[mercury-users] Pred defns

[John Griffith]

>> "FH" == Fergus Henderson <fjh at cs.mu.OZ.AU> writes:

FH> I agree that the hidden argument proposals do make it somewhat easier
FH> to write code.  Only somewhat, mind you, since often you will still need to
FH> *examine* every part of the code to see that it threads the hidden arguments
FH> in the right order, even if you don't need to change it.  In fact one
FH> could argue that by forcing you to change the code, it forces you to examine
FH> it, which is not such a bad thing.

FH> But my main criticism is that I think making it easier to add hidden
FH> arguments in this way often this comes at the expense of making the
FH> modified code harder to read.  Because the arguments are hidden, it is
FH> harder to understand the program's data flow.

FH> In general improving writability at the expense of readability is not a
FH> good trade-off for languages other than those designed specifically for
FH> rapid prototyping, because in long-lived programs, code is written once
FH> and read hundreds of times.  I remain unconvinced that it would be a
FH> good trade-off in this particular case.

Why not have a user interface that (optionally) shows you which hidden
arguments occur in a clause?  Wouldn't this give you the best of both
worlds?

-- 
John Griffith
Seminar fuer Sprachwissenschaft, Universitaet Tuebingen
email: <URL:mailto:griffith at sfs.nphil.uni-tuebingen.de>
www: <URL:http://www.sfs.nphil.uni-tuebingen.de/~griffith>