No subject

Tue, 24 Sep 1996 06:39:19 +1000

>From brd@netinfo.com.au  Mon Sep 23 08:09:03 1996
[203.8.127.196]) by wilma.netinfo.com.au (8.7.5/8.6.12) with SMTP id
IAA05435 for <link@charlotte.anu.edu.au>; Mon, 23 Sep 1996 08:08:58 +1000
(EST)
Date: Mon, 23 Sep 1996 08:08:58 +1000 (EST)
Message-Id: <199609222208.IAA05435@wilma.netinfo.com.au>
X-Authentication-Warning: wilma.netinfo.com.au: Host
betty-p04.netinfo.com.au [203.8.127.196] didn't use HELO protocol
X-Sender: brd@netinfo.com.au
X-Mailer: Windows Eudora Version 1.4.4
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: link@charlotte.anu.edu.au
From: brd@netinfo.com.au (Bernard Robertson-Dunn)
Subject: Architecture for Access to Government Information (Australia)
Sender: owner-link
Precedence: bulk

Re the Architecture for Access to Government Information report.

I have had a quick look at this document and I am a little
perplexed, maybe someone can help me.

On first reading, two questions come to mind:

1.   What objective is the architecture trying to achieve?

2.   Is the architecture a reasonable description of a
     comprehensive technology framework that will
     provide direction and guidance across all issues
     of access to government information?

On the first question, I cannot find in the report anything about
what the architecture is supposed to achieve - there do not seem
to be any requirements other than a few simplistic and vague
statements such as:

"Investigate how government information, irrespective of what
form it is in now, could be made more visible and for particular
holdings of information accessible, to either the public or to
officers of other agencies, subject to the provisions of the
Administrative Law suite."

and

"The Commonwealth Information Management Framework is a
high level framework through which the Commonwealth will,
through the innovative use of information, achieve its aims for
improvements in client service and government administration."

I can appreciate that it is not the job of the technical group to set
the requirements, but nowhere can I find reference to any
analysis that answers the question - "what is the objective?" I
cannot find where issues such as the following have been
addressed:

* Who are the people and/or organisations who want or need
  on-line access to government information?
* On what basis do you decide if a requester is allowed the
  information?
* How do you provide access to those without an internet
  connection?
* On what basis do you control access?
* How does a requester know that the information came from
  where it was supposed to come from and that the informatio
  is what it says it is?
* Should a charge be levied for some/all information? on what
  basis?
* Is it a requirement of the architecture to include provision for
  charging?

I would have thought that those tasked with creating an
architecture for access would have been given some indication
of the requirements, so that the solution can achieve the
objective. Does anyone know if such a document exists?

Without a clear objective, the implication seems to be that the
architecture should be comprehensive and complete so that it
can satisfy all requirements, once they are determined. The goal
seems to be - provide access for everybody to everything.
Somehow I do not think that this is acceptable to private
individuals and commercial organisations, both groups having
provided information to the government that they would expect
to be kept confidential.

>From what I can gather, although it is not explicitly stated in the
report, the architecture is aimed initially at information that is
currently available through normal government publications.
Does anyone know if this is true? If the information is already
currently available, what is the value or benefit to the
community or the government of making it available on-line?

This leads on to the second point. Is the architecture complete?
There are two important issues facing someone who is after
information that the government holds - where do I get the
information from and is it correct. The architecture covers the
first point but does not seem to address the issue of authenticity
of information as seen by the user. In the light of the recent CIA
and Department of Justice web page problems I would have
thought that this would have been a major area of concern and
addressed in considerable detail in the architecture. I can find
some discussion of the need for security and privacy but no
technical solutions or advice as how to ensure that the user gets
information that is correct and can trust it. Have I missed
something?

A fundamental assumption seems to have been made that the
state of the art of internet technologies is appropriate for the
government to make information widely available and that they
can guarantee its authenticity.

A quote from the report: "It was an assumption of the IMSC-TG
that as the Internet is already the dominant infrastructure
available to facilitate access to, communication with and
dissemination of government information, any technical
solutions would be based on the use of the Internet."

Does anyone know who made this decision? Does anyone
disagree with it? Does it follow that, as the dominant
infrastructure, it is appropriate for everything and everybody? Is
it true? I would have thought that there was a case for claiming
that the dominant infrastructure was the telephone/postal system.
Most people have access to a terminal device (ie a public phone,
mail box), there is a public index in most phone books, the
number of telephone calls to the government is probably greater
than the number of web hits (I must admit to guessing here),
more information transactions are carried out using a
combination of post/phone than on the internet ( a guess, again).
Is the Internet really dominant?

In addition to the above two questions, there are a couple of
other observations/opinions.

I am intrigued as to the definition in the document of
Information Management - "The management of all forms of
information using a common planning framework".

The definition currently being used by the Australian Defence
Organisation is that Information Management covers issues of
the use of information as a resource. It draws a clear distinction
between information and the systems that process information.
Defence Information Management sets the requirements for
systems but does not include systems.

The way the IMSC has interpreted IM seems to be aimed at how
information is managed, not why it needs to be managed. The
IMSC has focussed on the technology of management and is
greatly at odds with the Defence definition. Does anyone know
if the IMSC view is generally accepted in the rest of
government? If it is, then what is the term being used to describe
the management of information as a resource?

I am also puzzled by the use of the term architecture - the
document does not seem to follow any accepted definition of
architecture, certainly none that I have ever seen. In my view the
document is not an architecture, it is a discussion of issues and
recommendations as to a few standards. To quote from Clients
First, page 18 "Architecture defines the game plan for what is
unique and what is shared and provides a context and long-term
focus to ensure flexibility in the information technology
investment. An effective architecture is more than a set of
standards. ...".

As someone who has worked in the architecture field for many
years I rather take exception to the report on the grounds that it
is likely to confuse people who might think it represents an
architecture. It makes my job harder - I am currently trying to
assist Defence develop an IT architecture. Confusion is the last
thing I need.

One last point - The following statement is made in a discussion
on Access to Metadata - "A typical government department will
have a few hundred thousand PCs on desks around Australia"

Assuming a few is at least two and that there are at least five
typical departments then there would appear to be in excess of a
million PCs in the APS

According to Clients First (March 1995, page 86) a survey
identified 98,187 PCs (including laptops) in use in 97% of the
APS. Either someone has been spending up big in the past
eighteen months, or the IMSC-TG cannot add up.

My first impression of the report is that it is an incomplete
technology solution to an ill defined problem. Someone needs
to do some serious analysis of the problem before they start
throwing technology at it. The technology driven approach has
not worked in the past, I doubt if it will work now.

Does anyone else share this view?

The opinions expressed here are mine and do not represent those
of Defence or any other organisation.

regards
brd
+-----------------------------------------+
|Bernard Robertson-Dunn                   |
|Canberra                                 |
|Australia                                |
|brd@netinfo.com.au                       |
+-----------------------------------------+

______________________________________________
Email          : Tony.Barry@library.anu.edu.au
Vision & voice : CUSeeMe - ningaui.anu.edu.au
Voice          : +61 6 249 4632         Fax: +61 6 279 8120
Details        : http://snazzy.anu.edu.au/People/TonyB.html

Head, Center for Networked Access to Scholarly Information,
Australian National University Library, A.C.T. 0200, AUSTRALIA.