Oppose Internet Censorship
In this page I speak strictly for myself, -- not for RMIT and not for any other organisation with which I am affiliated.Don GingrichLast update 18 November 2008
Lecturer - System Administration and Operating Systems
School of Computer Science and Information Technology
Some thoughts on the proposed Internet filtering
Most of this was posted in 2009 and in December 2008. Most of us opposing the filter had more or less hoped that it would be dead by now. But it seems imposible to kill. The test results are back from the trial, and what is more interesting is what the filter doesn't do as compared to what it actually accomplishes. In spite of the declared ease of by-passing the filter, the reality that it only blocks a small part of the material that it claims to block, and the fact that it simply won't even try to block material on high traffic sites such as You-Tube, Senator Conroy continues to claim that Australia wants and needs this filter.
SAGE-AU Discussion of content and comparisons
Censorship in Action?
How do we "protect the children"?
Common Carrier Status of ISPs and what it means
International Responses to the Policy
Australian Senate Question Time
A non-discussion with a staffer in Anna Burke's office
A note to an opposition MP(Peter Costello, as it happens-- his electorate is next-door)
Note to Senator Fielding, following a discussion with a staffer in his office
Note to my local MP, Mike Symon, about why the filter should go(added 05/2010)
Put Conroy Last -- an election strategy for Victoria(added 05/2010)
References and Links to possibly useful material
Sign the EFA Petition against the filter
Introduction and "sound bites"At present, the plans of the current Australian government are not completely clear with regard to filtering of Internet content. It could be more or less intrusive depending on a number of details that have not been completely defined by the government at this stage.
What is clear is that it will not actually achieve the stated aim of protecting children. Protecting children is a vexed question in any case. It seems that the continuing argument, "What about the children?" is simply a cheap rhetorical trick that tries to demonise anyone who argues against the government position.
Remember, it is possible to "care about the children" and expose the flaws in the government proposal.
A little bit censored is like a little bit pregnant
The real concern about the whole question of censorship is that the list of what is to be blocked is secret. There is no way to know if legitimate content is being blocked, or if the government of the day is blocking material that disagrees with its policies.
If depictions of nudity are evil, by extension,do discussions of legitimate health issues that may need to be illustrated with nude images also get defined as evil and get blocked.
If, as some have suggested, this censorship is being driven by what is sometimes referred to as the "religious right", will that mean that educational and legitimate discussion of human sexuality including birth control will be blocked? If so, who are we protecting really, and what are we protecting them from?
The remainder of the material on this page won't be as organised as I would like. Unfortunately, at the moment I am in the midst of marking final examinations so I don't have the time to really organise the material. Most of what follows is a series of e-mails that I have sent to either the SAGE-AU mailing lists or directly to various politicians.
Note: Any and all links to sites external to this one are links to material that, in my personal opinion, may be of interest to other persons who are interested in the Internet filtering debate. The opinions expressed on these external pages strictly represent the opinions of the authors of that material. A link from this page does not represent an explicit or implicit endorsement on my part. It does say that I found the comments and information personally interesting and that you may also find it interesting.
I hate having to insert disclaimers, but I suppose it's necessary
This link, to press releases on the SAGE-AU website is one that I do endorse, particularly since I have contributed to some extent to the writing of these press releases.
In a SAGE-AU discussion, that was primarily centered on whether or not the Internet and Internet content is equivalent to films and books, someone said:> There have been a bunch of vociferous answers I'd paraphrase as:
> "Internet: Live Free or Die"
Would you be arguing for the same level of censorship of telephone conversations? I don't think so.
The whole problem is the question of how one defines the Internet. If the 'Net is defined as some nebulous cloud of "content" "out there" then arguing for the application of censorship standards may be reasonable. If on the other hand, the content is correctly, IMHO, viewed as something that is transmitted over the system of connectivity provided by the Internet, then it becomes a completely different question. Requiring ISPs to filter would be requiring a completely different an onerous regime for them compared to all other "common carriers" Or, since pornography is transmitted from state to state via Australia Post, are you going to require that the postie verify that the recipient of a "plain brown parcel" is over 18? Are you going to require that every DVD that arrives in International mail be checked to verify that it conforms to the ACMA standards? What about parcels via FedEx or other common carriers? What about calls to International phone sex lines? It goes on and on -- If an ISP is a common carrier, and according to what I have read, they *are* officially defined as common carriers, then creating an exception for ISPs is ludicrous.
If this makes me "Internet: Live Free or Die" then so be it.
I rest my case.
 the problem with "content" on the Internet is that it is so fluid. This page has easily changed 10 times since it was first created. It will change again as soon as I have something else worthwhile(IMHO) to say. Much of the "content" on the Internet is not created by corporations, but by ordinary people. So, who pays to have the material rated? Ratings aren't free.
 With appologies to James Tiberius Kirk, Star Trek: the Movie :-)
Filtering out the fury: how government tried to gag web censor critics
I wrote a note to my local MP:
I call your attention to the following article from the Sydney Morning Herald:
As you may notice from the bottom of this note, I am both a lecturer at RMIT University (I lecture in System Administration) and a member of the System Administrators Guild of Australia(SAGE-AU). As the organisation that represents the people who would be called on to implement Senator Conroy's filtering plan, the Guild has had some in-depth discussions of the implications of the plan. Mark Newton, mentioned in the article, has been an active participant in these discussions. I have found his comments to be well reasoned and insightful.
The critical problem is that the study on which the Senator is basing his argument is deeply flawed in a number of ways. Put simply, there is no way that it would be accepted for publication in a peer-reviewed journal. Arguing that censorship on the scale that he is proposing is feasible based on this study is ignoring any number of unpleasant (to the Senator) realities.
To put pressure on the Internet Service Providers in an attempt to silence Mr Newton's criticism is simply unacceptable in a society that values free discussion of political issues. I sincerely hope that you will make an effort to rein in this abuse of government power.
I would be pleased to make myself available to discuss the whole filtering issue in greater detail. I will be unavailable on the morning of 30 October and the evening of 11 November since I have students sitting exams at those times, but would be happy to meet at another time of your convenience.
Well, I believe in the earlier discussions a few months ago someone pointed out that there were already ISPs that provide filtered Internet for families. Would it not be better for the government to do more to subsidise and highlight these ISPs. Push an official "Family Friendly" rating and publicise the government website that has information about the program and lists of ISPs with the rating.
I've been thinking along those lines myself, coupled with pointing out ways that the proposal won't really "protect the children" i.e.
- drive traffic in nasty stuff underground in a way that would make it more difficult for law enforcement to track. But then again, that might be OK from a government perspective - out of sight - out of mind
- I believe that the free speech issue, may be a viable argument within reason examples of lists in other countries that have blocked breast cancer awareness, testicular cancer awareness, birth control (If you don't want abortions - then IMHO birth control is the alternative.) The level of typical collateral damage with any blocking list is typically a lot larger than the study suggested.
- Added material here The government study of filtering software was published in June. Many of us believed that the negative impact of the tested filters meant that no further action was likely from the government. The report is Closed environment testing of ISP-level Internet content filtering The most polite comment possible is that it would be unlikely to be accepted for a peer-reviewed publication.
- And, as I read it, that was a major flaw in the study -- there was a severely limited set of sites in the first place and the sites to be blocked were a subset of that list. When we start adding sites daily to a list that is several times (my guess is thousands times bigger) the size of the test list, how is it going to scale? Hopefully they are using some sort of b-tree algorithm, but even that will start to strain a bit as it gets really big. Maybe I'm looking at this wrongly, and I'm sure someone will point out the error of my thinking, but I'm looking at this a problem in real-time processing. And, if the performance starts to degrade to the point where the filter cannot keep up with the traffic - really bad things happen as in packets get dropped and need to be resent, which in turn increases the load and sends the system into a downward spiral that reduces traffic to a trickle compared to optimum conditions.
Is it possible (net engineers out there) that this was a difference between the acceptable systems and the really poor ones? That the acceptable ones had simply not yet been loaded to the tipping point?
Now all that I need is to get to see my local MP to explain the whys and wherefores, including what I learn in the responses to this.
But there is a deeply cynical part of me that thinks there is a good chance that this is all about appeasing Senator Fielding to get stuff through the Senate. In which case a different sort of reason and logic prevail. And, rational arguments about filtering may be met with, "My mind is made up -- don't confuse me with the facts...!"
> The Australian telco marketplace is almost a hundred years old now,
> and throughout that time carriers have been seen as "Common Carriers,"
> who are not responsible for any material carried by their networks.
Wait a minute.... I cannot believe that I have been missing something so obvious for so long. Doh!
If we allow the filter to go forward in any form we are allowing government to change the common carrier status of ISPs. This may not seem like a big thing, but removing the common carrier status changes many very significant aspects of how ISPs do business.
There was a series of fairly nasty cases back in the early 90s that specifically revolved around this very question. A person in England would post "flamebait" in Usenet groups, then sue anyone who took the bait (and their ISP). The Melbourne PC User Group got done in this, partially because the common carrier status was not clearly defined at that stage, and partially because this individual set the damages demanded low enough that it was cheaper to pay than to fight in a foreign court.
So, make no mistake, the common carrier issue is a big issue. And, if ISPs lose the common carrier designation, the next step to enforced monitoring of e-mail traffic or other sorts of traffic is a small one.
But, to be honest, I'm not so sure that there has ever been any formal designation of ISPs as common carriers. (Though I am well aware that the British system of law is much more reliant on precedent than the US system that I grew up with.) Perhaps it would be better to pursue a formal declaration of common carrier status for the Internet.
Comic for 9 November 2008
It seems that the debate is starting to attract international attention.
One of the better comments, if a bit light for technical detail.
But it's not the first time that Australia and Internet filtering have copped a serve from User Friendly. This one is from 1999.
Comic for 6 June 1999
Subject: internet censorship: in the Senate todayOK, the link to the live feed would not work anymore, but you can find the comments in the Hansard for 11 November 2008
From: "Ludlam, Scott (Senator)" <Senator.Ludlam@aph.gov.au>
Date: Tue, 11 Nov 2008 13:06:24 +1100
Thanks for getting in touch with the Greens regarding the internet censorship issue - I'm just letting you know that I'm going to be raising the issue in the Senate today, and you can watch the broadcast live online here:
In the Senate question time at about 2:30PM eastern summer time today I plan on asking Communications Minister Stephen Conroy about his project for mandatory internet censorship, the so-called 'clean feed' which you've written to me about.
If you miss the broadcast I'll have the transcript and a video clip here as soon as possible: http://scott-ludlam.org.au/
Depending on the Minister's answer, you might like to give him a call on (02) 6277 7480 or email him on firstname.lastname@example.org
This proposal has raised major concerns in the online community - thanks for taking the time to add your voice to this debate.
Senator Scott Ludlam
Communications spokesperson for the Australian Greens
I suppose that since this debate was on Remembrance Day, we should take a moment to be thankful for those who served in several wars protecting our right to actually be having this debate.
I've just got off the phone from a 20 minute conversation with the local MP's version of version of Sir Humphrey.
It looks to me like the Labor party is really closing ranks on the filter thing. The last time I spoke to the electorate office they were ready to throw Conroy to the wolves. This time they were spouting the Conroy line, chapter and verse.
I got the following arguments:
- But what about the children (about 10 or 15 times)
- When I suggested that a lot of the traffic in unacceptable smut isn't via HTTP the reply was that I wasn't looking in the right places. When I said "show me the proof" the reply was, "If I did I would be sacked" And so it goes...(See Catch-22 by Joseph Heller)
- When I suggested giving more money to the AFP to stop it at the source -- I was told that they were ineffective in any case
- When I commented that the current approach was swatting flies with a a 40Kg sledge hammer -- he said, but if it makes things better for one kid <sigh !> Wasn't thinking fast enough, should have pointed out that the "kid" has already been abused by the time it gets to the filter. If we're going to help "the kids" it has to be before it ever gets to the net.
- When I tried to explain the problem of dropped packets and the tipping point in filtering, the response was "<pat> <pat> <pat> on the head, trust me, the government has smart people who will make this work." But don't confuse me with the technical details of why it won't work.
- The comments suggest to me that the government has definitely determined to go full speed ahead with a test on a real network, wherever that will be.
I'm so frustrated I could spew at the moment.
I do feel a bit better for having had my whinge, though.
Thanks for listening.
Some problems with Internet filtering
From: Don Gingrich
To: "Peter Costello" <email@example.com>
Date: 12/11/08 11:58
Dear Mr Costello,
I recently tried to discuss this issue with Anna Burke, my local MP. I wound up talking to one of her staff, who kept repeating, "But what about the children?" as it it was some sort of mantra that made everything OK.
Yes, I am concerned about children, but by the time material is being accessed on the Internet, the child in question has already been abused. What is really needed is police responsiveness to rescue children before they are harmed, or further harmed. Filtering is not going to achieve this.
The other side of this coin, and the one that is seldom mentioned, is that there is a group of people who seem to be deathly afraid of sexuality in any form. In some ways it resembles the schizophrenic attitude that is common in the USA, where I spent the first 31 years of my life. There it's considered OK for television to broadcast thousands of murders a year, some fairly graphically depicted, but a national catastrophe that Janet Jackson's breast appeared on television. I don't know about you, but I've got a fairly strong idea about which viewing is more likely to be psychologically damaging.
Another aspect that has not been discussed with respect to the proposed blocking list is the question of legitimate sexual education sites. While, as a voter in Victoria, I am pleased that abortion has been decriminalised, I would be sincerely unhappy to see abortion become the primary means of birth control for young women. And it seems to me (And I certainly hope that I am wrong) that the people pushing this filtering issue seem to have an agenda similar to that of the so called "religious right" in the USA. In which case, are we likely to see web sites preaching abstinence as the only available web sites discussing sexuality for teenagers? It's been longer than I'd like to admit since I was a teen, but I can remember being a "walking hormone" and there is no doubt in mind that abstinence is another way of saying teen pregnancy and wrecked lives. "Think of the children!"
Finally, I'd like to take a moment to mention the technical side of the issue. Please accept my sincere apologies if you have a computer science and networking background. I'm going to try to keep this as simple as possible, close to the way I would describe it to first year students.
It is my understanding that the "block list in the trial consisted of about 1900 sites (described on pages 21 and 22 of the report(Closed environment testing of ISP-level Internet content filtering) as URLs (i.e. http://www.somesite.com). And there were about the same number of OK sites. That's all well and good for testing, but there are a several problems when we try to apply this to the real world. The first problem is that there are a lot more than 2000 web sites with acceptable content, and the second is that, depending on criteria, there will be more than 1000 unacceptable sites.
The increase in blocking complexity with increasing numbers of sites on the unacceptable list is not linear. That's actually partially good news for those who are proposing the filtering. If I were designing something like this, I would use an algorithm that converted the URL into a value that could be used as an index into a binary tree. This would give a performance degradation based on the base 2 logarithm -- i.e. 2 sites would have a factor of 1, 16 sites would have a factor of 4, but (when the bigger numbers kick in) 1024 sites would have a factor of 10 and 16384 sites would have a factor of 16. (Note that this makes no consideration of the conversion from URL to index, which would be a linear factor. i.e. 10,000 packets to test would be 100 times more difficult than 100 packets.)
There are two non-obvious points about how the Internet and the Internet Protocol works that make this much more difficult. The first is that, when a user asks for a document from a web site, the document is not transmitted as a single large block of data. There is a fixed maximum size of a block of transmitted data. This is dependent on a number of factors, including the particular hardware that is installed between the source and destination machine. For the purposes of discussion, most systems use a maximum packet size around 1500 bytes. So, a picture that is 150,000 (150 Kbytes) would need to be split into 100 packets, each of which would need to be blocked by the filter. And the filtering report, at 2.8 mega-bytes would need nearly 2000 packets, each of which would need to be examined and "approved" as I downloaded it from the ACMA site.
The second is that the URLs mentioned in the study are not actually anywhere to be found in the packets that are sent across the net. What is in the packets is a numeric IP address. As I mentioned earlier, this would be a good value to use in searching the list of blocked content, but there are several problems, one of which is that blocking entire sites for any bit of banned content is not likely to be popular. A further problem is that the "mapping" of URLs to IP addresses is not necessarily static. The mapping is managed by the Domain Name Service which runs on a large number of machines throughout the Internet. In the worst case, looking up an IP address in the DNS can require communication with 2-4 outside machines, with the attendant delays. And, if the filter is going to avoid blocking legitimate content, then it will need to re-map the URLs on a regular basis.
Another technical problem that I tried to raise with Anna Burke's staffer was that, according to a lot of material that I have read, the sites with "really nasty" material often do not have proper URLs. They are run on sites that do not actually appear in the DNS. In fact the information that I have is that they only appear for short periods of time to make it harder for law enforcement to trace them. Actually getting these mapped into the block list is problematic, at best. Yet these are the sites that are most likely to have the content that everyone agrees that we would most like to see blocked. His response, when I commented that, in nearly 20 years of Internet use, I had never seen any really objectionable material was to say that he had lists of URLs for really nasty stuff that's "out there" and that he couldn't demonstrate that it really does exist because he'd get sacked if he did. So it's strictly a "trust us --we're the experts" sort of response.
So where do these problems leave us as we move to a larger block list? There's a further non-linear problem with TCP/IP traffic generally. (And most Internet traffic uses TCP/IP -- the web's HTTP is a sub-set of TCP/IP) The problem is that TCP/IP was designed to automatically deal with network problems such as loss of data packets. If the source sends a bunch of packets and does not receive an acknowledgment from the destination, the source will re-send the unacknowledged packets. So, if a filter begins to be overloaded and starts to drop packets that it cannot process, the unprocessed packets will be resent from the source, increasing the load and increasing the number of dropped packets. This cascading effect can effectively reduce the actual effective throughput of a network to some fraction of its design capacity. This is the biggest potential problem with the proposed filtering. When it fails, it is unlikely to fail gracefully, but rather catastrophically.
I'd much rather see the government subsidise "kid friendly" optional Internet feeds from ISPs so that parents could use their parental discretion and use these services at equal or lower cost than the existing offerings.
I'd also like to see more resources given to the AFP and state police forces to enable them to deal with "cyber-crime" better so that hopefully, instead of blocking unacceptable content, we can stop it from being created in the first place and thus really, "protect the children"
If you've got this far, thanks for listening.
There is a perception, that may be incorrect, that various religious groups are strongly in favour of the proposed Internet filtering. I would like to direct your attention to the following URLs:
A Sydney Anglican Blog
and this post in the Whirlpool discussion forum:
Whirlpool discussion post
I'd like to introduce myself, I lecture in Computer Science at RMIT. My specialties include System Administration, Networking, Operating Systems and Security. As such, I would like to think that I am more qualified than most to discuss the technical implications of the proposed filtering.
I stopped by your office in Syndal today and had a productive discussion with the staffer who came to the front counter. (I'll be honest here -- I was pleasantly surprised to feel that my views received a fair hearing. Certainly it was a more positive hearing than I received from the local ALP MHR.) In discussing the proposals for filtering the Internet there are several important points that I believe should be taken into account:
- The Web (World Wide Web) is only a small part of the traffic on the Internet. (less than 1/2)
- The "really nasty material" for which there is consensus agreement about blocking generally does not travel via the Web's HTTP (protocol)
- Proposed filtering solutions mostly concentrate on the HTTP and thus won't stop the nasty stuff
- There is also an encrypted version of the HTTP which would be invisible to the basic filters. (Note that it is theoretically possible to institute what is described in cryptography as a "man in the middle" attack and thus see the contents of these packets -- we really do not want to go there.) If the filter included HTTPS, it would mean that communications with banks would no longer be secure. This cure is worse than the disease.
- The government is proposing to spend large sums of money to improve the speed of Internet access for Australians.
- The testing from last year demonstrated that the accuracy of the filter (blocking the undesired material, and only the undesired material) is inversely proportional to the reduction in data transfer. We can have an accurate filter, but at the cost of a significant degradation in performance.
- Combining the previous two points amounts to shooting ourselves in the foot as a nation.
- By the time that "really nasty material involving children" is transmitted across the Internet, the child in question has already been harmed. If this happened overseas, then, other than assisting the foreign country in prosecuting, there is little that Australia can do. It would be naive to assume that blocking this material from Australia would reduce demand. (In any case, not forcing the traffic further underground would give the AFP a better chance of tracing the Australian recipient.) The alternative, that an Australian child was abused in the production, is, in my opinion, an argument for better resources for the AFP to allow them to stop the abuse at the source.
Thank you for listening,
A note to my local MP about both filtering and the National Broadband Network
One of the things that education people tell teachers is that it is a good idea to say something positive to a student, before you tell them something negative. As it happens, I believe rather strongly that the proposed "Fibre to the Node" broadband network is a really good idea. So, in this letter to my local MP, I was able to use the classic "carrot and stick" in explaining what I think about Senator Conroy. I actually sent this as snail-mail since individually written snail-mail scores the highest level of response from politicians.
If you agree with what I said, feel free to use the ideas, but try to put it in your own words. Hundreds (well since it's my site, tens) of identical letters don't have the same impact as individually written letters that share common ideas. In any case, my letter is probably too long -- being a lecturer, I find it hard to resist pointing out all of the reasons why I think the filter is a bad idea. (And I still forgot to mention how easy it is to bypass the blinkin' thing. :-( )
Anyway, here it is. Hope that it contributes to the campaign.
A letter to Mike Symon, MP
A thought about the upcomming Federal election
I've been thinking about this for a while. The Internet filter is being driven primarily by Stephen Conroy, the Senator from Victoria, who is also the Minister for Broadband, Communications and the Digital Economy. If the Senator were to fail to be re-elected in the comming election, it would be a clear signal that the filter has strong opposition. He will be up for re-election this year, so this can work.
What is needed is a coalition of anti-filter persons who are prepared to produce "how to vote" cards explaining how to modify the major party's standard ticket to Put Senator Conroy Last. And then volunteers to distribute the sheets on election day. As far as I can tell this would be a legal thing to do. Please contact me at the e-mail address at the top of this page if you are interested in this effort.
To gather ammunition for the discussion of Internet filtering, I had a quick look for a number of relevant papers -- one that stands out considering that the Minister is quoting BT's CleanFeed as an example, is this PhD thesis from Richard Clayton of Cambridge University.
Cambridge University Technical Report 653
This is free and includes more than the first document that I found, which was in a Springer-Verlag collection at a cost of $25.
If you haven't looked at it, you might also like to take a look at:
Developments in Internet Filtering Technologies and Other Measures for Promoting Online Safety.
And of course, the one that kicked off all of this kerfuffal back in July:
Closed environment testing of ISP-level Internet content filtering
And the public submissions to The Review of Schedule 5 to the Broadcasting Services Act 1992 is not not particularly exciting reading, but since it is from 2004, it does show how long this issue has been "simmering".
Richard Clayton's home page has some other potentially interesting stuff:
And, if you have access (And RMIT students do through the electronic section of the library) to the Springer-Verlag online e-Books, there were a couple of further books and chapters that are relevant.
Finally, the original 1999 CSIRO report is at: Access Prevention Techniques for Internet Content Filtering
A paper by Michael Flood and Clive Hamilton of the Australia Institute. This one is central to the debate since it is cited as the rationale for the filtering. Regulating Youth Access to Pornography
You may note that there is little reference to child pornography, the primary stick that the government is using to beat its opponents.
This paper, from the US University of New Hampshire, paints a very different picture of the level of access to pornography by teens. Either it is a different Internet over there, or one or the other groups of researchers are fiddling the books. Given that this study was published in a peer reviewed journal, I know which way that I'm betting.
The Exposure of Youth to Unwanted Sexual Material on the Internet
And this is a discussion, by what purports to be a "family friendly" group in the US, of the discrepancies between the above two studies.
Online kids' exposure to porn: 2 studies in 2 countries
And this URL has links to a lot of useful and relevant material: libertus.net: about censorship and free speech
A Wiki on the overclockers web page:
Australian Internet Filtering
If you want to contact your local MP, you can find their e-mail address from this page: Parliament of Australia: Who's Who
The campaign by EFA (Electronic Frontiers Australia) against filtering is called No CleanFeed.
And EFA (with even more background information and links than I have) is at:
Feel free to use the ideas in this page in writing to your local member. But, for maximum impact, try to write your own note -- dozens of unique notes have more impact than hundreds of identical notes.
And just to point out that I am not necessarily categorically opposed to Internet filtering, just filtering that is imposed from "on high", I'll include a link that describes setting up a filtering system for a home connection. It is based on the use of a Linux firewall/router managing the connection and uses the Squid web proxy to do the filtering. I'm sure that there would be ways to evade this filter -- I'll need to do some experimenting. But -- it's better than a filter on the local machine and hopefully will require more effort to evade. It also has the advantage of allowing blocking of adware with the use of an additional ruleset. Which brings me back to the primary reason that I like this system. With this system I will be able to look at the rules and adapt them as required. It is a near certainty that I will find rules that I drop and possibly rules that I add after I have a good look at the list.
I'm not talking about filtering for University age people, but for teenagers. And, even for teens, there will certainly be links that I will think about removing from the block list.
It's at Linux.com -- A parent's guide to Linux Web filtering
Hope this is of use to others.