Standards to Support National Cooperation in Applying Technology to VET

[Previous section] [Next section]

The Spectrum of Standards

Analysis of standards in the telecommunications and information services industries reveals a range of standards which have different levels of 'openness' in terms of the characteristics in Section 3.6. A standard may be ranked by the response to a series of test questions, as follows:

Q1 - Is it a mandatory standard? For instance AUSTEL standards which had the force of law. AUSTEL has now been replaced by the ACA (Australian Communications Authority), which is working on telecommunications technical standards with the ACIF, the Australian Communications Industry Forum. These standards are tabled in parliament as disallowable instruments and have the force of an act of parliament. These standards control what customer equipment can be connected to the Public Switched Telephone Network. (AUSTEL standards resulted from a process similar to that of Standards Australia in Q3 below, but under AUSTEL, rather than Standards Australia. The new arrangements are still being finalised, but will be under the new industry body the ACIF, with ACA involvement mainly to give them legislative force.)

Q2 - Is the standard in a field which would normally be covered by the above standards, but is not an ACA/ACIF mandatory standard, but some other standard which the carrier specifies equipment must meet in order to connect to its network. (For instance particular phones or cable modems which do not meet any applicable ACA/ACIF mandatory standards, but which the carrier is happy to connect.)

In either case, there is no choice for VET or any other users of the telecommunications service which uses these standards; they must comply with the requirements of the law and the carrier who provides the service.

The remaining types of standards are those which VET is likely choice between:

Q3 - Is it a 'formal' standard? Did the standard result from an open, accessible, consensus-based public process, under the auspices of an 'official' standards body such as Standards Australia, as described in: http://www.standards.com.au/Information-Sheets/creatstan/createstan.htm

Note that such standards are by default voluntary standards. For instance if there is a Standards Australia standard for the design of shopping trolleys, then that doesn't mean that all shopping trolleys manufactured, sold and used in Australia must meet that design. Sometimes, these standards are given the force of local, state or federal law to set safety standards, for instance for child car seats, bicycle and motorcycle safety helmets etc.

Q4 - Is it a 'quasi-formal' standard? There is a second level of standards making bodies, which are not formal, internationally accredited bodies, but which do have open processes within or beyond their own membership? and which have the expertise and the clout to create IT technical standards of lasting value. For instance the Institute of Electrical and Electronic Engineers, whose standards making work is described at: http://standards.ieee.org/sa/index.html

IEEE committee 802 has provided LAN standards of great significance; for instance 802.3 is the formalised (and somewhat changed) standard of what used to be the proprietary Ethernet standard.

Other examples are the POSIX standard for application programs to interface to (typically) Unix operating systems and the C and C++ standardisation efforts at ANSI.

Sometimes these standards are adopted wholly or with modification by the international standards bodies; for instance IEEE 802.3 (as it is usually known) was adopted by ISO and called ISO 802.3.

The W3C and Internet Engineering Task Force produce standards which might be called, 'quasi formal'; they are not formal standards bodies, but their work is almost universally respected and their processes resemble the formal standards bodies in their openness.

Q5 - If the standard is published, but is not a formal or quasi-formal standard as described above, did it result from the work of volunteers, a large or a small consortium of companies or a single company? The case of a large consortium of companies is similar to the 'quasi formal' standards bodies mentioned above.

Note that some of these standards (or 'Publicly Available Specifications' as the formal standards bodies refer to them) might be deemed suitable for adoption by ISO / IEC according to criteria set out in: http://www.iso.ch/dire/jtc1/pas.html

Some standards which result from these processes involve licensing fees in order to implement the technology.

a - The work of volunteers, who publish their technical standards and often source code, but who do not necessarily meet the criteria of openness and public review which official standards bodies aim for. For instance PGP, PGP-Phone, the Linux operating system. This is particularly prevalent in the software field, where cooperative development can result in a software system which is widely adopted and which embodies certain protocols, interfaces and file formats which are effectively standards.

b - The work of a larger number of companies in a relatively open consortium of companies, but the process falls short of a public process, because of the costs and other barriers to involvement by any interested person or company. A good example is the DAVIC's (200 companies) work on digital video and to a lesser extent cable modems. Other examples might be the ATM Forum, the ADSL Forum and VESA for video display standards.

These produce published standards, but which may or may not involve licensing fees to implement the required technology.
http://www.davic.org/WHATIS.htm

c - The work of an consortium of a few companies working closely together.

This means it is not owned by a single company, but by a relatively small group of companies operating for mutual benefit. For instance Philips and Sony's work on the compact disc, which is a published standard which involves licence fees. For instance Cablelabs' work on cable modems, which will be published, but may or may not involve licence fees. Similarly the MCNS work on cable modems:
http://www.cablemodem.com/

which is results from work by four main companies, plus Cablelabs (which represents many cable TV operators, and so could be called a quasi-formal standards body).

d - The work of one company, e.g. Adobe's Postscript and PDF, or to-date Sun's Java.

This means that there is no public or external corporate input into the current state of the standard or to its future development. Another example would be the interface of application programs to proprietary operating systems such as the operating systems MS-DOS, Windows and Macintosh.

Q6 - Non published proprietary standards Where Q1, Q3, Q4 and Q5 do not apply: the technical details of the 'standard' are not published, and are presumably the product of the work of a single company. Most prominent is the file format for Microsoft Word, Corel Draw, Adobe Illustrator etc. Similarly the protocol and encoding details of the various proprietary voice, conferencing and groupware programs.

Q7 - Ad hoc 'standards' Where none of the above apply: there is no actual published standard, but more of a common understanding amongst designers of how to do things. An example might be the layout of the PC keyboard the QUERTY 'standard' dates back a century or so and was deliberately designed to slow down typists). Mac and Unix workstations substantially follow the PC layout.

The set of features and physical layout of the PC motherboard is probably an example of such a process of accretion of various functions in a way which resembles a standard, although probably no-one has ever written it all down and published it as such.

The things we are looking for in VET centre on:

1 - Interoperability between products from different vendors.
2 - Competition between multiple vendors to lower prices.
3 - The avoidance of marketing-led distortions in perceptions caused by one company having a financial incentive to push its proprietary product as the answer to every need.
4 - Relative stability of the standard,
5 - An evolution of the standard which is likely to suit the broad public interest.

Standards at the various levels map to the above properties in the following way:

[Top of page] [Next section] [Contents page]

[Home page] [Copyright statement] [Feedback]

Last modified on February 25, 1998.