On Thu, Sep 20, 2001 at 06:45:24AM +0100, Simon Cozens muttered...
: On Wed, Sep 19, 2001 at 05:39:26PM -0700, Paul Prescod wrote:
: > I propose, for example, "intval" for IV and "floatval" for NV.
:
: I reject. :) Larry has a neat little principle:
:
: Common operations should be "Huffman coded". That is, frequently used
: operators should be shorter than infrequently used ones. For how
: often it's used, the C<scalar> operator of Perl 5 is too long, in
: my estimation.
:
: I think the same goes for type naming. You only have to understand what
: "IV" means *once*, and then you can just type two letters all the time,
: instead of the beautifully accurate and obviously easy to understand
: integer_guaranteed_to_hold_a_pointer_t.
:
: Seriously, I think you underestimate programmers; they can learn and remember
: two letters. Now, sure, PV's confusing as hell. That's why we use "STRING".
: There's got to be a tradeoff.
Just speaking from experience, code is so much easier to understand if meaningful
identifiers are used and the code can be read rather than 'de-coded'. Its then
means that the programmers dont have to document the code (because it is self
documenting) which in turn means more people can pick up the source code and
understand and contribute, which is kinda the crux of OpenSource isn't it?
Now I have to say that when I got this slammed at me 4 years ago in a CS
class by a particularily clueful CS teacher I laughed at him thinking - why?
Unfortunatly there are far too many OpenSource projects that answer the
'why?' question and it's painful to try and modify or contribute to them, and
I consider myself a good programmer.
Just my two cents,
Regards,
Chris
--
Chris Ross (boris) [ [EMAIL PROTECTED], http://www.darkrock.co.uk ]
[ (chris|boris)@ferite.org, http://www.ferite.org ]
"Ferite - A small scripting engine with the traits of many."
