Future (fyū’chər) n.
- The indefinite time yet to come
- Something that will happen in time to come
- A prospective or expected condition, especially one considered with regard to growth, advancement, or development
I appreciate Fred’s comments regarding my recent posting. Fred appropriately challenged me for describing some of the problems but not more clearly articulating a solution along the lines of which he did in this April 2008 posting.
Fair enough, Fred. So here goes . . .
I would recommend that VistA create a truly open source governance council along the lines of the Apache or Mozilla Foundation. Oops! . . . I already proposed this back in December 2003. Working with two professors who studied open source for a living, I clearly articulated the concept of a council, with seven members of self-interested parties who each have a stake in the success of VistA. We described the purpose, the process, and how this organization would be funded – all based off of other successful projects. The reason that this never took off, I believe, was that there was never a benevolent dictator (executive director) who stepped up to herd the cats, lasso the politics, and “wrassle” with all the self-interested parties. It clearly is a very tough job.
Not to say that it was not attempted. WorldVistA, Medsphere, DSS, and several also-rans were launched at the same time. We actually created a VistA Software Alliance in sheer frustration when the natural leadership role that should have been provided by WorldVistA proved a study in frustration from an underfunded, understaffed, and under-visioned leadership team that could not garner the confidence of the business community. To WorldVista’s credit, however, they have hung in there and accomplished some fairly impressive milestones, including the ongoing meetings (just finished the 17th Conference) and most importantly helping VistA EHR achieve CCHIT status. Perhaps, now is the time that they can reassert themselves and rally the community around their leadership.
One thing is clear, however, that the VA will be exceptionally wary of going along in the way that Fred has proposed. There is no way that they are going to give up control of the development direction of VistA as it relates to their internal organizational objectives. Why would they? Furthermore, why should they? They are a sovereign organization that needs to maintain absolute control over their software. Also, I don’t believe a congressionally mandated council would be in the best interest operationally, politically, or efficiency-wise (can you point to a single congressionally mandated open source project?). Finally, I actually buy into the argument of “why is our government developing software in the first place”? The fact that the VA’s accomplishments are such a rare anomoly, appears to be another argument in favor of transitioning their “against all odds success” to more likely to be sustainable collaborative development process from which they would participate, not take direction.
This notion, I believe, is a much more effective and perhaps realistic approach. It allows the VA to transition from the original author of the software to an ongoing contributor to the story line (see my Act III comments). The point that the VA has never understood, from their former leadership on down the bureaucratic chain, is that the VistA movement has transcended the VA. Ever since they have been releasing versions of the software (the latest FOIA version released just this month), they were really transferring their legacy and the long term sustainability of the software to a much larger universe of people who began adopting it in the private sector. While these early adopters (Oklahoma Department of Veteran Affairs, Midland Memorial Hospital, West Virginia Department of Health, Texas Tech Health Systems, etc) have represented a huge infusion of new blood to the software, the next set of implementation will help transform VistA to a truly international EHR capable of powering entire countries (ala, the Hashemite Kingdom of Jordan).
This is really amazing to watch, as you consider it, that while the VA and DoD fumble around with the greatest information system in the world (including wondering if they should even continue to develop it), the rest of the world is going to be adopting it, improving it, collaborating around it, and creating a global VistA based information sharing network.
As I have already mentioned in my Diabetic VistA post, there are certainly parts of the system that need radical updates, enhancements, and a freaking graphical user interface for heavens sake. However, it is the underlying database, the degree of integration, and the clinical logic that has been imbedded in the system that makes it so powerful in actually improving care to a degree that is still unrivaled within the private sector at the scale of the VA (although Kaiser is fast approaching with their $5 billion dollar, MUMPS-based Epic Systems implementation – which again highlights the value of VistA given its relative price). So VistA should be considered a natural foundation from which to build, from which to improve, and from which to create a sustainable, ongoing stream of innovation. Why do we feel the need to continually re-re-re-invent the wheel? Why not pool all the intellectual, financial, and clinical capital to build around a common framework in a collective fashion through collaborative development?
In considering this possibility, I am not talking about building a one size fits all solution. What I am talking about is leveraging the beauty of the open source development process wherein you leverage the self-interested needs of each member in order to entice them to contribute. What if the entire VA software developers were directed to continue to work on their assigned projects, but allowed in Google like fashion to have some of their time working on VistA based open source projects like Pharmacy GUI, Laboratory upgrades, interoperability with the DoD. What if all the private sector hospitals who have deployed the system had a more effective mechanism to share knowledge, tips and tricks, templates, and process methodologies in a fashion where the experience of the one could benefit the many?
This is ALL possible, through the collaborative power of Open Source. It is ALL possible without detracting others from their core, self-interested focus. It is ALL possible and has been DONE repeatedly by other successful projects. An Open Source VistA project, complete with a project plan, has already been developed and on the web for further refinement, suggestions, and improvements since 2003.
So Fred, the only thing that I see as missing is the sustained leadership required to lead such an effort. I believe there needs to be a Linus-like VistA guru, a Mitchell Baker equivalent who can parse through the politics, or even a gravitational core like the benign brotherhood of the Apache Foundation itself. In the search for the White Knight, one thing is clear, it should be community not congressionally mandated leadership that is allowed to reign.
2 comments on “The Future of VistA: Community versus Congressionally Mandated Leadership”
Excellent points. In fact the only counter that I have to your idea, and the only reason that I did not suggest the same thing, is that Congress continually does things or allows things that really threaten the VistA lifecycle.
But you either have to solve one of two problems. You either have to get some FOSS organization be trusted MORE than the VA itself regarding VistA. OR you have to accomplish what I am suggested, embedding a Mozilla foundation directly within the VA itself.
Frankly, either one is pretty unlikely, which is why fear for the future of VistA. The nice thing about your plan, is that if the VA falls flat on its face, is that an “Apache Foundation for VistA” is much more likely to arise under those circumstances.