Hi Patrick:

there is not a centralized list of talks in the Boston area.
But when the time comes closer, I will forward some announcements to you.

Let's discuss the issues you raise when you are here.

Best regards,
-- Karl

>From patrikj@cs.chalmers.se Tue Apr  7 16:01:02 1998
>Reply-To: Patrik Jansson <patrikj@cs.chalmers.se>
>To: Karl Lieberherr <lieber@ccs.neu.edu>
>cc: dem@ccs.neu.edu, johanj@cs.chalmers.se
>Subject: Re: Polytypic and Adaptive
>
>[Sorry about the late answer - I was on vacation.]
>
>On Fri, 27 Mar 1998, Karl Lieberherr wrote:
>> 1+2. We promote the same useful idea:
>>    Programming language constructs to write programs which can operate
>>    on vastly different data types. 
>>    Polytypic for FP, adaptive for OOP
>I fully agree, and I think there ought to be opportunities for both
>approaches to benefit from one another. I have just read 
> "Traversals of Object Structures: Specification and 
>     Efficient Implementation"
>and I have skimmed two similar papers earlier. I am not very good at OOP
>but it seems that the class graph plays a similar role in OOP as a
>recursive datatype (or rather a group of mutually recusive datatypes) does
>in FP. 
>
>> 3. It seems to me that the Polytypic work can benefit from the concept
>>    of strategies for better high-level control.
>[...]
>>      length :: d a -> Int {via x to a}
>>    where d ranges over type constructors using x and a.
>>    length is undefined if the actual parameter for d does not use x and a
>>    and there is no "path" from the current type through x to a.
>
>Such a generalisation is indeed possible, but I think that a more
>natural solution for selectively traversing some parts of a structure in a
>FP language is to pass around a predicate (pred :: d a -> Bool). This can
>be done polytypically (using a paramorphism) and of course, defining a
>predicate that avoids paths with x in is not difficult.
>
>> 4. We are currently working on visitors and adjusters which are 
>>    parameterized by strategies. They are polytypic components which
>>    can be composed and maybe the polytypic framework you have
>>    can be used to formalize elegantly visitors and adjusters.
>In the Functional polytypic setting a visitor is a catamorphism (or a
>paramorphism) and an adjuster probably corresponds to a map. 
>
>> I was wondering how the ideas in PolyP can be best applied to oo
>> languages, like Java.
>Me too ;-)
>
>In spite of the obvious differences [most notably in language semantics
>and type systems], there ought to be a number of ideas that could be
>carried over to OOP. I will try to identify some similarities and
>differences before the talk so that we can discuss it in Boston.
>As it look right now I will stay in Boston for the rest of the week
>after my wednesday talk, so that should give me a possibility to talk to
>you, and others at NEU, (and perhaps at BU, MIT). Maybe you know of other
>talks or meetings I could attend when I happen to be in Boston?
>
>Patrik Jansson
>
>
>