The following are extracts from the Victorian Government's WWW pages.
'The communications and multimedia future is coming very quickly, and with it enormous possibilities for change. The task for the Victorian community will be to ensure that we have the infrastructure, the regulatory environment, the content and the capabilities to realise enduring benefit for Victoria.'
'Victoria has the capacity to be a winner in this new global economy. We have creative and knowledgeable people and enterprises working at the leading edge of technology. We have strong research and education institutions, and a rich, diverse cultural heritage. These provide the building blocks for success in the information age. But success is not guaranteed. Victoria's industry, employment, revenues and cultural identity will all be exposed to new threats if government, business and community institutions do not rise to the challenge.'
'While the opportunities are considerable, an agenda for the development of the communication and multimedia industries must tackle a number of significant constraints including:
It is concluded that the Victorian State Government recognises the importance of IT, and sees
as worthwhile objectives.
The Ubiquitous Public Network is an element of communications infrastructure designed to help achieve these objectives.
When considering allocation of the cost of information technology (IT) it must be recognised that IT differs from other products in three important respects. Firstly, an IT product such as data or software can generally be duplicated essentially without cost to the originator. Secondly, the value of an IT product frequently depends on how many people use that product. Thirdly, users of IT frequently make contributions to the technology, the content or both.
Recognition of these differences by technology organisations such as Sun, Netscape and Microsoft is evident in their marketing and charging strategies. Simplistic "user pays" methods are inadequate if optimum outcomes are to be achieved.
Two aspects of cost are assumed.
Cost is typically made up of a fixed periodic cost plus a usage-based cost. However, if usage-based cost is negligible, then users have a powerful incentive to use that service as intensively and creatively as possible. This results in a strong positive-feedback effect. The concentration of utilisation and creativity enhances the usefulness of the service, which in turn attracts more use and creativity.
The dramatic success of the PC is an example of this effect. The very limited penetration and slow growth of pay-by-use on-line services shows the converse effect.
It is concluded that a low cost, zero-marginal-cost communications infrastructure would result in the fastest growth and the most useful information technology structure.
Agreed standards are essential for effective utilisation and progress. Standards take time to develop. The most efficient shortcut to standards development is to utilise existing public-domain standards wherever possible.
The adoption of Internet standards in corporate Intranets is an example of the very rapid implementation possible from use of existing standards.
It is concluded that a communications infrastructure should be based wherever possible on existing public-domain standards.
Use of architectures may be either restricted by the developer, or made freely available for use by all comers. Freely available open architectures enjoy more rapid acceptance and development than closed architectures. One reason is that any manufacturer may produce any product for use anywhere within an open architecture, so competition results in lower prices and hence a larger market share for the technology. Another reason is that many users have suffered costly consequences from adopting proprietry technologies.
An example of open architecture's almost inevitable dominance can be seen by comparing the history of the Apple and IBM-compatible computers. Any organisation may manufacture IBM-compatible PCs. Only Apple licencees may manufacture Apple-compatible PCs. Despite Apple's initial technological lead, Apple's market share is now small compared to the IBM-compatible. Other examples of the success of open architecture are the Internet and UNIX. Examples of failures of closed architectures abound, and include offerings from the largest hardware and software companies in the world. The histories of IBM's Microchannel architecture, and more recently Microsoft's MSN, are enlightening.
Access may be limited to certain points in the architecture, either by the deliberate actions of a system owner or because the system becomes vulnerable if access is unrestricted. Irrespective of the reason, the result is that the service and its cost structure are controlled by a limited number of organisations, which prevents real competition despite the best efforts of regulators.
It is concluded that a communications infrastructure should be based on open architecture, access and protocols so that any part of the infrastructure can be provided by anyone, without restrictions.
Let us examine the indicators for Australia's current direction in communications infrastructure.
The following is an extract from the Victorian Government's WWW pages.
'Telstra and Optus Communications are spending $A6 billion to roll cable
past 70% of Australian homes by the end of 1998.
The new networks will provide fast, reliable digital switching and transmission
technologies. Synchronous digital hierarchy (SDH) transmission with asynchronous
transfer mode (ATM) switching is being deployed.
When combined with the use of Satellite and MDS technology that currently
provides mobile telephony, Pay TV services and a number of interactive
applications, Australia will soon have in place a multimedia support
infrastructure providing delivery and access to over 95% of households.'
Depending on the information source used, this appears to translate to some $1000- $1500 per household, and very much more per connected household. The telcos will need to recover this cost in their charges, along with their operating costs and profit. This gives some idea of the likely charges. It would seem that the charges are unlikely to be small.
All the services known to the author which are currently provided by these two telcos include a charging element based on usage. There does not appear to be a zero-marginal-cost option for the majority of ordinary users of the communications services which are or will be provided by these telcos. While ostensibly commercially sensible, this is a negative precursor.
It appears that our current direction will be towards neither low cost nor zero marginal cost, and so will not result in fast growth nor the most useful information technology structure.
Standards are not considered to be an issue with Australia's current direction.
Access to the telcos new networks will be restricted in that an equipment owner may not incorporate his equipment at any point he desires in the telcos' new cable networks. Real competition will not be possible regardless of the best efforts of regulators, and prices will reflect this.
It appears that we are headed towards a high-cost solution, which will inhibit use and creativity.
After June 1997, telecommunications in Australia will be governed by the Telecommunications Act 1996. The Act states:
'The main object of this Act, when read together with Parts XIB and XIC of the Trade Practices Act 1974, is to provide a regulatory framework that promotes:
(a) the long-term interests of end-users of carriage services or of services provided by means of carriage services; and
(b) the efficiency and international competitiveness of the Australian telecommunications industry.'
However, in reality the legislation appears to impose fees and restrictions which limit the field to the larger players, and thereby results in unnecessary costs for the users of telecommunications services. In particular, there appear to be constraints which would severely limit the sort of inventiveness and creativity which have characterised the most successful IT products yet developed, the PC and the Internet. The main problem with the legislation is that it directly or indirectly imposes costs on owners of "network units". The definition of a "network unit" is sufficiently broad to encompass almost any communications link between "distinct places in Australia".
While the stated objects of the legislation are admirable, its effects will be to unnecessarily increase the cost of IT to Australians, and to prevent us from utilising and developing optimum technologies.
Our current direction will neither result in low-cost communication, nor rapid development nor high utilisation of our communications infrastructure. It appears to be based on an economic model which, while appropriate for marketing wool or tourism, has historically yielded losses or dramatically diminished returns when used for IT.
We ignore the lessons of history to our detriment.
The value of IT is in the information content. However, without appropriate communications infrastructure, this content is of little value. This proposal relates only to this communications infrastructure.
A publicly owned and operated data communications network is proposed which would address some of the constraints to IT identified by the Victorian Government. This Ubiquitous Public Network (UPN) would result in
The concept can be thought of as a "networked city", where data can flow between any two connected nodes without marginal costs.
This UPN proposal has characteristics of the two outstandingly successful IT technologies of this century, the PC and the Internet. These characteristics are low, usage-independent cost, open architecture and open access.
This UPN would be publicly owned in that the individually-owned nodes would comprise the only hardware necessary for the network. All the network specifications, interface specifications and software would be public domain, and access to the network would not be controlled by a governing body or organisation.
While the Telecommunications Act 1996 effectively prevents the use of the optimum technologies for a public network, it appears that data links using visible light may not be encumbered by fees and conditions.
Participating households and companies would each own their own UPN node, which would comprise a controller (based on a PC) and optical link hardware linking the node to nearby nodes. Data would be transferred between geographically separated nodes in multiple "hops", in much the same way as data is transferred on the Internet.
For a fixed annual cost of around $200 p.a. for each node owner, a UPN could:
The UPN as proposed could not support high speed, low latency applications such as real-time video or voice. Further developments of the UPN may be able to support these services, although this seems unlikely in the foreseeable future.
For more details, see http://www.vicnet.net.au/~upnet/
derekw@alphalink.com.au
13 Jan 1997