The ACS-link Project - a Case Study
Author: Peter Talty BA MACS
ACS Canberra Branch Conference: MULITNET'95
Abstract: The first attempt by the Australian Computer Society to provide a network service for it's 15,000 members has not been without problems - but seven months since it was launched it is servicing 4,000 members widely dispersed across Australia (and even some overseas). This paper looks at the decision-making process within ACS leading up to the launch and some of the lessons learned to date. It finishes with some brave predictions about where this project may lead over the next few years.
Introduction
On 8th November 1994 the Australian Computer Society switched on a network service for about 2,000 Members across the country who had registered in the month since ACS-link was announced. The service allows them to dial a local number from anywhere in Australia and connect to a host computer from where they can also access the wider Internet. There is a period of 5 hours peak-time usage per month included in the registration fee ($50) or 10 hours off-peak usage - where peak time is between 8.00am and 6.00pm Melbourne time on weekdays.The host computer provides full Internet access and thus includes electronic mail, conferences (discussion groups), World Wide Web, and a repository of ACS information such as technical papers, regulations etc. Those with access to the Internet via some other provide (eg. at work or university) can telnet directly to the ACS-link host computer.
The service is provided by a Brisbane based company Pegasus Networks.
Background
The membership of ACS in the early 1990's was around 14,000. Of these about 7% were academics and since the Internet (in the form of AARNet) had been in all Australian tertiary institutions for many years (some since the late 1970's) it could be assumed that almost all these academics had access to the Internet. In addition, 1992 and 1993 saw a rapid expansion in the number of private and government organisations who had access to the Internet (via AARNet). My estimate of other members who had Internet access in 1992 is 3% - giving a total of 10% with access in 1992.1991 and 1992 also saw the emergence of more user-friendly ways to find information on the Internet - such as World Wide Web format. There was a concerted move to open up access to the Internet to the public.
The Role of the ACS Council
ACS Council is the ultimate decision making body in the ACS. Each State is represented and there are also two vice-presidents, the treasurer, the president, and several Board Directors. The State delegates hold by far the majority of votes since the number of votes they get is related to the number of members in the State Branch. eg. Victoria with 4,500 members holds about 26 votes, and the others go down from there. The President, vice-Presidents etc only hold one vote each.Another significant item of information about Council, and one which affected the progress of this project, is that State Chairmen can only hold office for two consecutive years. Since they are also normally the Councillors there is rapid turnover in Council and significantly, lack of continuity and what might be called corporate memory.
In 1992 Council passed motions recommending that something be done to provide email for members. The reader should realise that everyone at Council is there in an honorary capacity except the (non-voting) CEO. They are all holding down full-time jobs as well as doing ACS activities. Unfortunately, the project did not get any specific resources either in the form of a project manager or funds to progress it. Needless to say it languished.
The Canberra Initiative
Canberra Branch in 1993 had about 800 members. Of all the cities in Australia Canberra had (and still has) the biggest networks - largely because the head offices of Federal Government departments are located there. It follows therefore that there are more IT people in Canberra involved in networks.In March 1993 the Canberra Branch Committee voted funds to set up a basic Internet service for Canberra members using the expertise of and some equipment of one of the members. The initial connection was for email and News only - since the host only did a dial-up connection to one of the universities twice a day.
The service attracted keen interest from members. By the end of 1993 it was clear that members were using it, and most importantly a survey of users in late 1993 indicated that they were prepared to pay for a permanently connected service.
This permanent connection went ahead in February 1994 with two types of users:
- Those who only used email and News did not pay anything in addition to their annual membership fee;
- Those who wanted full access to the Internet paid a special annual fee of $200 on top of their membership fee - calculated as one fortieth of the annual AARNet connection charge at that time for a permanently connected 9600 baud line (forty being our best guess as to the number in Canberra who would pay $200 for the service).
As it turned out, more than 40 "full access" users were on the system within three months. By mid 1994 20% of the Canberra membership of 800 were using the service. Usage statistics were analysed once the system settled down - giving Council something on which to base requirement for free time per month in negotiations with service providers which was also happening about this time.
The Call to Industry
In February 1994 the Director of the Marketing Board sent out a request for expressions of interest to industry. This had been prompted by a Council meeting late in 1993. The response was less than enthusiastic but subsequent discussion firmed up a response from Pegasus. The proposal was unacceptable to Canberra Branch however since it would have meant a return to a very basic service from which they had now progressed.Because of the success of the Canberra system the then Chairman, Peter Talty, with technical advice and assistance from Charles Smith put together a proposal for a national system - primarily because the proposals from industry (discussed below) they considered to be inadequate for ACS members.
The Canberra Proposal
This proposal was put up to Council as direct competition with the proposals from industry - because those proposals were considered inadequate. The Canberra position was that the service had to be at least as good as what they had given to members. There was no going back to a lesser service. The Canberra service by then had full Internet access - including SLIP and PPP access so that members could use Windows-based tools like "eudora" and "mosaic".The "threat" of the Canberra proposal had the positive effect of getting significant changes in one of the industry proposals - the one from Pegasus networks. As a result Council in September 1994 accepted the revised Pegasus proposal with unanimous support from all Branches and began a fast-track implementation process so that the network could be launched on November 8th.
The Best of Both?
Pegasus Networks committed to delivering a service with all the features of the Canberra service - but without ACS having to be involved in establishing infrastructure and support services. Was it too good to be true? Many thought so - since all other service providers charged in the order of $200 per year - and here we were (and are) getting it for a one-off fee of $50 per member. Pegasus believe they can subsidise the service by selling advertising space to companies and individuals who are attracted by the buying power of so many IT professionals in one place.
What we got.
- The first national network for ACS members - both urban and rural.
- Good support via email and 1-800 freecall number
- A way to rapidly get thousands of ACS members up to speed with the Internet technology that is sweeping the world at the moment.
- Exposure for ACS as doing something useful.
- Our own mail domain "acslink.net.au" which can be retained as a permanent address for subscribers even if the service moves to some other host system.
What we didn't get.
- No network for ACS offices. Staff in ACS offices can dial-up ACS-link the same way members do but this is not workable for some.
- No guarantee of continuation of the service. At any time Pegasus can give six months notice of termination of the arrangement - though to date principals have clearly indicated they want to continue. (ACS can also terminate the agreement with six months notice)
Management considerations - present and future
The ACS Management Committee as well as Council have been very supportive since the ACS-link decision was made. It is clear to all that this is a service many members find worthwhile and it is also clear that there is great potential for changing the way we do business with members by making more use of the network. Some of the considerations so far:
- Ease of connection for members: ACS was able to negotiate a very good price on modems which about 500 members have so far taken up.
- Ease of use of software: up to date some members have experienced difficulty in downloading and installing some of the software. It is now available from Pegasus on diskette to make it easier.
- Documentation: All members who sign up with ACS-link get a detailed manual to help them get started with the Pegasus service. It is being upgraded to add new features as they appear.
- Training: This has been tackled on several fronts. Firstly, there is on-line help in the form of conferences where someone having a problem or wanting to know something can post a message and have it answered by either Pegasus support or another member. Secondly, the support staff at Pegasus are available by email or a 1-800 phone number - so that problems can be resolved as speedily as possible. Third, some Branches are now running regular training sessions on ACS-link topics - keeping costs low and attracting full classes.
- Development of conferences: From day one there have been discussion groups (conferences) available to ACS-link subscribers. Not all of them have been active but some have been. These forums cannot be artificially activated - it is up to members to generate the activity.
- Development of World Wide Web material: The ACS now has in place a useful amount of material on it's WWW server. Not all States yet have local material available but it won't be long before they do. Information that was on the IT Practitioner's Guide is now on the Web server - along with interactive membership and regrade forms, links to such things as IT World 95 information and IFIP 96 information. There have been some very dedicated and expert members involved in getting the WWW server to where it is now and it will probably take even greater effort over a long period to maintain in a manner useful to members.
- Funding: In 1994 the ACS spent about $70,000 to get ACS-link going - this covered publicity, legal fees, travel and more. This year running costs for ACS will be in the order of $25,000. There is no direct income to ACS from members joining ACS-link - but since it is attracting new members (and perhaps retaining others who may otherwise have left) there is an indirect benefit - not yet quantifiable.
Objectives progressed
It is important to highlight the connections with the ACS Strategic Plan. This project addresses much more than " Provide email for members". It is a project which is having an effect on many of the objectives in the ACS Strategic Plan.The section below lists the elements of the strategic plan which are potentially affected by this initiative - with comments about how each is affected.
Objective 1: To increase the professional membership of the society
Has ACS-link progressed this objective?
Yes... My estimate is 500 in just five months.
Objective 2: To have professional membership of the ACS accepted as a prerequisite for employment in the IT field
Has ACS-link progressed this objective?
By raising the profile of ACS this objective will be much easier to achieve. ACS-link has not directly progressed this objective.
Objective 3: To develop and promote professional standards in IT
Has ACS-link progressed this objective?
Yes... ACS-link is providing forums for members to discuss standards as well as many other issues.
Objective 4: To be, and be seen to be, a credible and authoritative commentator on professional and relevant industry matters.
Has ACS-link progressed this objective?
Yes... Internet is only one area of IT but the reputation we gain by providing ACS-link reflects in other areas.
Objective 5: To enhance the public image of IT professionals
Has ACS-link progressed this objective?
Yes... The public want access to the Internet and now see ACS as something more relevant to them.
Objective 6: To be recognised within Australia and the Asia-Pacific region as the pre- eminent Australian association for IT professionals
Has ACS-link progressed this objective?
Showing leadership in this form can only help the achievement of this objective.
Objective 7: To promote the code of ethics
Has ACS-link progressed this objective?
Yes... ACS-link will be an excellent vehicle to do this. So far there has not been much activity on this front.
Objective 8: To enhance the Society's financial viability and administrative effectiveness
Has ACS-link progressed this objective?
Yes... ACS-link will be positive on both aspects of the ACS operation - provided that opportunities for administrative improvements are taken and risks are properly managed.
With such a potential for progressing many of the Society's objectives, this initiative rated highly for commitment of ACS resources.
Lessons from this project
Some of the lessons relate to the internal functioning of ACS - in particular the way Council follows-up on decisions made. There is as well a series of lessons common to many other projects we get involved in as IT professionals:
- Assigning responsibility and resources:
All Council decisions should assign clear responsibility for deliverables and allocate appropriate resources. Otherwise nothing may ever happen (In this case one of the Branches took on the project and pushed it through over 18 months.) The model of assigning one Branch to pick up a project and be responsible for it may be a good one.
- Council recognising areas of weakness:
Council is unfortunately (by the nature of the selection process) comprised mainly of management level people - who are not necessarily very technical. This as much as anything I believe caused the most delay in getting this project going. It was not until Tom Worthington from Canberra came along in 1992 with particular experience in the required technology, along with Professor Gopal Gupta from James Cook University - that there was a strong call for action. To me this indicated a failure of the Council structure which includes Technical Boards. The Board Directors should have been able to get this project on the agenda and moving much earlier.
- Working with volunteers:
Getting agreement in all States to do something nationally is not easy for several reasons - this can include (as in any organisation) personality conflicts, lack of understanding by Branches and therefore lack of support, and last but not least - the problem of Councillors not having time to focus on issues either because of other commitments or because of too many other agenda items at Council and confused priorities. To a large extent the innovation of Chairman's meetings in late 1993 and early 1994 played a big part in the Chairmen having time to get to know one another, develop confidence in one another, and just having time to talk about issues of common interest - which because of time pressures often did not happen at Council. The email project was recognised by the Chairmen as a high priority one which they were all prepared to push forward and support.
- Managing expectations
This is well known as the key to any successful project. In the ACS-link case the decision to fast- track the launch of the project made it more difficult to manage expectations as well as would have been preferred. This has resulted in some negative comments from members who expected to be on "a super-highway" but got stuck on the on-ramp.
- The impact on Pegasus
This company doubled its client base overnight - from around 3,000 to 6,000. Of greater impact was the fact that most of their 3,000 ACS-link clients were IT experts who knew what to expect, knew what questions to ask, and knew how to get their money's worth out of the system. I think Pegasus were caught off-guard by the level of usage, the number of queries and the expertise of the new users. The effect of this was that the help-desk was under-manned for a couple of months, systems (such as SLIP/PPP) were not in place as quickly as expected, and the modem network was not brought up to scratch until about February 1995.
This modem issue caused subscribers in some areas to use AUSTPAC instead. This directly affected the finances of Pegasus for a few months because they pick up the charges for AUSTPAC calls.
Where to next?
With almost daily news breaks about who is doing what with whom in the way of control and delivery of network services in Australia it is not without risk that the following predictions are made. However, they should at least contribute to debate about where we are going - even if they don't turn out to be totally accurate:
- ACS-link will not continue to grow in percentage-of-membership terms much beyond 25%. Most other members will have accounts either at work or on other services which for one reason or another are more convenient.
- ACS administrators will start using the network this year for everything from member surveys to distribution of notices. This will grow in the coming years to providing access for members to software at special rates - delivered via the network of course.
- Electronic "newsletters" will make inroads to conventional ACS publications in the next two years - but it will be more like five years before the paper ones are replaced completely.
- ACS-link will continue to exist for the foreseeable future - whether it is with Pegasus or not is not all that relevant. The ACS-Pegasus venture is looking healthy at the moment and I hope it succeeds in the long term.
- ACS is about to broaden it's base with a new range of services for new levels of membership. At the same time there will be a tightening up of entry conditions to the "Professional Division". ACS subscribers in all the Divisions will for the next year at least find the ACS-link service very attractive. It will thus be one of the important "attractants" for new members.
Note that I have avoided predictions about whether Bill Gates or some other organisation or individual will have a big impact on the network world in the near future. My view is that network technology is relatively inexpensive and easy to set up - it is the value-added bit that will make the difference for subscribers. It is therefore important for Pegasus and ACS to continue to value-add to the ACS-link service. Only in that way will it survive in the long term.
Peter Talty
23 April 1995