From mira@ccs.neu.edu Mon Feb  9 11:16:09 1998
Received: from taboor.ccs.neu.edu (mira@taboor.ccs.neu.edu [129.10.112.115])
	by amber.ccs.neu.edu (8.8.6/8.8.6) with ESMTP id LAA20371;
	Mon, 9 Feb 1998 11:16:05 -0500 (EST)
From: Mira Mezini <mira@ccs.neu.edu>
Received: (from mira@localhost)
	by taboor.ccs.neu.edu (8.8.6/8.8.6) id LAA21902;
	Mon, 9 Feb 1998 11:16:05 -0500 (EST)
Message-Id: <199802091616.LAA21902@taboor.ccs.neu.edu>
Subject: Re: more integration
To: dougo@ccs.neu.edu (Doug Orleans)
Date: Mon, 9 Feb 1998 11:16:04 -0500 (EST)
Cc: dem@ccs.neu.edu
In-Reply-To: <199802070639.BAA02300@vega.ccs.neu.edu> from "Doug Orleans" at Feb 7, 98 01:39:50 am
X-Mailer: ELM [version 2.4 PL23beta2]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Status: R

Hi Doug:

> 
> Class1 {        // No name-mapping yet, assume there is a class
> 		// called "Class1".
>   traversal toJournal(JournalVisitor) { to Journal; }
>   void event1() = toJournal {
>     before Class1 (@ /* ... doSomethingBefore1 .. */ @)
>     before Journal (@ mystructure.register("Class1.event1"); @)
>   }
>   void event2(Par p) = toJournal {
>     before Class1 (@ /* ... doSomethingBefore2 .. */ @)
>     after Class1 (@ /* ... doSomethingAfter2 .. */ @)
>     before Journal (@ mystructure.register("Class1.event2", p); @)
>   }
> }
> 
> Is this roughly what you intended?  

Yes, this is what we roughly intended. Thank you for providing the
corresponding DEM/Java pseudocode for it. It helps clarifying
terms.
 

>If so, it seems like you want to
> add two things:


>   1. parameterization, i.e. you can pass a strategy in and refer to
>      its parts with class-valued variables, rather than hardwiring
>      them.
>   2. encapsulation, i.e. packaging the two traversals together,
>      arranging the code according to class rather than function, and
>      (potentially, but not in this example) sharing local data and
>      functions across multiple traversals.
> 
> Do I have this right?  Is there more to it that I'm missing?

I like your formulation. It says in a very exact form how adjusters 
differ from adaptive methods. 

- Mira