Digraph proposal from Denmark

D. Hugh Redelmeier uunet!redvax!hugh
Fri Mar 27 09:10:23 PST 1992


| From: Doug Gwyn (VLD/VMB) <gwynaBRL.MIL>
| Message-Id:  <9203271116.aa28336aVGR.BRL.MIL>
|
| The traditional name for && is "cand" (conditional AND),
| presumably for || it would be "cor" although I've never seen that used.
| My guess is that "and" and "or" were chosen because they are SHORT.
| Remember, the proponents of this stuff really intend that coders type
| it in, thus short names should be used for the frequently-used macros.

Doug is quite right.  I certainly have seen and used "cor".

I don't have a vote, and I have been letting this stuff pass by
because I have hoped it would go away.  I still hope that.

Furthermore, it seems too late to "improve" the proposal.  But since
the Danes seem to feel free to do so, I will throw in my two cents
worth of constructive comments.

- "cand" and "cor" are much better than "and" and "or".

- This would allow bitand and bitor to be renamed, but I think that
  they are fine.  I certainly would not name them "and" and "or".

- *_eq seem like bad names.  They suggest the characters they
  replace rather than the meaning of the operators.  What these
  operators do is not "equality" but "assignment".  It is one thing
  to use the = glyph, which can be verbalized as something other than
  "equals"; it is another to use the anti-mnemonic "eq".  Off the
  top of my head, I would suggest *_by or asgn_*.

- I *think* I would prefer that these things were in upper case.  I
  suspect this is because I consider them to be second-class to
  keywords.

Hugh Redelmeier
hughamimosa.com or {utcsri, uunet!attcan, utzoo, scocan}!redvax!hugh
When all else fails: hughacsri.toronto.edu
+1 416 482-8253



More information about the Numeric-interest mailing list