Trevor Kerr wrote:
> That's my question - having assembled your low-end servers running
> Linux, how long does it take to populate a modern database with the data
> from the old COBOL mainframe, making it available for massaging on the
> NT4 desktops?
The question of the hour! If I had to guess at it, I would say that the
easiest way would be to get some HL7 *2.x* interfaces running. After
all, that's what everyone else is doing in their best of breed worlds.
Obviously, the 2.x interfaces would be an example of programmed
obsolescence, but hasn't the health IT industry created its own trojan
horse by making everything HL7 2.x compliant. They did it so that all
the vendors could sell closed systems and claim they were open.
Wouldn't it be a pity if the systems suddenly became open.
Someone else might say, well if we get Corbamed up and running *it* will
get all that old COBOL data for us. Why reinvent the wheel? HL7 2.x is
already running, and all we want is the data. We *don't* want the
proprietary applications behind the data. COBA is an *application*
interface, but, by definition, we want no part of the applications, just
the data.
We want open source applications and legacy data. I say use HL7 2.x
What do others say?
John
