[2638] in Release_7.7_team
Re: Please strongly consider backing out the zephyr servers
daemon@ATHENA.MIT.EDU (Ginny Williams)
Wed Mar 7 15:49:01 2001
Mime-Version: 1.0
Message-Id: <p0500190bb6cc33a7dd7b@[18.162.2.64]>
In-Reply-To: <v04220802b6cc10818ad3@[18.18.0.128]>
Date: Wed, 7 Mar 2001 15:46:38 -0500
To: Ted McCabe <ted@mit.edu>, Garry Zacheiss <zacheiss@mit.edu>,
"Jeffrey I. Schiller" <jis@mit.edu>
From: Ginny Williams <ginnyw@MIT.EDU>
Cc: Garry Zacheiss <zacheiss@mit.edu>, "Susan S. Minai-Azary" <azary@mit.edu>,
Greg Hudson <ghudson@mit.edu>, John Hawkinson <jhawk@mit.edu>,
release-team@mit.edu, op@mit.edu, winzephyr-release@mit.edu
Content-Type: text/plain; charset="us-ascii" ; format="flowed"
All,
The upgrade to the zephyr servers highlighted two issues in my mind,
one that will be (I hope) unique to this particular software at this
point in time.
The first is that, even though Winzephyr is currently considered an
alpha product and is being tested by a small number of people within
the Software Release Process, its unclear how many previous versions
of it are out there and how many people are using it since it's been
in development for some time. So if the staff upgrading the server
were following the efforts of the recent release team, they would
have surmised that there were only a few alpha testers using it.
Unfortunately, that's not really the case because of all the people
who began using it during the entire development cycle. We're also
vulnerable to this with Plop, which has the same circumstances albeit
with a smaller user base.
At 11:35 AM -0500 3/7/01, Ted McCabe wrote:
>My apologies for continuing the disussion on this topic, but it
>strikes that during all of this discussion that there is a
>fundamental question that is explicitly unanswered.
>
>Who do we consider to be a customer?
>
>
>It seems to me that we don't all have the same idea what the answer
>is. I have the impression that ASO and ghudson addressed this
>zephyr problem from the point of view that a customer is basically
>someone who uses IS supported software, while Jeff and Susan et. al.
>seem to have been working with the idea that a customer is basically
>someone who uses IS' servers.
The second issue is that, from my experience, even though we identify
supported software and services it's very difficult to deny anyone
access to support if they request it. This is because most people
see themselves as customers of IS and therefore entitled to support.
In my time spent at the computing help desk we worked on many client
problems that involved software and services that weren't officially
"supported", and it was very rare that we were ever able to deny
support unless we were helping someone move to a supported product.
Instead we always provided the best information and help that was
within our expertise.
When we deny support we are seen as unresponsive and unhelpful, and
therefore not good citizens of the MIT community. In the purely
client/service provider perspective, one could argue this to be the
case. On the other hand, there are reasons why we direct people to
specific products; there are economies of scale, issues of quality,
availability of support, and overall stability of the MIT computing
environment.
Since the forces of customer service will always conflict with
control of the environment and good use of resources, the only
solution I see is to always understand what the customers are using
beyond what we officially support, and to strongly communicate what
we support and why. By knowing what clients are actually using we
can work hard to "banish" certain products and practices, or we can
add popular and useful products and services to our offerings.
I would also be interested in participating in a discussion of this
face-to-face.
Regards!
Ginny Williams
Software Release Team
