"So he's saying that the only way to stop the 'net from being placed under centralised control would be to place the 'net under central control?"
No, he saying where there needs to be some sort of aurhority, it should be a community of experts not some random government wonks.
The internet has no central control, it's edge controlled. There are a couple of single points of failure choke points like the root servers at the physical level, and ICANN at the political layer, but these can both be routed around and don't have any actual authority.
"John Nobody blathers on about something he knows very little from his summer place while sipping tea.
Why do Slashdotters buy the banal esoteric blather that comes from guys like these who have no real connection to reality?"
Huh?
Maybe you don't know anything about Zittrain but I do. Here's what I've seen.
In 1996 or so the Internic began charging for domain names. Immediatley there was consensus for new top level domains, or "life after.com".
The community fractured into two camps: one that wanted a central authority, which hapoened to be them, and the other half that believed the insisible hand of Adam Smith would permit a self regulating industry to flourish.
Thems what was in charge shut out the rest which caught the governments attention. They stepped in and announced it was their baby and would settle things, and insitagted a process by which the replacement for Jon Postel and IANA would be formed.
The Berkman Center for Law and Technology at Harvard, under Charles Neeson was interested in this at it appeared to be a form of internet governance. Larry Lessig was there at the time and was slightly involved althouh he left shortly after. Zittrain was there are a while and was the real time scribe at nearly all te ICANN meetings.
He's seen first had what happens when expert self regulation is displaced by the collusion of corporate and government interests.
At least that's what I've seen that gives me reason to feel he qualified to comment on this.
"That article sure uses a a lot of words to say 'the web should be communist'. "
Rubbish.
The point is internet technology is so complex very few people understand how all of it works, and how it works all together. The further away you go from technical to admisistrative skillsets the less likely are people to understand what's going on. That's the difference bewteen SMTP actually working and a sock puppet raising venture capital.
This has nothing to do with capitalism or communism and is inappropriate for a framework of discussion about technology and what kind of environment open standards and processes need to flourish.
"Why on earth should he think that "experts" are any better at self regulation than any other random group of people?"
Because they're "experts" and not a random group of people.
Jono's quite right: frame it in this context - who would you put in charge of managaing, say, the Linux kernel? A bunch of guys that knew it best or a governmnet committee of people qualified to do something else?
TFA is wrong though when it says "this almost happened with domain names". Substitute "DNS" for "Linux" in the above and you have ICANN.
Jono was there as well, to watch this all come up. In fact Lessig and Zittrain were involved in the process that led up to ICANN and were as surprised as anybody else when the government stepped in and said "We know you've been working on this for a year all over the world, but here's the baord and here's the organization. Thanks, but you can go home now".
That phones may be able to ray trace is news? Sounds more to me like intel was of reading in the news all week how inferior their graphics stuff was because of the Microsoft Vista debacle part eight - and suddenly we have an anonymous tip to a blog at intel saying ray tracing on phones there is "an opportunity to deliver more content in less time" and "Ray-Tracing may become a very popular technology in the upcoming years".
A popular technology? Like a working filesystem? They're real popular I hear. Or an on off button that actually works.
Slow news day + intel graphics dept astroturfing = ray tracing on phones is news.
"Although I do love watching them flip out when I show them how my 24 year old, 5000 lb, loaded with bells and whistles, diesel Mercedes sedan gets 38mpg on the highway. "
38? Which one is that? a 190D? My 300SD gets 33 mpg at best. Booth's US and Euro 190D 2.5's got 45 and 48 mpg respecitvely.
"Also, if you have an old engine you will probably need to clean the fuel filter after several thousand miles as the biodiesel will clean out the accumulated junk from regular diesel"
As if. A buddy of mine swiched in his 25 year old Mercedes diesel. He went through a case of fuel filters in 200 miles.
"Non-injected diesel engines don't have these issues"
You would have really hated 4Mhz Z-80's then. Or the PDP-11's that wern't a whole lot faster. They did get the job done letting C and Unix get written on them though.
Go rent a copy of "Who killed the electric car" then figure out what role the oil companies have in a hydrogen based economy.
Then understand why the Bush administration dumped millions into hydrogren resarch and never mind any running car is ten years off from whenever you ask.
We *might* be able to make hydrogen at home? Great. I *am* getting a lot of sunlight right now, and don't drive that much.
Where's my electric option to cut me loose from the oil infrastructure? You know, the one that's actually technically possible and even feasable right now?
Windows is the only non-unix OS of any mention any more and we've always seen PC ports to unix apps - that what you get paid to do, vry few pc programs or worth porting to unix. So, the closer Windows is to being some form of unix compatible the better.
And there'e precedent for it. The architecture of dos/win is derived from CP/M which was a was written while looking at unix but only doing what could be done with an 8 bit processor and 16K or very expensive RAM. Keep in mind too Microsoft had Xenix - the first commercial unix ever sold.
I can almost see how this happened. A few of the biggest MS investors corner Ballmer in a bar: "Unless your company does something like sell Linux, we think you have no long term future. Apple has switched, Sun hasn't died. Look like unix is everywhere not Windows like you guys said" to which Ballmer replied "bah, we can do everything Linux can"
"Run Linux programs?"
"We'll have it on two years" (Note to self: remind Bill I did this)
The last time I asked about google doing dns I was told by a vp there "they aren't ready for that yet".
This is probably more political than anything else. Think about it. If google says "use us for dns" and they gave the fastest most reliable answer within a couple of months most of the world would be using it. Do you know what happens when you have most of the world using you as dns? They see what you say they see is the answer. If google were to slip in a.goog would that be a bad thing? You'd have control of the root zone, and all names on the net.
Right now, the "root zone file" that holds the list of TLD servers is under the control of the United States government - specifically the department of Commerce, which has in the past rejected ICANN's suggestions of modification for the root zone allegedly in return for a Bush staffer's political favour.
There are things both Microsoft and Google - but probably nobody else except possibly OpenDNS - could do in the next little while that would put them in a position of this sort of control.
When there's a monopoly of dns services, and I'd say more than 50%, then that entity gets to say what the root zone is, and experience has shown people can be convinced of the sense of any new plan when it comes to adding new tlds as long as new tlds actually get added. You'll notice the decade old process from ICANN has done SFA here.
At some point the internet community will get sick of somebody else saying they're in charge and deciding what domain names can or can not be published in a system we all ourselves run and provide the infrastructure for!
So, what I would do is use my own dns servers. And you should use your own dns servers. Or maybe you and your friends could se up your own root server network. One of you grab the root zone from ftp://internic.net declare yourself primary for ".", have the other guys slave the "." zone from you and stick each others ip's in your root cache file. Poof, you're a root server network.
You're still going to have the problem that port 80 (and 443) service has, or will soon have a "trasparent web proxy" - these intercept web requests then do the dns lookup on the domain name so they're sure to only cache web content their dns thinks is valid. You need to use a web proxy on the other side of this device to get unfiltered internet. And your own DNS.
Anything else and you're letting somebody else decide what you see. Log in to internic.net with ftp and cd to "domain" to get the root zone file you need to primary the "." zone for yourself.
Name (internic.net:r): anonymous 331 Guest login ok, send your complete e-mail address as password. Password: 230-Zone files can be found in the domain directory. 230- 230 Guest login ok, access restrictions apply. Remote system type is UNIX. Using binary mode to transfer files. ftp> cd domain 250 CWD command successful. ftp> ls 500 'EPSV': command not understood. 227 Entering Passive Mode (198,41,0,6,114,5) 150 Opening ASCII mode data connection for directory listing. total 160 -rw-r--r-- 1 9998 213 657 Feb 26 16:56 INTERNIC_ROOT_ZONE.signatures -rw-r--r-- 1 9998 213 680 Feb 26 16:42 arpa.zone.gz -rw-r--r-- 1 9998 213 75 Feb 26 16:53 arpa.zone.gz.md5 -rw-r--r-- 1 9998 213 72 Feb 26 16:54 arpa.zone.gz.sig -rw-r--r-- 1 9998 213 2876 Feb 4 12:07 db.cache <---------- list of ip addresses of root servers, ie, the NS records for "." -rw-r--r-- 1 9998 213 43 Feb 4 12:07 db.cache.md5 -rw-r--r-- 1 9998 213 72 Feb 4 12:07 db.cache.sig -rw-r--r-- 1 9998 213 2879 Feb 4 12:07 named.cache -rw-r--r-- 1 9998 213 46 Feb 4 12:07 named.cache.md5 -rw-r--r-- 1 9998 213 72 Feb 4 12:07 named.cache.sig -rw-r--r-- 1 9998 213 2878 Feb 4 12:07 named.root -rw-r--r-- 1 9998 213 45 Feb 4 12:07 named.root.md
Youtube had a route for 208.65.152.0/22 (208.65.152.0 - 208.65.155.255), but Pakistan's main ISP in Hong Kong announced a route for 208.65.153.0/24 (208.65.153.0 - 208.65.153.255) to keep youtube off their net. What they didn't understand though is this really needs to be kept as a local routing policy so it only affected Pakistan, but it sorta snuck out and affected the entire network.
Bernstein is quite right as always, but you don't care and unlike Bush, don't need to panic. There's lots of milage left in the protocol and those that think we're running out of addresses just aren't looking at the packet headers. They're in meetings instead.
You'd have a hard, if not impossible case to make it illegal. Believe me, if they could do this they would have by now.
The great beautry of the internet is it's "edge controlled" and there is no central control, that is everybody who has the root password to their little piece of it gets to say what their machine does. As such time as the government subsidizes tool becomes not the best tool to use and some other tool is superior, it'll be used instead.
You have to think of the big picture here and consider what happens over time.
Say you're looking for porn and you're presented with example.com and example.xxx.
Which ones sounds dirtier? That is whose gonna look at example.com when they can see example.xxx instead?
Keep in mind also if google sees the same website with two different domains, only one makes it into the index. That is to say, if you have both then example.com needn't show up at all or could trivially be filtered out by google's "safe search" criteria because it knows the same site exits in the.xxx namespace. Think of it as tagging but with domain names.
It's not a question of "what is porn" or "how do you regulate it" those are non issues. Over time the com namespace will simply become less and less dirty with porn slowly migrating to a more appropriate namespace.
You only need look at usenet to see an example of how this has worked in the past: alt.sex and alt.binaries.pictures were created to get porn off of mainstream usenet do they could be filterd out easily by those who wished to do so. This was 20 years ago and it worked and worked brilliantly. There is very little porn in "regular" usenet and tons of it in "the proper place" all because of self regulation. "If you build it they will cum" so to speak.
And this will work just as well in the DNS namespace.
That paper is utter crap. It's sheer fud. Read it caefully and makes notes of the suntantive points. They're garbage.
This thing came out in the middle of the "DNS wars" and was a smear tactic. Lots of cogent rebuttals were penned, but Eastlake, who wrote it, made his an iformative RFC - it's not binding, but it sure looks "official" having an RFC number doesn't it? This is called "proof by authority" and is a well known logical fallacy.
Keep in mind the people that allowed this paper to become an rfc were in on the icann process to get what they wanted -.org - and they got it... as their reward for supporting the creation of tbat bastard child born of secrecy we laughingly refer to as icann.
Slashdot Mirror
" So he's saying that the only way to stop the 'net from being placed under centralised control would be to place the 'net under central control? "
No, he saying where there needs to be some sort of aurhority, it should be a community of experts not some random government wonks.
The internet has no central control, it's edge controlled. There are a couple of single points of failure choke points like the root servers at the physical level, and ICANN at the political layer, but these can both be routed around and don't have any actual authority.
" John Nobody blathers on about something he knows very little from his summer place while sipping tea.
.com".
Why do Slashdotters buy the banal esoteric blather that comes from guys like these who have no real connection to reality? "
Huh?
Maybe you don't know anything about Zittrain but I do. Here's what I've seen.
In 1996 or so the Internic began charging for domain names. Immediatley there was consensus for new top level domains, or "life after
The community fractured into two camps: one that wanted a central authority, which hapoened to be them, and the other half that believed the insisible hand of Adam Smith would permit a self regulating industry to flourish.
Thems what was in charge shut out the rest which caught the governments attention. They stepped in and announced it was their baby and would settle things, and insitagted a process by which the replacement for Jon Postel and IANA would be formed.
The Berkman Center for Law and Technology at Harvard, under Charles Neeson was interested in this at it appeared to be a form of internet governance. Larry Lessig was there at the time and was slightly involved althouh he left shortly after. Zittrain was there are a while and was the real time scribe at nearly all te ICANN meetings.
He's seen first had what happens when expert self regulation is displaced by the collusion of corporate and government interests.
At least that's what I've seen that gives me reason to feel he qualified to comment on this.
"That article sure uses a a lot of words to say 'the web should be communist'. "
Rubbish.
The point is internet technology is so complex very few people understand how all of it works, and how it works all together. The further away you go from technical to admisistrative skillsets the less likely are people to understand what's going on. That's the difference bewteen SMTP actually working and a sock puppet raising venture capital.
This has nothing to do with capitalism or communism and is inappropriate for a framework of discussion about technology and what kind of environment open standards and processes need to flourish.
"Why on earth should he think that "experts" are any better at self regulation than any other random group of people?"
Because they're "experts" and not a random group of people.
Jono's quite right: frame it in this context - who would you put in charge of managaing, say, the Linux kernel? A bunch of guys that knew it best or a governmnet committee of people qualified to do something else?
TFA is wrong though when it says "this almost happened with domain names". Substitute "DNS" for "Linux" in the above and you have ICANN.
Jono was there as well, to watch this all come up. In fact Lessig and Zittrain were involved in the process that led up to ICANN and were as surprised as anybody else when the government stepped in and said "We know you've been working on this for a year all over the world, but here's the baord and here's the organization. Thanks, but you can go home now".
Can somebody explain to me what it is you can do on an iPhone app you can't do through a web page on a regular cel?
That is, it just glitz? Or is there any actual functional characteristic the iPhone has that sets it apart?
"I went to Waterloo and got through without taking out any loans[/quote]"
Me too.
Did yo notice this in TFA: "The university has received applications from as far away as Wilfrid Laurier University"
WLU is down the street about 4 blocks.
That phones may be able to ray trace is news? Sounds more to me like intel was of reading in the news all week how inferior their graphics stuff was because of the Microsoft Vista debacle part eight - and suddenly we have an anonymous tip to a blog at intel saying ray tracing on phones there is "an opportunity to deliver more content in less time" and "Ray-Tracing may become a very popular technology in the upcoming years".
A popular technology? Like a working filesystem? They're real popular I hear. Or an on off button that actually works.
Slow news day + intel graphics dept astroturfing = ray tracing on phones is news.
"Although I do love watching them flip out when I show them how my 24 year old, 5000 lb, loaded with bells and whistles, diesel Mercedes sedan gets 38mpg on the highway. "
38? Which one is that? a 190D? My 300SD gets 33 mpg at best. Booth's US and Euro 190D 2.5's got 45 and 48 mpg respecitvely.
I've driven a diesel for over a decade now and live way way out in the country. I've never had a problem finding diesel in Canada, not once, ever.
The US... that's different.
"Also, if you have an old engine you will probably need to clean the fuel filter after several thousand miles as the biodiesel will clean out the accumulated junk from regular diesel "
As if. A buddy of mine swiched in his 25 year old Mercedes diesel. He went through a case of fuel filters in 200 miles.
"Non-injected diesel engines don't have these issues "
There are non-injected diesels? Where?
[quote]300 MHz is pretty slow[/quote]
You would have really hated 4Mhz Z-80's then. Or the PDP-11's that wern't a whole lot faster. They did get the job done letting C and Unix get written on them though.
So... if 15 or 20 million Linux users sent Adobe a dollar would that help? Do they take Paypal? Ship internatioally?
What about Paint Shop Pro ? Is it nearly good eough? They may only need $5M to do the port.
I prefer "synthetic factoid" or "man made fact".
My kids just accuse me of "making up facts again".
Go rent a copy of "Who killed the electric car" then figure out what role the oil companies have in a hydrogen based economy.
Then understand why the Bush administration dumped millions into hydrogren resarch and never mind any running car is ten years off from whenever you ask.
We *might* be able to make hydrogen at home? Great. I *am* getting a lot of sunlight right now, and don't drive that much.
Where's my electric option to cut me loose from the oil infrastructure? You know, the one that's actually technically possible and even feasable right now?
It makes sense.
Windows is the only non-unix OS of any mention any more and we've always seen PC ports to unix apps - that what you get paid to do, vry few pc programs or worth porting to unix. So, the closer Windows is to being some form of unix compatible the better.
And there'e precedent for it. The architecture of dos/win is derived from CP/M which was a was written while looking at unix but only doing what could be done with an 8 bit processor and 16K or very expensive RAM. Keep in mind too Microsoft had Xenix - the first commercial unix ever sold.
I can almost see how this happened. A few of the biggest MS investors corner Ballmer in a bar: "Unless your company does something like sell Linux, we think you have no long term future. Apple has switched, Sun hasn't died. Look like unix is everywhere not Windows like you guys said" to which Ballmer replied "bah, we can do everything Linux can"
"Run Linux programs?"
"We'll have it on two years" (Note to self: remind Bill I did this)
"which is pretty impressive for a type of minnow."
They're not a "minnow" they're a killifish you insensitive clod.
I watched nexus to terrorism and I thought it was a great movie; my girlfriend thought it was great too.
We saw the sequel, but it wasn't as good. I'm hoping to see "Nexus to Terrorism III" ASAP.
The last time I asked about google doing dns I was told by a vp there "they aren't ready for that yet".
This is probably more political than anything else. Think about it. If google says "use us for dns" and they gave the fastest most reliable answer within a couple of months most of the world would be using it. Do you know what happens when you have most of the world using you as dns? They see what you say they see is the answer. If google were to slip in a
Right now, the "root zone file" that holds the list of TLD servers is under the control of the United States government - specifically the department of Commerce, which has in the past rejected ICANN's suggestions of modification for the root zone allegedly in return for a Bush staffer's political favour.
There are things both Microsoft and Google - but probably nobody else except possibly OpenDNS - could do in the next little while that would put them in a position of this sort of control.
When there's a monopoly of dns services, and I'd say more than 50%, then that entity gets to say what the root zone is, and experience has shown people can be convinced of the sense of any new plan when it comes to adding new tlds as long as new tlds actually get added. You'll notice the decade old process from ICANN has done SFA here.
At some point the internet community will get sick of somebody else saying they're in charge and deciding what domain names can or can not be published in a system we all ourselves run and provide the infrastructure for!
So, what I would do is use my own dns servers. And you should use your own dns servers. Or maybe you and your friends could se up your own root server network. One of you grab the root zone from ftp://internic.net declare yourself primary for ".", have the other guys slave the "." zone from you and stick each others ip's in your root cache file. Poof, you're a root server network.
You're still going to have the problem that port 80 (and 443) service has, or will soon have a "trasparent web proxy" - these intercept web requests then do the dns lookup on the domain name so they're sure to only cache web content their dns thinks is valid. You need to use a web proxy on the other side of this device to get unfiltered internet. And your own DNS.
Anything else and you're letting somebody else decide what you see. Log in to internic.net with ftp and cd to "domain" to get the root zone file you need to primary the "." zone for yourself.
No that's ok it's slashdotted anyway, at least as of this moment.
"The route was announced by AS17557"
Youtube had a route for 208.65.152.0/22 (208.65.152.0 - 208.65.155.255), but Pakistan's main ISP in Hong Kong announced a route for 208.65.153.0/24 (208.65.153.0 - 208.65.153.255) to keep youtube off their net. What they didn't understand though is this really needs to be kept as a local routing policy so it only affected Pakistan, but it sorta snuck out and affected the entire network.
Routing is the soft underbelly of the net.
"I think this article by Dan Bernstein is a pretty good read regarding this subject - http://cr.yp.to/djbdns/ipv6mess.html"
Bernstein is quite right as always, but you don't care and unlike Bush, don't need to panic. There's lots of milage left in the protocol and those that think we're running out of addresses just aren't looking at the packet headers. They're in meetings instead.
http://rs79.vrx.net/interests/computers/net/v6failure/
This is a non-issue.
You'd have a hard, if not impossible case to make it illegal. Believe me, if they could do this they would have by now.
The great beautry of the internet is it's "edge controlled" and there is no central control, that is everybody who has the root password to their little piece of it gets to say what their machine does. As such time as the government subsidizes tool becomes not the best tool to use and some other tool is superior, it'll be used instead.
You have to think of the big picture here and consider what happens over time.
.xxx namespace. Think of it as tagging but with domain names.
Say you're looking for porn and you're presented with example.com and example.xxx.
Which ones sounds dirtier? That is whose gonna look at example.com when they can see example.xxx instead?
Keep in mind also if google sees the same website with two different domains, only one makes it into the index. That is to say, if you have both then example.com needn't show up at all or could trivially be filtered out by google's "safe search" criteria because it knows the same site exits in the
It's not a question of "what is porn" or "how do you regulate it" those are non issues. Over time the com namespace will simply become less and less dirty with porn slowly migrating to a more appropriate namespace.
You only need look at usenet to see an example of how this has worked in the past: alt.sex and alt.binaries.pictures were created to get porn off of mainstream usenet do they could be filterd out easily by those who wished to do so. This was 20 years ago and it worked and worked brilliantly. There is very little porn in "regular" usenet and tons of it in "the proper place" all because of self regulation. "If you build it they will cum" so to speak.
And this will work just as well in the DNS namespace.
That paper is utter crap. It's sheer fud. Read it caefully and makes notes of the suntantive points. They're garbage.
.org - and they got it... as their reward for supporting the creation of tbat bastard child born of secrecy we laughingly refer to as icann.
This thing came out in the middle of the "DNS wars" and was a smear tactic. Lots of cogent rebuttals were penned, but Eastlake, who wrote it, made his an iformative RFC - it's not binding, but it sure looks "official" having an RFC number doesn't it? This is called "proof by authority" and is a well known logical fallacy.
Keep in mind the people that allowed this paper to become an rfc were in on the icann process to get what they wanted -
"he'll just fork over the cash for xxxample.xxx and operate both."
So?
You ever have two identical sites with two different names?
Guess what google does with them? It ignores one.