[LRUG] Topic for Next meeting
benjohn at fysh.org
benjohn at fysh.org
Mon May 22 04:49:58 PDT 2006
>
> That sounds great Benj, I'm sure people would like to see what you've
> got. BTW, have you had a look at Mongrel?
I've come accross it. I don't actually use Ruby (atm) for web scripting,
so Mongrel's not in my immediate locus of attention :)
What I had in mind was a fun and easy setting for trying to show a few
reasons why I like the language.
> I hope I didn't come across as being rude/forceful etc in my last email,
> I apologise if I did, you know how emails are.
:) Not at all.
> Would people be interested in looking at a whole unfinished/experimental
> DSL for finite state machines? I could compare it to 'finite state
> machine generator'.
>
> Also, any comp sci heads out there want to talk about continuations? I
> figure if they can have a talk about it at the RubyConf then we can have
> a talk about it here :) Also Monads, Functional programming etc
I am (among other things) a comp sci, and I'm proud of it. Those are
interesting topics, and various approaches would be amenable to doing
them.
> We could also have a general forum about various topics, I'd be
> interested in finding out about what 'real object oriented programming'
> is - Tell, don't Ask and all that
If you don't ask, you wont get told?
Wish I new. Personally, I think OO's realisations are flawed. Class
hierarchies seem to model the universe in quite a brittle way, and
tucking state away in objects seems brittle too.
What OO does get right (absolutely spot on, IMHO) is that it's really
useful to have a context (an object) within which you can ask abstracted
questions ("What colour are you?") or make a change ("I change your
state to being an email that is ready to be sent!"). I question the
mechanism that makes that work: classes, inheritance, virtual functions,
multi methods, state stored in frames...
...But I'm at a zero beer level at the moment, so I'm not in the right
mood for a fight about it :)
Cheers,
Benj
More information about the Chat
mailing list