[Baypiggies] The (lack of) design patterns in Python

[Alex Martelli]

On Tue, Aug 4, 2009 at 1:30 AM, Shannon -jj Behrens<jjinux at gmail.com> wrote:
> On Sun, Aug 2, 2009 at 7:42 PM, Tony Cappellini<cappy2112 at gmail.com> wrote:
>> We have openings for November and December..
>>
>> Would anyone here be interested in hearing Joe Gregorio's presentation
>> on "The (lack of) design patterns in Python" ?
>> I think it would be good for the group to hear, since many didn't get
>> to go to Oscon.
>>
>> If yes, are any of the Google employees in touch with him and would
>> you be willing to ask him?
>
> +1
>
> However, I must admit, it'll probably drive me crazy.  I read the gang
> of four cover-to-cover.  I think design patterns are language
> specific.  For instance, I think that monads are a design pattern that
> are really helpful in Haskell.  I get really bent out of shape when
> people say, "My language is so powerful, it doesn't need design
> patterns!" *sigh*  For instance, using class methods for providing
> multiple constructors is something that Alex Martelli taught me, and I
> think it qualifies as a design pattern.  (His example was
> DateTime.from_gregorian_date(), etc.)

Indeed, as another deep lover of design patterns, I was thinking a
debate format would be much more appropriate than a lecture format,
for presenting such a controversial thesis -- maybe 5 minutes to each
side for an initial exposition, then alternate
question-answer-antianswer-rebuttal with a moderator (audience would
submit questions and vote for them in advance e.g. via google
moderator, a human moderator would exercise editorial judgment to
pick, sequence and address an actual set of 6 questions or so,
possibly adding some of their own if the dory doesn't offer enough,
with about 10 carefully-timed minutes devoted in all to each question)
and finally a little freeform Q&A from the live audience at the end.

I volunteer to take the pro-patterns side if we do set things up that way.


Alex