Hoodwinking the Public:
Australia's Internet Censorship Regime

Last Updated: 6 November 2002


This paper reviews the operation and effectiveness of the Australian Government's Internet censorship regime which commenced on 1 January 2000.

Contents:


Summary

  • There is no evidence or indication in Government reports to support the Minister's claim on 21 August 2002 that the Internet has been made safer as a result of the Government's Internet censorship regime.
  • The ABA spent 83% of its Internet censorship efforts investigating content on overseas-hosted websites over which it has no control.
  • Approximately half of the prohibited items designated as hosted in Australia were found in world-wide Usenet newsgroups, most likely originated outside Australia, and were not taken down from the Internet.
  • The ABA's refusal to provide the URLs or titles of taken-down Australian-hosted web pages, on the ground that such information would enable a person to access prohibited content on the Internet, indicates the ABA believes such content has not been taken down from the Internet.
  • Ministerial statements trumpeting the success of the scheme have been, by the Minister's own admission, based on erroneous statistics.
  • Misleading statements have been made by the government about the proportion of prohibited content that is actual child pornography.
  • The scheme exaggerates the outcomes by claiming newsgroup postings removed from one Usenet newsgroup server as content that has been removed from the Internet.
  • The referral of prohibited content to scheduled filter vendors is not followed up to ensure that the vendors add the content to their filter blocklist.
  • The effectiveness or otherwise of the complaints system would be clearer if the outcome of investigations resulting from legislatively valid complaints (i.e. from Australian residents), and information received from other entities such as overseas hotlines, was reported on separately.
  • No information has been made available by the government about successful prosecutions, if any, resulting from the scheme.
  • The estimated $2.7M annual cost of the scheme is difficult to justify given the limited outcomes achieved.

Up ArrowGo to Contents List


Introduction

In releasing the fourth report on the government's Internet censorship regime on 21 August 2002, the Minister for Communications Senator Richard Alston proclaimed: "Internet safety for Australians continues to grow".

In this paper EFA analyses government reports on the regime, in light of the Minister's admissions to the Senate that official reports contain statistical errors exaggerating the alleged effectiveness of the scheme, and reviews the overall operation and effectiveness of the scheme.

We conclude there is no evidence to support the Minister's claim that the Internet has been made safer.

While the Australian government will no doubt continue attempting to hoodwink the public, EFA hopes that the increasing number of Australian adults who use the Internet is resulting in less parents being lulled into a false sense of security by the government's claims. The government cannot make the global Internet safe, or even 'safer', for children.

Up ArrowGo to Contents List


Location of Prohibited Content

Australia -vs- Overseas

During the two years ended December 2001, the Australian Broadcasting Authority ("ABA") prohibited 756 items of Internet content. These items consisted of pages on the World Wide Web and messages posted to Usenet newsgroups. The World Wide Web contained an estimated 800 million web pages in February 1999 and the growth rate has since been explosive. Today, the popular search engine Google provides Internet users with immediate access to over 3 billion Web documents. There are approximately 40,000 Usenet newsgroups. As well, Google provides access to a 20 year Usenet archive consisting of more than 700 million messages.

The ABA spent 83% of its Internet censorship efforts investigating content on overseas-hosted websites over which it has no control.

The remaining 17% of the ABA's censorship efforts were directed to investigating content deemed "Australian". As defined by the relevant law, this includes not only content on Australian-hosted web sites, but also messages posted by people all over the world to the global Usenet newsgroup network.

The graphs below illustrate the number and outcome of investigations concerning "Australian" and overseas content.

Graph: Investigations - Australian Content Graph: Investigations - Overseas Content

Content was prohibited as a result of 68% of the investigations into overseas-hosted content, and 43% of the investigations into Australian-hosted content.

Australian content: web sites -vs- global newsgroups

Approximately half of the "Australian" prohibited items were found in world-wide Usenet newsgroups, quite probably distributed by persons outside Australia. This situation is illustrated in the graphs below for the year ended June 2001.

Graph: Complaints Investgated 2001 Graph: Items Prohibited 2001

(Note: Information tabled in the Senate suggests the percentage of prohibited items found in newsgroups may be higher than shown in the above graph. Newsgroup items may comprise around 15% and Australian Web content 10%).

As shown above, investigations into complaints about "Australian" content locate proportionally more prohibited items per complaint than investigations into overseas content. Although only 6% of the complaints about prohibited content concerned "Australian" content, these resulted in finding of 25% of the prohibited items. The reason for this is provided in the following section.

Up ArrowGo to Contents List


The Exploding Statistics Phenomenon

In EFA's May 2001 analysis (Regulatory Failure: Australia's Internet Censorship Regime) of the second government report, we noted an exploding statistics phenomenon. Although the ABA's Internet Content Complaint Form states "Note: You can complain about only one item of Internet content in each complaint form", the ABA had reported 67 items of prohibited content from the 6 complaints about Australian-hosted content, in contrast to only 136 items from 133 complaints about overseas hosted content. We remarked that "it could be concluded that the ABA is ramping up the 'item count' in order to make the figures look respectable in relation to Australian content".

Subsequently, it has become clear that a significant part of the reason for this phenomenon is the treatment of Usenet newsgroups as Australian content, although the vast majority of messages in newsgroups are not posted by Australians.

Usenet is a world-wide network of public discussion and exchange forums called newsgroups. There are tens of thousands of newsgroups covering a huge range of topics, and hundreds of thousands of messages are posted to newsgroups each day. New newsgroups can be quickly created and are automatically disseminated around the world using ad hoc, peer to peer connections between computers, called Usenet servers. Many Internet Service Providers (ISPs) host Usenet servers and some web sites provide archived copies of messages distributed via Usenet servers. When a person somewhere in the world posts a message to a newsgroup, the message is automatically forwarded to and temporarily stored on Usenet servers all over the world, where it can be read and/or replied to by other people. The identity and origin of messages can be easily forged. Messages are periodically and automatically purged from Usenet servers to make room for new messages.

According to the Minister for Communications, Senator Richard Alston (in answers to questions on notice asked by Senator Brian Greig):

"[W]here a complaint relates to an entire newsgroup, rather than a single posting on it, the ABA investigates a sample of the postings contained in the newsgroup... For the purpose of reporting, each of the postings sampled is then counted as an item in the statistics. Similarly, the ABA may investigate a sample of the content on a World Wide Web site...and each page would be counted as an item in the statistics. However, investigations relating to complaints about World Wide Web content are less likely to involve more than one item of content, as complaints relating to such content generally pertain to a specific page of content."

and

"Three of the six complaints [in the Jul-Dec 2000 period] about Australian hosted content related to Internet content in Usenet newsgroups...All complaints about content hosted outside Australia related to World Wide Web content."

Hence, the reported number of 'Australian-hosted' items investigated and/or prohibited is meaningless. The numbers can readily be manipulated as they depend on the amount of time that ABA staff choose to spend searching through the thousands of messages in newsgroups.

For example, the second government report (Jul-Dec 2000) stated that the ABA had referred 45 of 67 Australian-hosted prohibited items to State or Territory police. EFA's analysis of the report remarked that "at least 14 of the 45 are likely to be newsgroup posts found in one newsgroup on one ISP's system".

The Minister has since revealed statistics stating that 64 (not 67) items were prohibited and 28 of 44 (not 45) items referred to State or Territory police were newsgroup postings. Furthermore, the statistics show that approx. 14 of 22 prohibited items not referred to police would have been found by the ABA while sampling posts in newsgroups. In summary, 28 of 44 'Australian-hosted' items referred to police were newsgroup postings and at least 42 (66%) of the total of 64 'Australian-hosted' prohibited items were newsgroup postings.

Similarly, in the third six months Jan-Jul 2001, 17 of 34 (50%) of the 'Australian-hosted' prohibited items were newsgroup postings according to the Minister. Relevant statistics for the first and fourth six-month periods have not been made available by government agencies or the Minister to EFA's knowledge.

While the vast majority of messages posted to newsgroups contain innocuous information, there is no doubt that a considerable amount of highly illegal material is also disseminated via newsgroups. However, counting newsgroup postings as "Australian" content and referring these to State/Territory police is unlikely to be an effective means of reducing the amount of illegal content distributed via newsgroups, as State/Territory police powers are limited to their own jurisdictions and most newsgroup postings originate overseas. Government reports indicate only overseas-hosted web site content is referred to the Australian Federal Police (AFP) for referral to overseas law enforcement agencies.

Up ArrowGo to Contents List


Type of Prohibited Content

Errors in Government Reported Statistics

The ABA and Minister have frequently claimed that the majority of items prohibited involve child pornography. While this might be so, the numbers concerning such material have been exaggerated in government reports. Moreover, a larger quantity of material that is legally available to adults offline has been prohibited online than reported. Some of this material has been counted and reported as child pornography although it is not. Furthermore, these false claims have in effect been used to support refusal to provide information under Freedom of Information law about the type of content being prohibited.

In EFA's May 2001 analysis, we questioned the accuracy of information in the second government report covering the six months ended December 2000.

Subsequently the Minister for Communications, Senator Richard Alston informed the Senate that numbers in the second (and third) government reports were wrong. Information in this regard is below.

Number of prohibited 'Australian-hosted' items

According to the Minister, less items of Australian-hosted content were prohibited than claimed in the second and third government reports and the numbers reported in each classification category were also wrong, as detailed below.

Table 2: Jan-Jun 2001
Number of items of 'Australian-hosted' prohibited content
Graph

Six-month Report
issued
Feb 2002
Minister's Answers
tabled
Jun 2002
R18 6 7
X18 3 10
RC 28 17
Total 37 34

 

Table 3: Jul-Dec 2000
Number of items of 'Australian-hosted' prohibited content
Graph

Six-month Report
issued
Apr 2001
Minister's Answers
tabled
Aug 2001
R18 5 11
X18 12 9
RC 50 44
Total 67 64


R18 prohibited content

As shown above, almost twice as much content was classified R18 than reported in the official six-month reports. Approximately half the items classified R18 (detailed information about aspects of adult life that may be disturbing or harmful to minors) were apparently reported in the same category as child pornography instead of in the R18 category.

Some of the material that was classified R18 and prohibited is equivalent to material legally available in magazines classified "Unrestricted", that is, in magazines that can be legally sold in Australia to persons under 18 years. This occurs because the regime applies film censorship guidelines to static images and text, instead of publications guidelines. As well, when a web page displays only one image that infringes the law, and many that do not, the entire page is prohibited. Unlike offline magazine publishers, online publishers are not notified of which specific images or text infringe the law and given the opportunity to remove those. Instead, the entire web page is banned.

X18 prohibited content

Material classified X18 (non-violent sexually explicit material depicting consenting adults) has also been incorrectly counted in the number of items reported to involve child pornography.

In EFA's analysis of the second six-month report (Jul-Dec 2000), we remarked that:

"It is reasonable to assume that the items referred to police concern child pornography, particularly given that the ABA's breakdown of items by category supports this contention, e.g. takedown notices were issued in relation to 45 items that were categorised as either 'exploitative/offensive depiction of a child' or 'paedophile activity'."

However, it has since become apparent that it is not reasonable to make such an assumption.

For example, the third six month report claimed that 23 Australian-hosted items classified 'RC' were referred to State/Territory police because they involved "exploitative/offensive depiction of a child".

However, the Minister subsequently informed the Senate the correct number was 13 (not 23) and that, in addition, 10 items classified 'X18' (non-violent sexually explicit material depicting consenting adults) were referred to State/Territory police.

Whether in fact the 10 items classified 'X18' were referred to police is questionable. If they were then, according to six-month reports and the Minister's answers to questions, 57 or 58 items classified 'RC' plus 10 items classified 'X' were referred to State/Territory police in the year ended June 2001, i.e. a total of 67 or 68 items. However, the ABA Annual Report 2001 states only 59 items were referred to State/Territory police.

The purpose of referring items classified 'X' to State/Territory police is unclear given it is not illegal to make 'X' content available to adults online under State/Territory laws, except under Victorian law (as at October 2002). Perhaps all four of the complaints about 'X' material in the six months coincidentally happened to concern content hosted in Victoria. However, referring 'X' material to police in States/Territories where it is not illegal to make available, and incorrectly reporting it as 'RC', is a means of ramping up the item count so that it appears more Australian content involving child pornography is being found and referred to police than is factual.

Number of items involving child pornography referred to police

As noted above, the number of items allegedly referred to State/Territory police because they involved "exploitative/offensive depiction of a child" has been exaggerated in government reports. However, the correct number remains unclear.

In relation to 'Australian-hosted', three different numbers have been claimed in government reports for the year ended June 2001:

  • the two six-month reports state 68 items involving "exploitative/offensive depiction of a child" were referred to State/Territory police,
  • the Minister's answers state the number was 57, and
  • the ABA's Annual Report 2001 states the number was 59,
suggesting the six-month reports exaggerated the number by approximately 16%.

With regard to content hosted outside Australia:

  • the two six-month reports state 209 items were referred to the Australian Federal Police,
  • the ABA's Annual Report states the number was 176,
suggesting the six-month reports exaggerated the number by approximately 19%.

Further exaggeration of the number of items categorised as "exploitative/offensive depiction of a child or paedophile activity" is found in the DCITA Review Issues Paper which states: "Of the items Refused Classification during the first 24 months of the Scheme's operation, 492 involved exploitative/offensive depiction of a child or paedophile activity. This represents 59 per cent of the items Refused Classification under the Scheme".

The number 492 bears no resemblance to the numbers previously claimed in government reports, nor in the Minister's answers to questions. According to the four six-month reports, the number was 432, not 492. However, according to the Minister's answers to questions on notice, the numbers in at least two of the four reports were wrong and, at most, 421 items involved exploitative/offensive depiction of a child or paedophile activity.

Furthermore, if 492 represents 59 per cent of the items Refused Classification ("RC") as stated in the Issues Paper, then a total of 833 items must have been categorised RC. However, only 756 items were prohibited according to the Issues Paper and at least 150 of the 756 prohibited items were not classified RC.

In relation to accuracy of the number of items allegedly referred to police, the Minister was asked in the Senate why the ABA informed a Senate Estimates Committee (on 30 November 2000) that 91 items had been referred to State or Territory police in the eleven months to 30 November 2000, while the official reports state less items (89) for the full calendar year. The Minister answered (on 8 August 2001):

"The discrepancy appears to relate to misclassification and/or double-counting of some items in the ABA's complaint database at the time the November 2000 statistic was calculated. The complaint data are regularly reviewed for accuracy and the inaccuracy was corrected prior to calculation of the December 2000 statistic. ..."

Nevertheless "misclassification and/or double-counting", or other errors, in relation to statistics for period ended December 2000 was still occurring in 2001 as shown in Table 3 above. Evidently, misclassification and/or double-counting continued to occur in reports concerning the 2001 year and in preparing the DCITA Review Issues Paper in 2002.

Errors and discrepancies in governmental statistical reporting undermine the credibility of the government's claims regarding the number and type of items located and prohibited. Moreover, as the errors noted above came to light from questions asked about only Australian-hosted content in one year, it seems likely there are also errors in reports for other periods and also in relation to overseas hosted content.

Although there has been much governmental fanfare implying the legislation is effective in reducing the quantity of child pornography material on the Internet, there is no support for this contention in government reports. Such material was already illegal to possess and distribute, both offline and online, under pre-existing law in all Australian States and Territories. Reporting on numbers of items referred to police, even an exaggerated number, does not demonstrate effectiveness. Notably, government reports contain no information on the outcome of police investigations, if any, regarding content referred to police by the ABA.

Distribution of child sexual abuse images via the Internet is a serious problem that requires sufficient funding of police forces to enable specialised training and adequate resources. EFA considers funds used in locating and categorising merely controversial speech as "prohibited" would be more usefully directed to increased funding of police engaged in tracking down those who distribute globally agreed illegal material such as child sexual abuse images.

Up ArrowGo to Contents List


Effectiveness of the Scheme

Take-down Notices: Australian hosted content

As defined in the Act, 'Australian-hosted' content includes content on web sites hosted in Australia and also messages posted by people all over the world to the global Usenet newsgroup network.

In relation to Web site content, although government reports claim that prohibited 'Australian-hosted' content found by the ABA has been taken-down, both the government and the ABA believe the prohibited content remains on the Web. This is demonstrated by their refusal to provide the URLs or titles of prohibited Australian-hosted Web pages under Freedom of Information law on the ground that the information would enable a person to access prohibited content. Obviously, if the content had been taken-down from the Web, the URL would not facilitate access to it.

In relation to prohibited content in global Usenet newsgroups, the ABA only sends a take-down notice to one ISP, that is, to the complainant's ISP. This is evident from documents released to EFA under Freedom of Information law and also in the Minister's answers to questions in the Senate. However, as discussed above, messages posted to newsgroups are automatically distributed to many ISPs' news servers. Hence, the same content remains available on news servers of numerous other Australian ISPs (there are approximately 650 ISPs in Australia). Nevertheless these items, which account for approx. 50% of "Australian-hosted prohibited content" in the year ended June 2001 are counted and reported as having been taken-down from Australian servers.

Referral to Filter Makers: overseas hosted content

As noted above, around 83% of the ABA's Internet censorship efforts have been directed to investigating content hosted on overseas web sites. Prohibited items are then notified by the ABA to approved filtering software vendors listed in the Internet Industry Association (IIA) Code of Practice. ISPs are required to provide an approved filtering product to any customers who wish to obtain/use same, at a price that does not include profit for the ISP.

However, neither the ABA or IIA are required to verify that approved vendors update their censorware blocklist as a consequence of the ABA's notifications and no reports have been issued indicating that either the ABA or IIA has ever checked whether blocklists are updated accordingly. Moreover a report on the effectiveness of approved filtering products commissioned by the ABA in 2001 states: "The list of sites [used in tests] does not include any that are included in the list of notified sites sent out to registered filter vendors by the ABA".

Most of the approved vendors do not sell filtering products that are reliant on vendor provided blocklists. Only 8 of the 17 do so (as at October 2002), while the remainder leave it to the purchaser to compile a blocklist, or use solely other means of blocking content such as keywords or skin colours.

Referral of prohibited items to filtering product vendors represents a government subsidy to a largely US-based industry that is most likely well ahead of the government anyway. For example, by March 1999 one censorware vendor that had been in business for three or four years had reportedly identified 8.5 million offensive sites (according to Senator Alston during a radio interview) without Australian tax-payer funded assistance. In comparison, during the two years ended December 2001, the ABA notified merely 529 items to censorware vendors. It seems very likely the vendors would already have known, or would have found out, about those 529 items without the ABA's notification.

In all probability, the ABA is entirely wasting its time referring URLs to approved censorware vendors, especially in the absence of any verification by the ABA as to whether censorware vendors update blocklists in accord with ABA notifications.

Complaints System

During the year ended November 2000, 2.7 million Australian households had home Internet access and 6.9 million Australian adults accessed the Internet (ABS 8147.0).

The ABA reported receiving 937 complaints during the two years ended December 2001 and approximately half (487) of these resulted in the ABA finding prohibited content.

While the number of 'complaints' reported is already extremely low in comparison with the number of Australian Internet users, the government reports over-state the number of complaints received.

Under the legislation, only Australian residents are entitled to make complaints. Clause 25 states:

"25 Residency etc. of complainant
A person is not entitled to make a complaint under this Division unless the person is:
(a) an individual who resides in Australia; or
(b) a body corporate that carries on activities in Australia; or
(c) the Commonwealth, a State or a Territory."

It could therefore be assumed that all 'complaints' were made by Australian residents. However, this is not so.

According to the Minister, when the ABA receives information from persons located overseas, or the ABA decides to initiate an investigation of its own accord, these are counted as 'complaints' and included in "Table 1: Outcome of investigations...(number of complaints)" in the six-month reports, although they are not valid complaints under the legislation.

For example, in response to questions in the Senate, the Minister said:

"The ABA initiated 33 investigations during the period 1 January - 31 December 2000. All 33 investigations were initiated following receipt of information which could not be formally considered to be a complaint (because the person was not entitled to make a complaint or the complaint did not contain all the required information), but which the ABA considered warranted investigation due to the apparent serious nature of the content concerned.

Thirteen investigations were initiated as a result of information received from persons who were not Australian residents, mainly from overseas complaint hotlines. The remaining 20 investigations were initiated after receipt of anonymous complaints."

Given that most overseas hotlines deal only with content relating to child pornography and paedophilia, and would presumably only refer Australian-hosted content to the ABA, it seems likely that 13 of the reported 22 'complaints' (in the calendar year 2000) that resulted in finding Australian-hosted prohibited content were not valid complaints, but referrals from overseas hotlines.

It would not be surprising if most, possibly all, of the content relating to child pornography and paedophilia had been referred by overseas hotlines rather than being an outcome of the complaints system. By encouraging Australian citizens to report child sexual abuse material to the ABA, the system places complainants at risk of committing a criminal offence, since mere awareness of the existence of such material implies that it has been downloaded into the user's computer. There is nothing in the Broadcasting Services Act which grants immunity from liability for users who report illegal content. In fact, there is at least one known Australian case where an Internet user has been prosecuted in exactly these circumstances.

EFA does not suggest that the ABA should not investigate referrals from overseas hotlines. However, it appears that the actual effectiveness or otherwise (including cost effectiveness) of the legislatively established complaints system would be clearer if the outcome of investigations resulting from valid complaints, and from other information, was reported on separately.

Prosecution Rates

On 8 August 2001, the following question and answer concerning prosecution rates was tabled in the Senate:

Q. "(7) In relation to the 89 items referred to state or territory police, and the 156 items referred to the Australian Federal Police, in the [calendar year 2000], has the Minister inquired about the prosecution rate of offenders and received information from police that would support claims that the Internet has been made safer; if so, how many prosecutions have commenced in Australia."
A. "(7) Law enforcement investigations relating to Internet content that the ABA has referred to a State or Territory police service are matters to be determined by the relevant agency in question. The issue of feedback on content referred to the Australian Federal Police (AFP) will be addressed in a new service level agreement to be negotiated between the ABA and the AFP, replacing the current Memorandum of Understanding between the two organisations. As part of the new agreement, the ABA will receive regular reports on action taken in relation to matters it has referred to the AFP. Where appropriate, the ABA will report on the feedback it receives."

Although fourteen months has since passed, neither the Minister or ABA have reported receiving any information from police that would support claims that the Internet has been made safer. Furthermore, it seems likely that if any Australians had been caught and prosecuted as a result of the legislation and ABA's activities, this would have made headlines, at least in a Ministerial media release.

Up ArrowGo to Contents List


Cost of the Scheme

For the financial year 1999/2000, the ABA spent $915,000 on its Internet regulatory activities, which included only the first 6 months of the scheme's operation. This included $100,000 spent on classification fees charged by the OFLC.

For the calendar year 2000 [1], the ABA spent over $618,319 and NetAlert spent $786,984. (The ABA amount probably does not include the $85,040 spent by the ABA on OFLC classification fees.)

For the first half of 2001, the ABA spent $531,478 and NetAlert spent $813,955. (The ABA amount probably does not include the $27,030 spent by the ABA on OFLC classification fees.)

This suggests a total cost in the region of $2.7 million per year, approximately $1 million p.a. by the ABA and $1.6 million p.a. by NetAlert.

Notes:

  1. The cost for the calendar year 2000 is under-stated in comparison with costs for the other two periods. It does not include corporate overheads nor a number of other costs included by the government when reporting costs for the other two periods.
  2. The government has not made figures for the financial year 2001/2002 publicly available.

Up ArrowGo to Contents List


Conclusion

There is no evidence or indication in government reports to support the Minister's claim that the Internet has been made safer as a result of the Federal Government's Internet censorship law.

Although government reports claim prohibited Australian content has been taken-down from the Internet, the government's desire to amend Freedom of Information law, to place a thicker veil of secrecy over operation of the regime than already exists, demonstrates the government is well aware that "taken-down" content remains publicly accessible on the Internet. The emergence of Internet archival services, which document web pages as they existed at various points in time back to the very beginnings of the Web, makes a mockery of any attempt by governments to achieve the modern-day equivalent of burning books.

The vast majority of content found and prohibited by the ABA is on overseas websites and continues to be made available by persons beyond the reach of Australian censorship legislation, which prohibits online publication of information that is not prohibited online in any country broadly comparable to Australia in terms of democratic political systems and cultures, and also prohibits information online that is legally available to adults offline in Australia.

A legislatively required Code of Practice for ISPs was developed by the Internet Industry Association (IIA) and approved by the ABA, which has the power at whim to make it compulsory on any ISP. Breach of the IIA Code, to which few ISPs have actually subscribed, is punishable by draconian fines and threats of taking away the ISP's business. The outcome is that ISPs must link to an IIA guide for parents, and offer one of a number of software filters, that may or may not block content prohibited by the ABA, given they are not tested for effectiveness in blocking same.

Filtering software has been available to Internet users who wish to use same since several years before the government introduced Internet censorship legislation. By March 1999 (according to Senator Alston during a radio interview) one censorware vendor had reportedly already identified 8.5 million offensive sites without Australian tax-payer funded assistance. The 529 items of overseas-hosted content that the ABA notified to filter vendors over two years is a drop in the ocean of the millions of web pages unsuitable for children.

After several years, censorware has not improved and cannot be relied on as a technological nanny. As reported in a study commissioned by NetAlert and ABA (issued in March 2002) "all products will pass through some content they should have blocked and block some content that should have passed through".

Internet-specific censorship law was not necessary to deal with globally agreed illegal material such as child sexual abuse images. Such material was already illegal to possess and distribute, both offline and online, in all Australian States and Territories. As noted by the NSW Standing Committee on Social Issues in June 2002, banning child pornography twice will not make it go away any quicker. It is doubtful that any Australian website would host child pornography material, since website ownership is too easily determined. Hence, it is unsurprising that most of the "Australian" material of that nature is found in Usenet newsgroups, most likely originating from outside Australia.

The few, if any, prosecutions of Australians initiated as a result of ABA investigations have been handled by State/Territory police under pre-existing laws; the ridiculously few web sites actually shut down by the ABA have made no difference to the millions of controversial web pages online.

The legislation has failed dismally insofar as its alleged objective of making the Internet safer for children is concerned.

It has, however, resulted in Australian adults self-censoring their speech more than required in offline publications in Australia or paying overseas, instead of Australian, Internet Service Providers/Content Hosts to host their web pages. The restrictions on Australian sites send Australian money offshore because Australian adults visit overseas sites to access content equivalent to material that adults can legally purchase offline in Australia, including material that is legally sold in Unrestricted magazines.

While the Australian government will no doubt continue attempting to hoodwink the public, EFA hopes that the increasing number of Australian adults who use the Internet is resulting in less parents being lulled into a false sense of security by the government's claims. The government cannot make the global Internet safe, or even 'safer', for children.

Protection of children on the Internet is a serious community problem, requiring adequate police resources and a commitment to community education, including practical Internet lessons for children and for adults unfamiliar with computer and Internet technology. The Government's Internet censorship scheme has been all fanfare and no substance.

EFA considers Schedule 5 of the Broadcasting Services Act should be repealed and the costly and failed Internet regulatory apparatus should be dismantled.

Up ArrowGo to Contents List


Related Resource

EFA Submission to DCITA review of the operation of the Internet censorship regime, 8 Nov 2002
(Contains EFA's comments on a number of other matters, additional to those covered in the above review of the scheme, in the second part of the submission).


References

Internet Content (Question No. 3590) re 'Six-Month Report on Co-Regulatory Scheme for Internet Content Regulation July to December 2000', Senate Hansard, 8 August 2001.

Information Technology: Internet Content (Question No. 223) re 'Six-Month Report on Co-Regulatory Scheme for Internet Content Regulation January to June 2001', Senate Hansard, 17 June 2002.

Six Month Report on Co-Regulatory Scheme for Internet Content Regulation January to June 2000, Department of Communications, Information Technology and the Arts, released 5 Sep 2000.

Six-Month Report on Co-Regulatory Scheme for Internet Content Regulation July To December 2000, Department of Communications, Information Technology and the Arts, released 19 April 2001.

Six-Month Report on Co-Regulatory Scheme for Internet Content Regulation January to June 2001, Department of Communications, Information Technology and the Arts, released 13 February 2002.

Six-Month Report on Co-Regulatory Scheme for Internet Content Regulation July to December 2001, Department of Communications, Information Technology and the Arts, released 21 August 2002.

Issues Paper, A review of the operation of Schedule 5 to the Broadcasting Services Act 1992, Department of Communications, Information Technology and the Arts, released 27 September 2002.

Internet content regulation, Lolita, interview with Senator R Alston, Transcript of John Laws program, 19 March 1999.

Answers to Questions on Notice, Australian Broadcasting Authority Supplementary Budget Estimates 2000-2001, Senate Environment, Communications, Information Technology & the Arts Legislation Committee, 23 and 30 November 2000.

Australian Broadcasting Authority Annual Report 2000-2001
  "From 1 July to 30 June 2001, the ABA referred 176 items of Internet content hosted outside Australia to the Australian Federal Police, and 59 items to the relevant State or Territory police service."

Effectiveness of Internet Filtering Software Products (PDF 1591 Kb), prepared by the Commonwealth Scientific and Industrial Research Organisation (CSIRO) for NetAlert and the Australian Broadcasting Authority, September 2001 (issued in March 2002).
   "The list of sites [used in tests] does not include any that are included in the list of notified sites sent out to registered filter vendors by the ABA."

Safety Net? Inquiry into the Classification (Publications, Films and Computer Games) Enforcement Amendment Bill 2001 (PDF 183 Kb), NSW Standing Committee on Social Issues, 6 June 2002.

Regulatory Failure: Australia's Internet Censorship Regime, Electronic Frontiers Australia, 5 May 2001.

8147.0 Use of the Internet by Householders, Australia, Australian Bureau of Statistics, 16 Feb 2001.

8153.0 Internet Activity Australia, Australian Bureau of Statistics, 18 Sep 2002.

Accessibility and Distribution of Information on the Web, Steve Lawrence and C. Lee Giles, Nature, Vol. 400, 1999.

Google Offers Immediate Access to 3 Billion Web Documents, Google Press Release, 11 December 2001.