I forwarded my last post to the six trial participant ISPs and promptly received the first comprehensive response from one Andrew Robson, Managing Director of TECH 2U, which is no surprise given the terse explanation I included:
I write to register my disapproval of your participation in the Internet Service Provider (ISP) Filtering Live Pilot which I consider to be a full frontal assault on the rights of all Australians as well as a potentially serious inhibitor for the next generation of cloud computing solutions. I have written about this important issue on my popular blog and called for a boycott of your services due to your participation in this pilot.
Your support of this ill-conceived program is inexcusable and I hope that the Australian public will exact a heavy toll on your business as a result.
I figured it only fair that he be given the right of reply here, though not without a response from me. I did consider the possibility that (like larger ISPs iiNet, Internode and Telstra) they were not ‘supporting’ the filtering as such, rather gathering data and possibly even preparing a case against it, but at least one of them prides themselves as being “Australia’s first content filtered ISP”. It’s also not at all clear what financial support/incentives were provided (it doesn’t help that they “are not contractually able to discuss” the details either) and times are tough in this cutthroat industry so it wouldn’t take much to convince a smaller ISP to participate. So without further ado:
Thank you for your email outlining your concerns regarding the particpation of TECH 2U Pty Limited in the upcoming government internet filtering trial.
We note from our files that before writing to us in such a tone you have not actually asked if we are in favour or against internet filtering. Nor have you asked why we are particpating in the trial in the first place.
Over the last six months we have read with real interest the debate over internet filtering and have noted the lack of any hard evidence either way of its operation in a real world enviroment. We like everyone else have firm views on what is possible and what is not.
This of course puts aside the very real question of whether or not the internet should be filtered at all. Nor whether parents should take responsibility for their own children.
In this case we see this trial as the first real test of a real world scenario and the first real chance for people to either prove or disprove the various theories. Provided with a golden once in a life time opportunity to be part of this process we found it hard to decline. It was better to be part of the test team than to be on the sidelines yelling (or emailing) while someone else determined our destiny.
We spoke to our customers who agreed. Better to be part of it than not.
We will of course conceed that not many of our customers are “cloud computing” application users and supporters such as yourself. (In fact many would never have heard of it). It therefore follows that in our test base no impact will be seen on “cloud computing” at all. Mathematically and technically correct but in fact perhaps a skewed test result. As you would agree. When the test results are colated they will show no impact on “cloud computing”. More likely they will not even mention it.
My suggestion is that you would be better to actually join one of the ISPs particpating in the trial than not. Once you join ensure you are part of the test trial and then be part of the proof that it does not work. Have your say and make it count. Fill out the customer surveys, answer the questions. Demonstrate why it caused a problem and not just state an opinion. Governments want hard facts to back their case either way (read here cover arse) and sometimes emotion alone while admired does not do it.
We joined this trial to obtain the facts.
The real problem we now face is that the very people who should be in the trial with their expertise have turned their back on it. Worse still they are now “calling for a boycott of your services”. If this really does occur it may distort the test results in a way that we will all regret later on. Watch this space as they say.
For your information we are not contractually able to discuss the specific tests involved but it would take about 500 -1000 users to have any impact on the test results for any individual ISP. The tests will actually vary from ISP to ISP but most will occur in the March to May 09 time period.
TECH 2U Pty Limited
What I took from this was that they don’t take a firm stance either way on Internet censorship, but they are aware of its limitations and they preferred to be involved than a spectator. I’ve said before that I don’t think the trials will be representative as without significant load performance may well be adequate, without extended duration outages may be avoided and without those of us having the requisite expertise (like myself and many of you) the feedback will be largely useless.
Andrew claims that “not many of [their] customers are ‘cloud computing’ application users” but he probably didn’t realise that cloud computing includes Facebook, Google Apps (including Gmail), Hotmail and many of the web applications that your average Internet user uses every day. These applications will all suffer to varying degrees with Internet filtering – it is simply not possible that their performance could improve (as may have previously been the case with caching proxy servers) as each request is dynamic and needs to be processed by the servers as quickly as possible. With more ‘moving parts’ reliability will also suffer – again it is not possible that inserting complex (relatively) unreliable systems into the data flow will help.
Furthermore it should not be too difficult to mount a (deliberate or inadvertent) denial of service attack against these devices. It is well within the realm of possibility that one of the many viruses in the wild today could generate enough requests to take down even the largest filtering system, and that’s just in the course of their ‘work’ spamming blogs, wikis and websites – let alone a malicious attack consisting of many small, random requests. Indeed it is likely that someone able to control even a relatively minor botnet could ‘take down the Internet’ for a large portion of the Australian population, leaving the ISPs essentially powerless to stop it without running afoul of the law.
Anyway I appreciate the time that Andrew took to respond and encourage people to take up the matter with their ISPs even if only so as to give them the right of response.