Larry,
My only fear here is that varchar will eat up a lot of memory. These are about 8 fields (Columns) in a row. Also, if the pointers get broken, it is a disaster - these are medical records. I am still pondering how to do this. If there was a longer row thrshhold, I would probably stick with notes in this particular case.
BC
| Lawrence Lustig <[EMAIL PROTECTED]>
Sent by: [email protected] 05/05/2005 09:31 AM
|
|
> They are (and should be) much more concerned with the medical information,
> not the look and feel.
> My job in this project is to come up with what the doctors want to see and
> make it mindless for the data entry nurses.
>
> If I can ever figure this out to where it works seamlessly and elegantly,
> then I can port the technology to physician office dictation systems and
> actually sell some copies of R:Base.
I'm coming to this conversation late and haven't gone back through all the old
messages.
Are you considering switching from NOTE to VARCHAR? I love VARCHAR -- at least
in principle. They are stored in the #4 file, separate from the row data, and
don't count against either the 4K row length or the 2G file size.
You should be aware that there are some substantial limitations on VARCHAR
columns. In particular, they cannot be used in expressions. That means that
you can't "build" the text in a VARCHAR column like this:
UPDATE Table SET VarColumn = (VarColumn & 'Case discount applied.') +
WHERE PackingType = 'Case'
You also can't SRPL or SLOC on them.
In addition, I've had some trouble with the line end character conventions
being different in NOTE and VARCHAR fields.
In general, VARCHAR is great if the primary access to the data will be through
bound controls on forms. If you want to use the data as maleable text values,
however, you may be better off sticking with NOTE fields.
--
Larry
