Background – I wrote this circa 2004, but its previous web address became extinct, so I have been asked to repost it. An interesting bit of early Australian and Asian Internet history
The Pegasus Story
Rainforest Activist becomes Connect Activist
In 1984 I was devoting a lot of time to the Rainforest Information Centre, a radical organization to protect international rainforests. In those days we were trying to get underway a fledgling international network of Rainforest Action Groups, and publishing a periodic magazine called World Rainforest Report.
In December of 1983 in Singapore I had purchased my first computer –an Apple 2 Plus lookalike known locally as a “Pineapple”. It had no hard disk (they came a few years later), a huge 64k floppy disk, a word processor called Wordstar, and no Microsoft software at all (MSDOS appeared on my horizons about a year later to replace the CPM operating system). I began experimenting with modems when a local technician re-wired a Microbee modem to work with my Apple computer.
People kept talking to me about starting a local bulletin board, but I only had one thing in mind – to connect to Econet in the United States. With no tutors, very few Australians using email and modems, and the vagaries of Austpac international connections, it took me some time to get the connection right.
The transition from an environmental activist to a connect activist had begun. It was a natural one for me; the moment I discovered the information processing power of the computer and could begin to envisage its communication power I needed to know more.
By 1985 I was well and truly underway, and when I went that year to India to a regional meeting of environmental organizations, I was very keen to see a regional network established. The United Nations Environment Program agreed, and in 1987 I had my first networking assignment – to link key environmental NGOs in Sri Lanka, Malaysia, Thailand, Indonesia, Philippines, Nepal, Hong Kong and Japan to a network.
That was some challenge – only a few of these groups were yet experimenting with personal computers. It was much the same with Australian environment groups in those days – some were just beginning to use computers, and there were no networks to connect to. Modem was not in the vocabulary.
The UNEP DDC network, as it was called, had mixed success, but it did teach me a lot about both the potential and the possibilities.
I remember for instance in 1987 we were organizing global demonstrations on October 31 to support the Penan, a rainforest dwelling tribal group in Malaysia whose homeland was being logged. These forest dwellers had taken to sitting in front of bulldozers and getting arrested, much like their Australian counterparts arguing for protection of the Daintree rainforest in North Queensland and Franklin River in Tasmania. There were some hundreds of demonstrations planned around the world to coincide with court hearings for arrested Penan. But at the last moment the court hearings were postponed, and we had to get a message out to over 100 cities quick.
You have to have lived in a pre-email world to know what email meant to us then. Without email, I would have had to sit through time zones, trying to catch people by phone, including one key organizer who was “somewhere in Europe”. It would have cost a fortune and taken 12 hours – with email it was a matter of 10 seconds and one message to the group. This was useful technology!
In 1988 I was in Costa Rica for a conference organized by the World Council of Churches for South American liberation theologians to talk about a thing called “integrity of creation” (which basically said that since God made the forests we should respect them). Radical stuff. Anyway, as I was in that part of the world, I organized my itinerary to visit Econet in San Francisco. While in Costa Rica I dropped in to visit Barry Roberts at the University of Peace and he told me about the Association for Progressive Communications, and a meeting he, Peter Gabriel, and two people who were later to be come close friends of mine, Mark Graham and Mitra (who doesn’t have a surname) had a year previously. This meeting had decided to set up an international network for non-government organizations, and was active in England and the United States. Their idea was to set up “nodes” in various countries and link them together. They already had a transatlantic network linking GreenNet (in the UK) with PeaceNet and Econet in San Francisco.
That night in my room in the La Selva forest I thought about PeaceNet, Econet and GreenNet – that gave me the PEG of my acronym. I then started to think about who wasn’t covered – Asia, South America, USSR, South Africa – There was my acronym – PEGASUS. The mythology of the flying horse as a messenger and the vision of a global network had totally captured me.
I had been thinking of moving to San Francisco to work with Econet. But during the time there it became very clear to me that what I wanted to do was to set up Australian connections to the emerging international network. APC were establishing a global network, and were doing so by porting UNIX technology which allowed Internet connections to the PCs of the day (then called the 286 chip). Before that, only expensive mini-computers could be used with emerging Internet protocols.
An Australian connection suited everyone I talked to, and I left USA quite sure what I as going to do, and with a lot of information as to what I would have to do to do it.
Starting a business from scratch
The first thing I knew I had to do to start Pegasus was to gather $150,000 – with a personal bank balance of nil, I managed to convince a few close friends to assist me to print some basic material on what I was planning and set off to tell Australia.
Those first meetings in Sydney and Melbourne still sit strongly in my memory – environment groups, aid groups, visionaries, this was new to all of us as I would crawl under desks to connect my laptop and its modem to overseas networks – nobody was too sure that they weren’t getting enormous telephone bills as I was connected to overseas networks, and I wasn’t sure I would be able to pay my petrol to the next stop!
I really thought everyone would be so excited that they would just put in some money and we would be up and going in no time. Not so, I learned there was a lot more to do – and the hardest lesson of the lot was that until I could prove that this thing would work financially there was no chance of it getting off the ground.
With great reluctance, I took on the three month task of writing a business plan. It was the best thing I could have done. The process taught me a lot, refined my thoughts considerably, and led me into contact with many people whose expertise I would need if this was going to succeed.
After three months, I finally thought I had a document that outlined the idea and how it would work. I tested it on a few people, spent the last available $20 to print it up nicely and make a few copies, and just wondered what would happen next.
Not long after that a partnership with Andrew Campbell, principal of a group called Fourth Wave Investments, was sorted out. Timetables to get underway were drawn up, and the real fun began!
We moved into the Epicentre in Byron Bay in May 1989 and in June 1989 – coincidentally at the same time as AARNET was commencing operations in Canberra – we signed up our first customers and began operations. Maxine Cole became the first employee of Pegasus. In late July 1989 Mike Jensen from The Web in Canada arrived in Australia armed with the technical equipment we could not purchase here and we scaled up in earnest.
Just to digress here – in the year in which we were planning this and negotiating with overseas colleagues only twice did I telephone overseas, for a total of five minutes, and never did I travel to negotiate agreements. It all happened by electronic mail and conferencing. I was really starting to realise what a powerful and effective business tool this was.
We decided to give Australia wide coverage at a local call dialup cost by using a Telstra service called Austpac. They billed us for use of it, and we billed our users. This enabled 3 speeds of dialup according to your modem speed – you dialed one number for 300 baud, another for 1200, and a third for 2400 baud. You can imagine what it was like with these low speeds – one two-hundredth of todays modems – but as most messages were text only in those days you got used to it.
On September 14 1989 we had out formal launch at Terania Creek in the rainforest, with a laptop computer, solar powered bus, and modem connected to a cellular phone. Some clever technician made the box for us to allow a connection from a cellular phone – and in fact we had to get Telstra to do some last minute changes to allow the connection to Austpac (the data service) from the cellular network. No-one had used data services from a cellular phone before that! The launch was covered by the local media and the ABC’s 7:30 Report.
AARNet was not yet operating in Australia when we first started, so we used to make phone calls overseas every hour or so to collect and send email, and one long off peak call each night to collect newsgroup material. This used a protocol called UUCP – or Unix Universal Communications Protocol. Overseas polling schedules were set up, users started to join, and away we went, innocents cast abroad in a brave new world.
We had a bit of fun getting it together technically. I just set off in our solar powered van with demonstration facilities to sign up Australia, and suggested to Maxine she set up billing software. Very few people could have done that, but Maxine did.
Managing the computers and networks was another issue to be dealt with, with Mike Jensen about to head off for Africa (where he established early Internet networks in a number of countries). We had a series of attempts to get someone to perform this important role, including a famous Australian radical called Albert Langer, and it was great relief a few months later when Paul Wilson joined us and began to stabilise and improve that side of our operations. Paul is still around the Internet business, at the time of writing as Director General of APNIC, one of the three global Internet registries.
Our team was expanding – Keith Stewart, Param Berg, Andrew Garton and Gert Gast all played key roles at a staff level, and Andrew Campbell, Ian Mathieson, Robert Rosen and others provided me with business advice.
How we joined AARNET is an interesting story. In the early days at one of our meetings in Canberra a person by the name of Geoff Huston sat patiently through a demonstration I was doing for environment groups and came up to talk to me afterwards. We were at that time connecting through AARNET’s predecessor, ACSNet. Geoff wanted us to take on a permanent connection via AARNET as soon as it got underway, but there were a few rules we had to get around to get access through AARNET and indeed to the Internet, which did not allow commercial traffic in those days. We sneaked in via sponsorship by University of New England and Melbourne University, and our first permanent connection was underway. Shortly after with a facility called telnet, logging into computers outside Australia became a reality. Geoff Huston did much to foster the early growth of the Internet in Australia, and to encourage businesses such as Pegasus to get underway.
Our attitude from the beginning was that, although we wanted affordable rates, we were going to do so with a quality service. We reasoned that anything less than a quality service would eventually see the larger and more wealthy users leave us for a more sophisticated service – and that that was in neither our financial interests or in line with our higher aspirations for the organization. We therefore concentrated funds quite heavily on improving our reliability and stability, as well as introducing new and more user- friendly facilities. Later on that quality of service broke down as Pegasus became squeezed by its competition, and I believe that that more than anything let to its demise.
At the end of our third year, we returned a small profit, which was re-invested in plans for expanded activities. We had close on 2000 users, a substantial bank loan, a full time staff of 8 people, and an exciting future. We were beginning to fulfil many people’s dreams for use of this media. At this point I stood down as Chief Executive Officer, and Paul Wilson took over. This enabled me to concentrate on forward development of the network.
We had already established strong activist networks, and the pattern of Australia wide networks with global links was what was impressing people. These later became known as virtual communities – people who would act like communities in the old geographical sense in terms of regular association, but who were separated by distance.
It was probably the only sensible way to bring a lot of people on board in those early days. Now, everyone wants email because everyone else has it, but in those days we had to find people that other people would want to email. So to associations and movements, “virtual communities” made a lot of sense. Some of the pioneering ones were
Earthnet, linking groups working on peace, human rights, environment and development aid issues
Landcarenet, linking rural based groups working on land degradation
Councilnet, launched in October 1992 to link municipal government employees and elected representatives
Artsnet, which also had connections to some early overseas arts networks
AOLIN, the network of the Open Learning movement
ACSLink, the network of the Australian Computer Society (this did not start till September 1994 – it was interesting that the computer professionals were actually not early adopters).
There were other networks, many of which I guess we never knew about anyway. Networks were forming in the ether we were beginning to call cyberspace and new groups were being established each day.
Most of these virtual communities used what we called “conferencing” software. This software could be configured for a number of uses, but the primary ones were public or private discussion rooms, which were very popular and informal, or information databases, which could be indexed to provide links to a wide range of information sources (something like a text only world wide web page with index links).
It wasn’t easy to use networks in those days. Typing on line was awful – people had all sorts of problems with simple things like getting backspace keys to work. You had to set a terminal type for your computer to get the characters to appear properly – modems had to be programmed for the settings of the network – it really was pretty difficult to use those early networks, and training and user support became a big part of out work.
It says a lot for how much people wanted to network that they actually put up with those interfaces.
Right from the beginning we were doing a global network and had gathered a series of global connections from the Electronic Networkers Association (ENA) which preceded the Internet Society as the forum for network growth. From the beginning Pegasus staff were involved across Asia with non-government organisations, and offered discounts to Asian NGOs to help get them on line. From what I know, Pegasus had a big hand in establishment of early Internet networks in Thailand, Malaysia, Indonesia, many of the Pacific Islands, Mongolia, and probably a few other countries as well. We were actively involved in the UNDP plans to establish networks in South America, and in 1993 I spoke at the first gathering of the Internet Society about the concept of “sustainable development networks”.
In 1990 Pegasus formed an alliance with various people active in NGO development in the Asian Pacific Region called Pactok. Pactok was originally a Fidonet gateway – in those days Fidonet was on PCs, but software to connect to the Internet wasn’t generally available. So Pegasus allowed transfer from Fidonet nodes across Asia Pacific to Internet.
By 1992 Pegasus had subscribers dialing in from Japan, India, Thailand, Indonesia, Malaysia, New Zealand, Fiji, Solomon Islands, and Papua New Guinea.
Later, Pegasus became involved through a relationship with CIDA (Canadian Aid) in establishment of networks in Cambodia and Mongolia.
The move to Brisbane and the last four years
By about the end of 1992 it became apparent that Byron Bay posed a few problems for us. Apart from the surf and the laid-back atmosphere, we suffered from unreliable telecommunications and unreliable electricity. After having fried a few computers and lost a few customers, we knew that we had to provide a more reliable service for our growing customer base.
So not without some misgivings, we set out and relocated in Brisbane, first in Fortitude Valley and later when bigger premises became necessary again in Southbank.
The growth of competition
In the early days, Pegasus was virtually on its own in the Internet Service Provider field, and there was a time when Pegasus had 85% of the Internet market in Australia. But not for long – once a market had opened up other players entered.
Some of the significant early players were Connect.com, started by Hugh Irvine in 1992, which more than any other organisation opened up the business market in Australia, Magnadata, and Ozemail, (which from the beginning had lots of cash). These were all pre World Wide Web.
By the time the www started to take off in Australia in 1995, lots of people were entering this field. Personal computer penetration was rapid, Microsoft had finally adopted Windows software, and modems were now 14400 or 28800 bps – getting pretty quick! All of these factors made for some very rapid growth, and soon there were about 800 Internet Service Providers in Australia. An industry had been spawned.
Many of these were small and regional, only a few maintained Australia wide presence.
During this period Pegasus grew to having 30 staff and 12000 customers, quite a significant business. That growth was very rapid, and Pegasus had a lot of trouble maintaining market share and profile as the field got larger. Pegasus just didn’t keep pace and soon became a target for a takeover.
So in many ways it was probably a relief that a buyer for Pegasus was found in the form of Microplex. Microplex at the time brought out a number of the twelve or so biggest ISPs and eventually sold them all to Optus, a telecommunications carrier. Within a short period of time, all of the major ISPs were brought out by telcos. The end of the electronic frontier was nigh – and the era of the electropolis and the cybermall had begun.
I would probably put the demise of Pegasus down to avoidable lack of capitalization, accompanied by some lack of vision. The capitalization was a continual frustration when growth was needed, investors willing to pay a fair price for shares were available, but the principal shareholder didn’t want to dilute his holding. Thus while others were expanding rapidly, Pegasus started to perform poorly. Just after bringing on board the Australian Computer Society’s national membership, a real coup that could lead to further growth, Pegasus started to suffer from clogged bandwidth and poor system performance. The people who could have helped so much in a next stage of expansion started to leave. The media who had supported Pegasus to that point of time started to compare us unfavourably with others. And we turned down further investment in favour of an attempt at immediate profits.
I came out of Pegasus a little disillusioned, with very little money, but it was a great experience and a fantastic thing to do. To hold memories of the growth of the Internet in its early stages, and to be one of a handful of people who can honestly say “we started the Internet industry in this country” is absolutely terrific.