I couldn't agree more with what Matt said. Perhaps it's incumbent upon us to
provide some examples/whitepapers/proof/etc of how the VA will benefit from
this proposed relationship. It's very likely that they do not understand that
the virtuous spiral which worked so well inside the VA to develop DHCP/VISTA
can be extended to include a larger domain and all will benefit.
----- Original Message -----
From: Matthew King
To: open-ehealth-collaborat...@googlegroups.com
Cc: Hardhats ; openhealth@yahoogroups.com
Sent: Friday, July 10, 2009 12:42 PM
Subject: Re: Open VistA Community Proposal
Awesome Fred!
I have a list of things, but Nancy will have a better one, so I will defer to
her. I also think Nancy would make a great point person on our side of the
arrangement. On the other hand, a small list of key actionable items is
preferable to an exhaustive laundry list.
IMHO, the most important piece by far is the recognition inside the VA that
partnering with the outside community is a real benefit to their mission, which
is ONLY to the veterans. Until that is internalized with some document, it will
be hard for the VA to have a consistent interface with the community and any
gains can be lost with a simple change in personnel at any one of many levels.
By establishing that benefit in some "official" way, MOUs and other agreements
with be much easier and gains will be easier to sustain.
We desperately need the complete VEHU site and the ICD lexicon
(OK, I couldn't resist putting at least a couple of things in.)
matt
On Fri, Jul 10, 2009 at 1:29 AM, fred trotter <fred.trot...@gmail.com> wrote:
Hi,
At the CONNECT conference, Brian Behlendorf, Kolodner and several other
VA employees who might wish not to be named helped me get a 2 minute audience
with the CTO of the VA Peter Levin. We exchanged information and agreed to
email. The very fact that there is a CTO of all of the VA, rather than just
seperate departments is a huge development, this is a tacit acknowledgement
that the VA is a technology creating organization (what a CTO does) rather than
a technology managing organization (what a CIO does).
A few days later I sent him a typical (for me) rant, with about ten
links to the various things I have written about the VA and VistA. I told him
that I though the attempt to proprietary portions of the code was a bad idea
and that the centralization of the development of VistA (as opposed to other VA
IT functions) was a bad idea. If any of you have read my blog you can have a
pretty good idea of what I put in the email. I basically complained about every
failing of the VA that I have heard at the WorldVistA or other FOSS
conferences. I tried to have a more positive tone than my standard
over-inflammatory style at the prompting of Nancy (thanks for that) and other
more calming voices.
The response that I got back was amazing. He essentially said that he
agreed with many of my points, and even sent me some of his own writing that
correlates with some of the ideas in my letter. His one criticism was that I
was not -asking- for something that he could act on. He specifically asked for
a shorter actionable proposal to fix the problems between the VistA community
outside the central VA and the VistA community inside the central VA.
So what do we want from the VA? I can think of several very specific
things that I might include in a response. I will throw them out here for
community comment and then send a letter based on the consensus (if there is
one) on the issue. Here is what I would like to see the VA do.
a.. Create a bridge-person: Create a role to interface between
VistA-inside and VistA-outside. Fill that role with someone who is capable of
speaking VistA and open source. Someone like Brian Behlendorf, a federal
employee who serves as the community manager for the CONNECT project. That
person would be expected to go to WorldVistA conferences etc etc and provide a
face for this collaboration.
b.. Overturn the moratorium of local VA hospital VistA development.
c.. Reinvest in local VA hospital VistA instances. Centrally managed
instances of VistA, with locally deployment. Flawed VistA modules from one
hospital should not take down the VistA instance of another hospital.
d.. Empower the bridge-person with a VistA Community Portal. That portal
should provide the following services:
a.. Allow for the submission of improved VistA components back into the
VA, to be evaluated as Class III code for possible adoption by local VA
hospitals.
b.. Those submissions should always be public unless they are security
issues, and then they should be made public immediately after being
confirmed-patched/denied-ignored
c.. Publish a list of approved licenses for contributing VistA
components back (probably from proprietary friendly licenses like Apache,
Mozilla, BSD, EPL etc etc, or just chose one if that is easier).
d.. Organizations that submit patches, or improvements should expect
that the bridge-person will publicly comment on reasons for rejection for a
particular patch or software, fi the VA will not adopt the software.
e.. Have a feature request system, that is accessible only to groups
who are or represent live VistA instances outside the VA. This should include
local VA hospital programmers and CACS, people from IHS, representatives from
foreign organizations like Mexico and Jordan, and private hospitals running
VistA. This should provide a means for the community to give feedback to the VA
about the consequences of central VA development decisions. However, this would
not put the VA in the position of accepting feature requests from people who
merely 'might' use VistA.
e.. The features and contributions should be analysed against the current
VA 'modernization' plan to create a new modernization plan that at least
partially considers the needs and contributions of outside-VA VistA users.
Note that this plan is far more conservative than what I have advocated in
the past.
http://www.fredtrotter.com/2008/04/18/what-do-about-the-va-crisis-the-aboveground-railroad/
I would call this a light-weight "open VistA community process". It does
not presume that the central VA should give up any control that they currently
hold. Rather, it just creates a formal means for the VA to be open about its
development decisions regarding VistA.
But I do not want to put something like this forward without taking
comments from the community.
Rather than have this debate on just Hardhats, I want to open it up to
everyone.
You can now comment on proposal using a co-ment instance (co-ment is the
successor to stet which was used to take comments on the GPlv3) that is
available through http://LibertyHSF.org
Regards,
-FT
--
Fred Trotter
http://www.fredtrotter.com
[Non-text portions of this message have been removed]
