PEP308: Yet another syntax proposal

[Alexander Schmolck]

Erik Max Francis <max at alcyone.com> writes:

> Alexander Schmolck wrote:
> 
> > Frequency * expressivness comes close to measure the usefulness of a
> > construct. So for something that adds near 0 expressiveness, the
> > frequency
> > argument is extremely convincing.
> 
> So now to use these frequency statistics (which are contentious by about
> an order of magnitude), all we have to do is come up with a metric for

Even with an order of magnitude uncertainty (and I'd think one could do
better), the frequency doesn't look to impressive, does it?

> expressiveness that everyone will agree on to legitimize these
> statistics.  Hmm, right.

Even without bothering to come up with some formal and quantifiable criterion
of expressiveness, couldn't we just reach the consensus that something that
can be trivially and compactly rewritten with existing language constructs
isn't all that expressive?

So what's wrong with arguing against polluting and complicating the language
with a construct that seems to be both used infrequently and be easily
replaceable in those rare instances were it is used?

Do you have a better criterion to from the basis for debate (maybe one with a
more easily defined and empirically investigated metric)?

(BTW, although I don't think its immediately relevant here, there *are* formal
definitions of expressivness:
http://citeseer.nj.nec.com/felleisen90expressive.html)

alex