From skotthe@ccs.neu.edu  Mon Nov 18 17:19:38 2002
Date: Mon, 18 Nov 2002 17:19:37 -0500
From: Therapon Skotiniotis <skotthe@ccs.neu.edu>
To: lieber@ccs.neu.edu
Cc: "Wupc@Ccs.Neu.Edu" <wupc@ccs.neu.edu>,
	"Lieberherr@Ccs.Neu.Edu" <lieberherr@ccs.neu.edu>
Subject: Re: FW: I wouldn't be supportive of AOP?

Well , I do agree that most of the remarks made in the paper about
program evolution (Section 6), are the same issues that AOSD is currently
addressing (claiming to address). 

Also, I find that the description (I did not see details in terms
    of implementation to some of the ideas, e.g., how do I define a
    'probe') has all the necessary ingredients to categorize the
    whole system as an AOSD system. 

I would also dare to take it a bit further and say that some of the
component/unit ideas seem to be also there (although I am not sure
    if all of the now known cababilities fo component/units are
    covered by this system).

To put it simply, this system sounds great ... could we read more :)
! 

-- Theo 
On Sun, Nov 17, 2002 at 10:17:02PM -0500, Karl Lieberherr wrote:
> 
> It appears to me that this paper by Ivar can be considered an early AOP
> paper?
> It even touches on dynamic changes to a running system by something similar
> to an aspect?
> 
> This paper appeared in the first OOPSLA conference.
> 
> How do you view the paper wrt AOSD?
> 
> -- Karl
>