All the twists and turns in the process so far have pointed in this direction.

The report that underpins the government's decision to legislate did not offer any new ammunition to the proponents of the scheme.

But, although not obvious at first sight, the report is actually a positive outcome for the schme's critics.

The report's findings reinforce that dynamic content filtering - or inspection of data as it passes through the system without any prior knowledge about the content - is not viable.

There is a clear trade-off between accuracy and performance penalty, and current technology does not perform at an acceptable level. The findings also confirm current wisdom that list-based filtering can be done at large scale, and without significant performance penalty.

This is great news since not only does it undermine the endeavour to enforce dynamic filtering, it also provides a political escape route to list-based filtering of refused classification (RC) content.

Critics will see good things in the government abandoning an attempted white-washing of the internet through complex and inaccurate filtering schemes - and instead presenting a package of solutions, including a more limited filtering scheme, as well as other measures of education and policing.

However, yesterday's move by the Government does not mean that all problems with mandatory filtering have disappeared -- far from it.

The decision to legislate has now made it urgent to raise public debate and seriously discuss the different positive and negative aspects of the legislation in an open and constructive manner.

Many difficult questions remain even after abandoning the initial notion of 'making the internet child friendly'.

There have been a plethora of issues presented so far in the public debate but perhaps the most pressing issue lies in the management of a black list and the definition of the term RC.

Using a classification does not protect against inclusion of any material, as long as the classification itself is not well defined and properly protected against changes.

It is not clear how the Government proposes to get buy-in from the Australian people regarding the definition of RC. Nor is it clear how the classification, once made, can be protected so that future Governments cannot start to misuse the powerful weapon blanket censorship constitutes.

Furthermore, the internet has some fundamental differences from other media channels that originate within Australia.

Some information that may be considered as RC material is actually material that should be made available to the general public. Go into any public library and you will find many examples of information that would fall under the current RC classification.

Even the late news often broadcasts imagery of severe violence, murder, abuse and atrocities straight into our living rooms without us having to search for it. Stopping such material, information about the holocaust, inquisition, torture and the Belanglo state forest has other far more dangerous consequences. It is far from obvious how to manage this difficult issue.

Children see violence, and they have a hard time understanding the context. On television, RC content is broadcast late at night when children are asleep, but the internet does not have morning, midday, afternoon and night; it is global and all phases of the day occur at once.

It is also very interesting to note that the trials in the report were carried out with a focus on web-based content on a mix of old and current technologies.

It has probably not escaped anyone that the government is currently rolling out a national broadband network to the tune of $43 billion. Most technologists agree that this will enable a shift away from traditional web-based content to a far richer mix of multimedia content, including video and audio.

Some more future-looking technologists see the emergence of immersive media and completely different ways of interacting with content and information.

The capacity of the NBN will allow much more information to be generated and be made accessible.

It is not necessarily so that a list-based scheme will negatively impact the performance of the NBN, but list-based filtering may very well be rendered useless. At least, I trust the NBN will bring an online world that has little in common with the current Web and modem based Internet, or I want a tax refund.

I hope the Government will invite the general public to a proper debate regarding all remaining issues. Judging by the strong reactions this debate has generated even before this legislation, the aftermath of yesterday's announcement will be interesting, to say the least.

Associate Professor Bjorn Landfeldt is with the School of Information Technologies at the University of Sydney