OK Rick. I'll take that as an invitation to rant...
I currently do support at the Washington DC VA Medical Center. I work mostly with COTS packages and solve problems that require multi-disciplinary skills. A good example of a long term task I recently accomplished was making electronic consents work without any wrinkles throughout the hospital. It took 2 years and required cooperation with many different people. There were networking issues, problems with the vendor software (that they have admitted, but only tangentially), issues with Sentillion (Vergence Locator -- we now have a new version that the users can't misuse) and user education problems. There is a very long story on this.
A medium term project was writing an M application to assign responsibility for equipment to users. A national package that did the same thing (and has been a lot harder for our users here to work with and understand) then preempted this.
I have also been given the responsibility for interacting with Surgery. I spent a long time being nice, helping with device and software issues and generally making sure that whenever they had a problem, they got instant service. I have, therefore, learned OR protocols and how to put on a Smurf suit in less then 2 minutes in order to go into an OR to make equipment work properly just before a surgical procedure.
Previously in the VA, I worked at the Washington DC IRMFO, previously known as the Washington DC ISC. I came from a medium-sized MUMPS shop and had done tracking and billing systems. With my team, I traveled around and installed this software at a lot of hospitals. When I was hired I was put onto some tasks doing screen managed applications for managers that weren't impressed with what was the VA standard then, roll and scroll. I met the Kernel people and worked with them to get the Terminal Type file to include escape sequences for screen control and eventually introduced them to escape processing (which is how screen managed software allows the user to navigate with tabs, back spaces and arrows...), wrote a prototype work processor, which served as an example of what could be done and how. I also introduced the idea of the FileMan reader (DIR) and wrote the prototype. This innovation allowed Alerts to be announced to users no matter what they were doing in VistA and helped maximze the use of program buffers as most users are usually "read hung" and, therefore, only one routine needs to be in the buffer for all of them.
In the process I met a lot of powerful, smart people. I was very lucky to have known and worked with these people. They were very interested in results, and had the confidence to teach others and cooperate with anyone that had something to offer. VistA was a team effort that included clerks, technicians, medical service providers and administrative staff. In those years everybody wanted to help. People were being taught FileMan so they could write their own reports and databases. It was Camelot.
Because the "national email node" (FORUM) was in DC, I was given responsibility for the MailMan package when Tom Munneke left. (BTW: I had dinner with him a couple of months ago and you might want to see what he is doing now -- still quite the innovator). I worked with Milt Roberson and Jack Divers and made FORUM capable of handline up to 1,000,000 message units a day. I worked with people at Austin, where the national data center is, to speed up transmission of data via email, which was originally across dial up lines and developed new transmission protocols. Later I worked with Micronetics to develop the first ability in the VA to receive email across a TCP/IP channel (basically a clone of Inet-servers -- the original Unix code), a capability that generalized for all potential services, and allowed us to develop the Broker.
I left the VA in 1995 or so and worked with some private companies. I did pharmacy robot interface support and development. I learned SQL and Oracle PL/SQL and managed a large resevoir of PL/SQL code that converted old data into a new database for the Coast Guard Fleet Logistics System. So, I now understand RDBMS.
I was an accountant before all of this. And I have since earned a MS in Telecommunications management. Therefore, I can also do cost accounting and have been known to write an occaisional rant (managers call them studies or analyses) that includes references and a bibliography.
Right now, I have a couple of ideas for development. The model for the future of VistA is "Open VistA". I can buy into that. I believe that this idea is supposed to make if possible for more COTS products to enter the VistA environment. I see that as a plus and a minus. But I think that creating an environment that makes it easiest for all who do development in VistA to use VistA tools is the best model. Therefore, the following.
FileMan needs an option that allows data to be "published". The current print capability needs to remain. But when data is reported and distributed to less computer oriented people, including medical staff, managers and especially users, it needs to be structured and easy to read. There needs to be the ability to output the data in a way that it can be printed to distributed electronically. Right now HTML seems to be the best candidate. I envision a process that includes the design of some default style templates and allows users to clone these and create their own so that reports can be output through with the style template as the model for the document. I do not think this will be terribly difficult. And I think that if the user starts in a GUI environment, the output can be directed in many different directions.
|