Truth In Erroring: IETF Proposal Includes New 451 Censorship Error Code
from the it'll-probably-be-nixed-by-the-Bradbury-estate dept
As nations around the world work to chip away at the liberated internet for such compelling reasons as “um, children!”, or “um, cyberwar!”, or even “um, piracy!”, all while lambasting other countries for censorship mind you, the general population may still be shrugging their shoulders in confusion. One reason for that may be a lack of an in-your-face approach to educating people as to what all this proposed policy may mean in their lives, which have only become more reliant on a functioning internet.
That's where internet developer advocate Tim Bray comes in. Business Week has the story of Bray's proposal to the Internet Engineering Task Force to add a new error code for websites: the 451 Censorship Code. You can read Bray's formal proposal here, but the short version is that any site that is unreachable due to legal reasons (normally because of ISP blocking, but not always), would return a 451 error along with a short explanation as to why the connection was blocked. So, for instance, if an ISP decided to block The Pirate Bay, you would get the 451 error along with an explanation that copyright fights were preventing the traffic (or something like that). The 451 code, of course, is an homage to Ray Bradbury's “Fahrenheit 451,” a book that most people think is about censorship, but Bradbury insists is just about how everything you love sucks, including the internet.
Many of you may have seen this circulated in June, but Bray is now saying that it will go before the IETF beginning this weekend:
“I've been told by the chair of the IETF HTTP Working Group that he'll give the proposal some agenda time at the next IETF meeting,” Bray told CNNMoney by email. “It's not a big proposal; shouldn't take long.”
When asked why he felt compelled to put this proposal together, Bray pointed to Terrance Eden (a UK blogger) who had put a call out for a censorship code after having a connection to The Pirate Bay blocked and receiving a common 403 error:
“Eden's Internet provider had been ordered to block out the site, but Eden wasn't happy with the 403 error response it generated.
'As far as I am concerned, this response is factually incorrect,' Eden wrote on his blog.”
That's because a 403 error is what you get when the server you're connecting to won't let you in. But that isn't what's happening in Eden's case. Eden's ISP is blocking the connection. That's a distinction that should matter to the consumer, who ought to be properly informed of who is keeping them from going where they want on the internet.
And that's Bray's point. It's not that traffic should never be blocked, but the user should know what's actually going on. As he says himself:
“I think most people agree that censorship is sometimes justified, but it's just common sense that when it happens, it should happen out in the open.”
Filed Under: 451, censorship, error codes, free speech, ray bradbury, tim bray
Comments on “Truth In Erroring: IETF Proposal Includes New 451 Censorship Error Code”
No body?
Right, because end users are actually looking at the HTTP response, instead of, you know, the actual body.
If the ISP blocking content couldn’t be bothered to display a useful message, then what makes anyone think they’d be bothered to return a 451 error code?
I hope this is more a homage to Ray Bradbury and not more nonsense IETF posturing.
Re: No body?
But this isn’t being sought by the IETF. This is a request TO the IETF for it to add the code to the spec for the HTTP protocol. How could that possibly be posturing by the IETF?
Re: No body?
No. Either a body is sent along with the code like most such http response codes or the browser would display a page based on the http response code.
My understanding is the text explaining why the site was blocked would be the message body.
>”I think most people agree that censorship is sometimes justified, but it’s just common sense that when it happens, it should happen out in the open.”
Censorship is *never* the answer. (imho).
Re: Re:
OF course. I understand that. Mike understands that. However, that’s an ideal right now. So, in the meantime, we have to come up with potential stopgaps.
This is a decent one. It would be especially interesting, if it had a link to the ongoing case reference (barring exceptional circumstances, of course).
Re: Re:
a desire to avoid riots and violent unrest due to inflammatory individuals behaving in a way designed to cause such for no good reason says otherwise. (to be fair: things censored by the chief censor’s office of NZ in the last few decades amount to the following: a shirt deemed to be excessively offensive to damn near everybody, a book which denied the holocaust took place at all (among other issues), said book’s Author (actually, i’m not sure if that was the chief censor or not, and it wasn’t entirely about the book) and various pornographic material deemed unlawful (mostly child-porn) they’re also responsible for asigning ratings to films, games and tv programs and the like, the R ratings being legally binding and carrying a penalty on selling them to those who are under-age. additionally, before something is even considered for censorship (except porn and things which require ratings stickers) someone has to actually have complained about it. then the censor has to also deem it to be a major problem.)
also: in times of war, it can be quite significant (and that’s where the office actually comes from.)
note this: there are a number of commonwealth countries where this office exists. most of them have substantially less censorship than the USA 😛
(that said, it’s VERY much one of those jobs where anyone who actually WANTS it should not get it, and probably should be arrested on sight… One of my mother’s relatives(from memory. certainly someone she knows.) actually had the job for a while. retired when he simply could not stomach the material which had to be checked any more. this is the stuff deemed so bad that it couldn’t even be given an R18 label.)
this post was funded by the World Society for Parenthetical Statement Awareness.
😛
IETF-sanctioned censorship?
Alas, such a code would mainly serve to legitimize the act of censorship. How can you say that censorship breaks the Internet when you have an official error code for censored websites?
Internet 3.0, now with built-in censorship capabilities.
Another non-political group making political statements. It’s disappointing to say the least.
Re: Re:
If you are getting censored why shouldn’t you be informed that you are being censored? Does it somehow magically remove the block or provide a different experience of inaccessibility?
Re: Re: Re:
Yes. I’d have about 1-2 seconds of pun-level amusement at the IETF’s wiseassery first, rather than just going straight to calling the ones man-in-the-middle-ing me cockbiting fucktards.
It’s like having to eat a shit sandwich and being told you can have it with or without bacon. Might as well take the bacon.
Re: Re: Re: Re:
For all the good it will do. Said cockbiting fucktards generally don’t read Techdirt or your personal blog.
If you said it to their face, on the other hand…
More seriously, I very much want to know the difference between error codes 403 and 451. In the former case, I’m not allowed in your house because you didn’t want to let me in and you wanted to tell me not to try again. In the latter case, it’s because you fear that Cato the Elder will take away your horse if you allow me in.
Re: Re: Re:2 Re:
not really, in the second case the owner of the house has nothing to do with it, you *are* allowed into the house, but cato is personally keeping you out.
Re: Re: Re:
Why bother? There is no need for it. It’s not a protocol’s position to decide what is censorship and what is just country blocking, as an example.
How are they going to determine specifically that it’s censorship?
It’s a bullshit political ploy by a group that should be neutral and STFU.
Re: Re: Re: Re:
Simple, really. Censorship is censorship. “Just country blocking” is censorship too.
Re: Re: Re: Re:
Sounds like someone has a vested interest in making sure those who are censored stay censored.
Re: Re: Re:
If you received 403, you’re not sure what you could do unless you have VPN channel to proxy at another ISP that can open it.
If you’re sure the website you’re going to is blocked accidentally (say, by having some keywards), you can write a complaint to ISP and request them unblock it.
That’s the difference.
Re: Re:
Yeah, I too wish for the time when everyone just shut up and did nothing about anything.
“The 451 code, of course, is an homage to Ray Bradbury’s “Fahrenheit 451,” a book that most people think is about censorship, but Bradbury insists is just about how everything you love sucks, including the internet.”
Well, it works both ways. It’s either the “censorship code”, or the “fuck you internet code”.
This post has been censored, for the cyber pirate children’s sake.
Re: Life of ???
I think it should be for the sake of the pirate cyber children, Splitter!
I am trying to figure out how they figure out what is censorship, and what is just “not available current” or “not available in your country”. Will accessing Hulu from outside the US get you a 451?
Will they form a committee and review each page before assigning it the code? Is there a way to detect censorship in code that will allow the to be returned automatically?
I doubt it.
Bullshit political stand from a non-political group.
Re: Re:
This is not a censorship-only code. Read the actual proposal:
http://www.tbray.org/tmp/draft-tbray-http-legally-restricted-status.html
It’s just to differentiate when an ISP is legally required to block stuff. So, sure, if some moron refuses to license something to certain countries, 451 provides a more precise error message.
Hulu, AFAIK, does not return any error _code_. Because the actual code is irrelevant in most cases. Instead they give you a nice 200 OK and pop up a little message. Really, the error code is totally unnecessary, and is more of a “cute” thing than anything serious.
Re: Re: Re:
Well, the fact is that licensing should be BLANKET licensing to release things in any country where that thing is legal.
None of this “License for the U.S.”, “License for Australia”, and “License for South Africa” bullplop. That is just an attempt at copyright owners trying to make their stuff appear more ‘valuable’ by making companies/people pay multiple times for content.
Re: Re: Re:
There could be requirement added to pass thru a request to the server so that the server can log the error allowing the server admin to know how many requests were blocked. The pass through request should include…
1. Requested URL
2. Origin IP
3. Blocking IP
4. Something to indicate if it is government ordered, ISP ordered, or a 3rd party requested block like DMCA.
This could be useful as site owners would then also have access to log data that could possibly be used in challenges to the blocks.
Re: Re: Re: Re:
There could be requirement added to pass thru a request to the server so that the server can log the error allowing the server admin to know how many requests were blocked. The pass through request should include…
That would be a nice to have, but it is doubtful, given the current climate, that that would possibly occur. ISPs are told to block a site, which they do. There is nothing in the protocols that allow part of the connection to go through and then deny the rest, unless they do some flag blocking (like allow the SYN to go through and then kill everything else.) However, that would damage the internet as it would put some serious resources on the server in question (who is queuing up the connections based on SYN packets, and must time-out those connections when the final ACK is not received,) and anything the ISP does to modify the SYN packet adding error information will be lost because the software doesn’t record anything (or expect anything) in the SYN packet. This would only be something they could really accomplish via email, or some other protocol, and I don’t see them building an infrastructure just to alert a 3rd party to their blocking.
Besides, it would be much easier to offload that on their customer, who can contact the site through other means and let them know they are being blocked.
Re: Re: Re:2 Re:
I wasn’t suggesting allowing part of the packet to go through necessarily. It could be mandated in the spec that a provision be made that allows a new request to be sent through with information retrieved from the original packet, much in the same way that email servers send bounce messages back to the original sender. However I do agree with you about the climate thing except those that write RFC’s generally are on this side on the side of the issue where transparency is seen as a good thing so getting something like this added to the spec could IMO actually happen. Of course just because you put it in the spec doesn’t mean there won’t be those that choose to ignore that part. However, I think adding a provision to the official protocol spec that allows for both visitor AND site owner to know when a request is being blocked by a third party for legal reasons would be a good thing. And of course passing through info when a block is implemented to mitigate an attack of some sort, however those are usually done at the request of the site holders anyway so they already are aware of what is going on.
Re: Re: Re:3 Re:
However I do agree with you about the climate thing except those that write RFC’s generally are on this side on the side of the issue where transparency is seen as a good thing so getting something like this added to the spec could IMO actually happen.
I think transparency is a great goal, and agree that there should be something technical that can be done to add transparency. I agree though, most ISPs will ignore any RFC, but I hope some would follow it, and they would get my money (if they offered service in my area.) We have so few choices though. I’d happily move to KC, Mo right now though, since they are soon to have a kick-ass ISP.
Re: Re: Re:4 Re:
However, adding it to the RFC DOES do one thing. It allows a basis by which those who follow the RFC can claim that they are compliant with the accepted standards and criticize those who do not follow them.
Actually, censorship is NEVER justified. 99% of the time when someone is talking about ‘censoring’ something, they are actually hoping to push it ‘underground’ so that it won’t come out, become popular, and become normalized.
They tried it with interracial marriages, they tried it with homosexuality, hell…. they even tried it with heterosexuality outside of ‘marriage’ (a purely human invention).
Bottom line is that we should not allow any censorship and tell people “You don’t like X? Fine, you don’t have to do it, but leave the people alone who want to do it unless they are physically forcing someone else into something against their will!”
Another good error code is 1984.
Re: Another good error code is 1984.
Had to be either 4xx (client-side error condition) or 5xx (server-side error condition).
Re: Re: Another good error code is 1984.
4891
I agree. I would like this error code to be implemented, especially on file-sharing sites. It would be nice if the file-sharing sites listing the party/parties that hit that link too.
Re: Re:
Nah, for file sharing sites, it’s more like:
499: Paypal processing removed, site closed.
File lockers are falling like flies..and Mike is nowhere to be seen on the issue.
it'll get ignored, just like the 410 error
Seriously. Look up the spec. The entry for 404 says, “The 410 (Gone) status code SHOULD be used if the server knows, through some internally configurable mechanism, that an old resource is permanently unavailable and has no forwarding address”. When’s the last time you got a 410?
Re: it'll get ignored, just like the 410 error
A site I manage serves 401 where applicable.
Google already does a form of internet censorship. It’s called “I think there is Malware here!” with a bright red screen and ominous looking text. It scared the hell out of me. Especially since I hosted my web site at GoDaddy and was rewarded with their service of Malware infection. The problem is getting it removed. Since Google has no phone number listed anywhere I can find. Who do you call? We just closed down the site and hosted our own servers from then on. We were forced out of business by Google!
Google Malware Warnings
There is a process for handling those…
http://support.google.com/webmasters/bin/answer.py?hl=en&answer=168328