Archived News for April 2009

Brian Edwards on Quality

Brian Edwards is a New Zealand broadcasting legend. And he's gone and started blogging, over at Brian Edwards Media.

His most recent rant, Killing Boston Legal, begins like this:

Many years ago I had a heated debate with Neil Roberts who was then Head of Television at TVNZ. The debate was about standards, about quality, about good and bad programmes. Neil’s view was that there was no such thing as a ‘good’ or ‘bad’ programme. The terms were essentially meaningless. There were programmes that people liked and programmes that people didn’t like. It was purely a matter of individual taste. ‘Standards’ were arbitrary and personal with no objective validity. ‘Quality’ existed solely in the eye of the beholder.

There is logic to this argument, just as there is logic to the argument that ‘good’ and ‘bad’ are themselves entirely subjective concepts, that nothing is inherently good or bad. But you can’t live like that. You end up with anarchy and social collapse.

Edwards' rant is mostly about his firm belief that we can say whether a program is good or bad. And I agree with him. But I don't think the particular argument he puts forward in this part of his rant is a good one.

Check out that second paragraph. Here's what I think Edwards is saying:

P1 If good and bad are subjective concepts, you end up with anarchy and social collapse.
P2 Anarchy and social collapse are Bad Things.
C Therefore, good and bad are not subjective concepts.

The problem with this argument is that it's not deductively valid. That is: even though social collapse is indeed very bad, and even if subjectivism about quality would lead to it, that would still have nothing to do with whether or not quality subjectivism is true.

What makes the argument particularly vexing is that it assumes a form of subjectivism itself. An objectivist holds that nothing we think about a proposition affects its truth or falsity; nothing we believe about the height of Mt Everest will move that mountain up or down. But Edwards, in this argument, seems to be saying that because (we think) quality subjectivism is a bad thing, that makes it false. There's a word for the idea that our perception of a proposition affects it's truth: that word is subjectivism!

So, Edwards attempts to show that quality subjectivism is false by assuming another form of subjectivism is true. To an objectivist like me (and him, I suspect) that's pretty tough to swallow.

After following through this line of reasoning, I thought I wasn't perhaps being as charitable as possible in my interpretation of Edwards' view. I think we can make a better argument using the same first premise. Here's my take on it:

P1 If good and bad are subjective concepts, you end up with anarchy and social collapse.
P2 We don't have anarchy or social collapse.
C Therefore, good and bad are not subjective concepts.

This is certainly deductively valid: it's modus tollens. But I think the problem with this argument lies in the first premise. I think it begs the question. The only way someone would accept that quality subjectivism leads to anarchy and social collapse is if they already believe that quality is objective.

To elaborate, imagine trying to convince a quality subjectivist of the first premise. "If good and bad were subjective," you say, "we would end up in chaos. Things would go crazy!" "But," replies the subjectivist, "they are subjective, and things seem perfectly fine." I don't think there's any sort of evidence you could give for the first premise that wouldn't already rely on the fact that quality subjectivism is false. Perhaps that's why Edwards doesn't give any evidence for his claim. If he did, the argument's circular nature would be more apparent. If he did, we'd see that's he's not actually providing any reason for us to accept quality objectivism beyond his say-so.

Why is this important? Well, I often see a similar argument applied to moral subjectivism and religion. In such an argument, religion is equated with moral objectivism, and then the above argument seeks to show that moral subjectivism is false because it would lead to chaos. While the first part of such an argument is usually easier to attack (you don't need to be religious to think there are objective moral principles), I think it's important to see that the second part of such an argument (subjectivism would be anarchy) is pretty weak too.

Glorious VPN using TAP instead of TUN

I have spent a bit of time recently working on creating a VPN for my home network. I now have a Cobalt RaQ, firewalling my network (instead of silly little D-Link inbuilt fire wall). On top of this i now have a VPN using a TAP instead of a TUN interface. I won't go into the details of why i am using TAP instead of TUN, except for the following point:

TUN does not support the ability to use the Broadcast IP This creates problems for creating LAN games and using windows file sharing etc on your VPN.

So i decided to go ahead and use the good ol OpenVPN client/server. I am a fan of open source and the fact that pretty much every website i went to recommended it. Also, due to its popularity, OpenVPN has a whole lot of guides and HOW-TO's laying about. The only problem i ran into is the fact that not many of the HOW-TO's were windows server based and/or were for an outdated version of OpenVPN.

The only thing you need is the OpenVPN 2.0.9 installer available on their website, I am using this stable version instead of the Beta as i really didn't want to sit there and try and figure out their new commands. The only other thing you may need to do is sit on their website and read the VPN presentation they have on their main page if you are curious. That and Pinholing your router.

The instructions below apply to a Windows Server:

Step 1: Install OpenVPN (default settings are absolutely fine)

Step 2: Bridge the new TAP interface in your control panel with your internet connection. NOTE: When you do this, it assigns a new MAC address to your new NIC (bridged connection) and thus any DHCP reservered settings will need to be changed to reflect this.

Step 3: Follow the instructions on the OpenVPN page on how to create client/server keys

Step 4: Create a new file called server.ovpn (name is unimportant) and put the following text in it:

mode server
port 1194 #or your desired port
proto udp #better to use UDP as using TCP is double redundancy
dev tap
ca ca.crt
cert server.crt
key server.key
dh dh1024.pem
ifconfig-pool-persist ipp.txt
server-bridge #<OpenVPN Server IP> <Net mask> <DHCP Assign From> <DHCP Assign To>
client-to-client #let clients access clients
keepalive 10 120
status openvpn-status.log #log in the log
verb 3 #provide feedback. 3 is good for debug, 1 is good for operation

The server-bridge statement is a pain. Took me a while to get around that. An example of this is:


This says the OpenVPN server has an IP of and it gives clients one of 10 IP's between the range of -> So change these settings to your network settigns and you are away laughing.

Step 5: Right click your file (server.ovpn) and click "Start with OpenVPN"

Step 6: Get clients to connect

For the Client side:

Step 1: Install OpenVPN (default settings work)

Step 2: Create a file called client1.ovpn (name unimportant but good to diferentiate from others) with the following text in it:

dev tap #must match server setting
proto udp #must match server setting
remote hostname port #for example, 1194
resolv-retry infinite
ca ca.crt
cert client1.crt
key client1.key
verb 3 #once again with the verbose above

Step 3: Right click your new file and click "Start with OpenVPN"


  • This will not work if you are on the same network, client must be on an external connection otherwise IP conflicts for some odd reason

Other than that, Dom sorry for the long post, Everyone else, deal with it. Any questions, let me know.

Newest Posts