----- Original Message -----
From: "Gregory Woodhouse" <[EMAIL PROTECTED]>
On Jan 15, 2006, at 11:14 AM, Kevin Toppenberg wrote:
application programmer by trying to push as many variables as possible to
the global frame. Such thinking is outmoded, and is not conducive to
sound development practices.
Isn't the issue here that a new version of GT.M. or Cache cannot break VistA
or RPMS? Aren't there ways to introduce new elements into the language that
will not break old code?
My biggest beefs with M would be the issue of variable scoping,
allowance of single letter commands leading to code that is very
difficult for newcomers to read, and limitation to 8 characters in
variable names etc (made worse by the need for namespacing.)
I agree. Both restrictions need to go. More importantly, though, using
symbol variable names to implement namespaces is a very sloppy solution.
Making namespaces (or packages or modules) first class objects is a much
better solution.
I know
that this last point has been fixed in Cache' and GT.M, but is still
not used in the VA.
Doesn't this mean that the issue of 8 character names is not a modern M
issue at all. It is just an issue of the VA SAC. Why blame it on Mumps
anymore?
Jim
-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems? Stop! Download the new AJAX search engine that makes
searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
_______________________________________________
Hardhats-members mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/hardhats-members
