcome for my mates.

Brad