4 ms·

Open source, not $19 billion, may be best health care stimulus

11 points by edw519 17y ago

bavcyc 17y ago
$125 million to translate from one computer language to another. I'm underpaid.
- algorias 17y ago
  That's more than a thousand man-years. Absurd.
- yellowbkpk 17y ago
  The general rule with medical software is to estimate the cost in money and time and multiply it by 100.
bebop 17y ago
VistA is an extremely complex piece of software written in a time that readability was not a top priority. Not to mention that there were file size limits, which spread modules over many different files, and the fact that most of the MUMPS keywords can, and are abbreviated adds to challenge of understanding and translating. On top of the often hard to read modules, most of the real work is stored in "Globals" basically persistent sparse arrays. So if you really want to trace the code you have to spend half your time reading code and half your time looking through the "Database" for the rest of it. I guess what I am trying to say is its a really hard job to bring VistA to a different platform without a large sum of money and some really smart people.
blhack 17y ago
The fact that "stimulus" has entered your standard issue citizen's lexicon as something that the government gives to a failing industry to keep it from failing is one of the most horrifying things to come out of the 2008 election cycle. WHY is this looked at as okay? I get that the headline was a joke...but why is it even funny? I'm sorry, I know this is off-topic (and more appropriate for a political forum like reddit), but I'm 22 years old. Fuck you, to the government that is taking away money that I haven't even had the chance to EARN yet so that companies that I have no interest in can continue existing. How did we get to a point where this is looked at as acceptable? Won't somebody please actually think of the children? You, government, the very people who are supposed to be looking out for me, are robbing from me in broad daylight, then having the audacity to tell me that it is for my own good?! What the hell is happening?!!?
- RyanMcGreal 17y ago
  When a large finance company goes down, an awful lot of innocent people are dragged down with it. When enough large finance companies go down at the same time, the entire economy goes down with them, leading to massive suffering. The bailout sucked. Not doing the bailout would have sucked far more.
  - blhack 17y ago
    I agree with you, which is why I was 100% behind the troubled asset relief package. That was to help out the banks, which needed to remain liquid. I was extremely extremely vocally for TARP... This is something else entirely.
sachinag 17y ago
For fuck's sake, the code is out there: http://sourceforge.net/projects/openvista/ http://sourceforge.net/projects/openvista/ EDIT: apparently the main project is at WorldVista now: http://sourceforge.net/projects/worldvista/ http://sourceforge.net/projects/worldvista/ There are for-profit companies selling install/support services around it: http://www.medsphere.com/ http://www.medsphere.com/ (And here's their .org for their OpenVista repository: http://medsphere.org/index.jspa http://medsphere.org/index.jspa) Yeah, VistA is a PITA. What we need are data standards for non-imaging data (where DICOM has been a raging success). I've long suggested that all records have to be readable in VistA - it doesn't force people to use Medsphere, but it means that there's a lowest common denominator FREE software package that works.
- siculars 17y ago
  agreed. i'll just paste my comment from an article on the oreilly radar blog, http://radar.oreilly.com/2009/05/who-will-cut-the-gordian-knot.html http://radar.oreilly.com/2009/05/who-will-cut-the-gordian-kn... . In my opinion, one thing that government can do to accelerate electronic record keeping in the healthcare arena is mandating that solution providers must make means available to both import and export data to/from their systems. Mandating a specific format is less of an issue and should not be the main goal. If vendors were obligated to open their systems and encouraged to do so via some [fiscal/tax] incentive then the problem would solve itself. IT professionals either on their own or contracted would create solutions that connected the hodgepodge of systems out there. These systems are infinitely difficult to implement on any scale (by scale I mean number of people using the system - not technical challenges). At the moment I am tangentially involved in a major enterprise installation at a very well known institution. From my vantage point one of the major stumbling blocks is the difficulty in which the many, many systems already in place will interact with the new system (and with each other). The inability of these systems to interact with each other directly affects the cost of doing business in the modern day hospital. Many of the people in positions to make purchasing decisions are not fully cognizant of the concept of "vendor lock in" and "data portability". These poor decisions are paid back many fold not only in cold hard cash but also in opportunity cost over the lifetime of the system purchased, which may very well be decades. Government can make this all go away by simply requesting that all players in this space provide mechanisms for programmatic, unfettered data access. Providing fiscal incentive will all but ensure that those providers in compliance will be preferred. Mandating a specific system, or data exchange format is not the solution or, frankly, the "American way". Evening the playing field by ensuring openness through financial incentive is much more palatable to a broader spectrum of participants. The market place will ultimately decide and coalesce around its own standards and formats.

11 points by edw519 17y ago

bavcyc 17y ago

$125 million to translate from one computer language to another. I'm underpaid.

algorias 17y ago
That's more than a thousand man-years. Absurd.
yellowbkpk 17y ago
The general rule with medical software is to estimate the cost in money and time and multiply it by 100.

bebop 17y ago

VistA is an extremely complex piece of software written in a time that readability was not a top priority. Not to mention that there were file size limits, which spread modules over many different files, and the fact that most of the MUMPS keywords can, and are abbreviated adds to challenge of understanding and translating. On top of the often hard to read modules, most of the real work is stored in "Globals" basically persistent sparse arrays. So if you really want to trace the code you have to spend half your time reading code and half your time looking through the "Database" for the rest of it. I guess what I am trying to say is its a really hard job to bring VistA to a different platform without a large sum of money and some really smart people.

blhack 17y ago

The fact that "stimulus" has entered your standard issue citizen's lexicon as something that the government gives to a failing industry to keep it from failing is one of the most horrifying things to come out of the 2008 election cycle. WHY is this looked at as okay? I get that the headline was a joke...but why is it even funny? I'm sorry, I know this is off-topic (and more appropriate for a political forum like reddit), but I'm 22 years old. Fuck you, to the government that is taking away money that I haven't even had the chance to EARN yet so that companies that I have no interest in can continue existing. How did we get to a point where this is looked at as acceptable? Won't somebody please actually think of the children? You, government, the very people who are supposed to be looking out for me, are robbing from me in broad daylight, then having the audacity to tell me that it is for my own good?! What the hell is happening?!!?

RyanMcGreal 17y ago
When a large finance company goes down, an awful lot of innocent people are dragged down with it. When enough large finance companies go down at the same time, the entire economy goes down with them, leading to massive suffering. The bailout sucked. Not doing the bailout would have sucked far more.
- blhack 17y ago
  I agree with you, which is why I was 100% behind the troubled asset relief package. That was to help out the banks, which needed to remain liquid. I was extremely extremely vocally for TARP... This is something else entirely.

sachinag 17y ago

For fuck's sake, the code is out there: http://sourceforge.net/projects/openvista/ http://sourceforge.net/projects/openvista/ EDIT: apparently the main project is at WorldVista now: http://sourceforge.net/projects/worldvista/ http://sourceforge.net/projects/worldvista/ There are for-profit companies selling install/support services around it: http://www.medsphere.com/ http://www.medsphere.com/ (And here's their .org for their OpenVista repository: http://medsphere.org/index.jspa http://medsphere.org/index.jspa) Yeah, VistA is a PITA. What we need are data standards for non-imaging data (where DICOM has been a raging success). I've long suggested that all records have to be readable in VistA - it doesn't force people to use Medsphere, but it means that there's a lowest common denominator FREE software package that works.

siculars 17y ago
agreed. i'll just paste my comment from an article on the oreilly radar blog, http://radar.oreilly.com/2009/05/who-will-cut-the-gordian-knot.html http://radar.oreilly.com/2009/05/who-will-cut-the-gordian-kn... . In my opinion, one thing that government can do to accelerate electronic record keeping in the healthcare arena is mandating that solution providers must make means available to both import and export data to/from their systems. Mandating a specific format is less of an issue and should not be the main goal. If vendors were obligated to open their systems and encouraged to do so via some [fiscal/tax] incentive then the problem would solve itself. IT professionals either on their own or contracted would create solutions that connected the hodgepodge of systems out there. These systems are infinitely difficult to implement on any scale (by scale I mean number of people using the system - not technical challenges). At the moment I am tangentially involved in a major enterprise installation at a very well known institution. From my vantage point one of the major stumbling blocks is the difficulty in which the many, many systems already in place will interact with the new system (and with each other). The inability of these systems to interact with each other directly affects the cost of doing business in the modern day hospital. Many of the people in positions to make purchasing decisions are not fully cognizant of the concept of "vendor lock in" and "data portability". These poor decisions are paid back many fold not only in cold hard cash but also in opportunity cost over the lifetime of the system purchased, which may very well be decades. Government can make this all go away by simply requesting that all players in this space provide mechanisms for programmatic, unfettered data access. Providing fiscal incentive will all but ensure that those providers in compliance will be preferred. Mandating a specific system, or data exchange format is not the solution or, frankly, the "American way". Evening the playing field by ensuring openness through financial incentive is much more palatable to a broader spectrum of participants. The market place will ultimately decide and coalesce around its own standards and formats.