RE: Re[2]: Is 'testability' a useful architectural constraint?

I agree with both you here, but I think that the main point is slightly
different. And that is that the cleanliness of an architecture can (in part)
be demonstrated by showing that it can be tested in a reasonable way. The
converse is also true: the cleanliness of an architecture is suspect if a
reasonable test suite cannot be created.
-Aaron

-----Original Message-----
From: Chris Lilley [mailto:chris@w3.org]
Sent: Monday, January 07, 2002 8:35 AM
To: Mike Rajkowski
Cc: www-tag@w3.org
Subject: Re[2]: Is 'testability' a useful architectural constraint?


On Monday, January 07, 2002, 5:03:07 PM, Mike wrote:

MR> Chris wrote:
MR> "Thus, testability aids clean architecture."

MR> I actually believe the opposite.

MR> Clean architecture aids testability.  A good architecture need to
MR> address what occurs when it is improperly used as well as when it is
MR> correctly used.  If something is designed with improper/limited criteria
MR> for the error cases, then dealing with these cases will be hard to
MR> implement or even test.

Well I elieve that statement too, which implies that there is a
feedback loop there  - improving architecture improves testability
which improves architecture of other specs that join on to that one.
Sort of like ripples in a pond.

-- 
 Chris                            mailto:chris@w3.org

Received on Monday, 7 January 2002 13:11:32 UTC