Talk:Warrant canary

Legality
Many countries subscribe to the common law, including the USA. The common law provides for punishment for violating the spirit of the law, rather than its letter. So it would seem that a warrant canary, as a loophole, could be closed by the courts, under the argument that saying that you have not, and then removing the statement is the same thing as saying that you had. And, indeed such decisions regarding other issue have been made.

However, at least in the USA, the argument is that the First Amendment trumps the common law here, meaning attempting to enforce the spirit of the gag order is impossible. Disallowing putting up the canary in the first place violates free speech restrictions. There is no law preventing it, and even if there was, it would be unconstitutional. Forcing a false update of the canary violates the restrictions on compelled speech. Compelled speech has been allowed in the past, but only if it is truthful. The usual use of this is to force true warning labels and ingredient lists. There has never been a single case in the US where compelled false speech, or even compelled speech that is neither true for false, has been upheld. In fact, it has been struck down every time. The only form of compelled speech allowed in the USA at all is truth. This is the legal case for warrant canaries in the US. Such stringent protection against compelled speech may not exist in other countries, and the legality of a warrant canary there cannot be determined so easily. Foreign governments may or may not have the ability to compel false speech. If they do, a warrant canary is useless. 74.211.60.182 (talk) 22:33, 23 September 2016 (UTC)

Usage
A quick google gives me the impression that rsync.net is the only provider ever to have used this method. If there are others, the article would benefit from mentioning them. 92.195.60.23 (talk) 16:16, 25 January 2009 (UTC)
 * Anyone spotted one of these? --Elvey (talk) 09:49, 1 December 2009 (UTC)
 * My public libraries used these after the patriot act was enacted. --Gmaxwell (talk) 01:09, 13 June 2013 (UTC)

Image
I favor continued inclusion of the free, CC-licensed image, and don't quite understand the rationale for its removal. Discuss? --Lexein (talk) 16:20, 13 June 2013 (UTC)

"Tested in court" citation
I question the inclusion of "The legality of this method has not been tested in any court.[1]"

ArsTechnica article cited: http://arstechnica.com/tech-policy/2013/11/apple-takes-strong-privacy-stance-in-new-report-publishes-rare-warrant-canary/

The citation is basically circular. The ArsTechnica article links back to this wiki page, which cites that article as a source, and neither has anything more definitive. How do we know it was not tested in a FISA court under seal? The quote in the article is written as the author's knowledge, and not quoted from his expert source. Please let me know if there is a more constructive way for me to raise this issue. This would be my first substantial edit, so I don't want to just change a page that is probably getting a lot of hits right now without some validation by other editors.

Gabe (talk) 10:08, 8 November 2013 (UTC)Gabe Kneisley (gabekneisley)
 * Concur. If nothing comes up in a search for "warrant canary" "case" on Lexis/Nexis or other legal journal or even notable-author legal blogs, that claim (and maybe ref, if it's not used elsewhere) can be deleted, IMHO. "When in doubt, leave it out", and see WP:CIRCULAR. My searches turned up no comment by a notable legal commentator or Lexis/Nexis. --Lexein (talk) 11:02, 8 November 2013 (UTC)
 * Followup: I asked at WP:RX. Some commentary found at EFF, but nothing definitive about use in court yet. --Lexein (talk) 23:49, 10 November 2013 (UTC)
 * Further followup: I support removing the claim because it was unsourced since the article was created in 2008, and is now suspected with good cause to be circular, since it appears to have been quoted without attribution by ArsTechnica, which was then cited here. I've removed it. --Lexein (talk) 03:37, 15 November 2013 (UTC)


 * Lexein, I promised to follow up if I came across anything further, and here it is. The Yale Law Journal Forum has posted a student piece on warrant canaries.  The article mentions a pending lawsuit by Twitter, which seeks to establish the legality of warrant canaries.  The Twitter lawsuit, incidentally, calls them by the more formal title, "transparency reports," rather than the less formal but more distinctive "warrant canaries."  The suit is currently pending in the U.S. District Court for the Northern District of California.  John M Baker (talk) 18:59, 29 January 2015 (UTC)


 * Hmmmm. Should we consider merging/redirecting Transparency report into Warrant canary? --Guy Macon (talk) 16:26, 30 January 2015 (UTC)


 * Yes, I would think so. "Transparency report" is actually a broader term than "warrant canary," but the transparency report article seems only to cover warrant canaries.  John M Baker (talk) 14:03, 31 January 2015 (UTC)

Canary in a Coal Mine
The phrase "warrant canary" alludes to the proverbial 'canary in a coal mine'. Both canaries expire when poisoned, and thus alert their watchers. See page 168 of a 1968 US Department of the Interior publication Walter Dufresne (talk) 15:28, 23 December 2013 (UTC)

Potential source material
Here's some potential source material or material that may lead you to source material for the article:
 * 4 January 2004 Washington Post - These subpoenas are secret
 * August 26, 2005 ACLU seeks to "lift a gag order preventing the target of the secret FBI subpoena from revealing itself."
 * August 26, 2013 Court accidentally leaks Google's name on request to reveal a secret subpoena.

-- Jreferee (talk) 15:39, 3 January 2014 (UTC)

Confusing word
In the first paragraph, the following appears: "A warrant canary may be posted by the provider to inform users of dates that they have not been served a secret subpoena" The word "dates" doesn't make sense to me. Should that be "data"? I'm not changing it, because it might be correct, but in a way that I don't understand. Bunthorne (talk) 02:34, 29 January 2014 (UTC)

No, it means dates, as in "We haven't been served a warrant or NSL during January of 2014", "We haven't been served a warrant or NSL during March of 2014" with the conspicuous absence of February being a clear message that they *were* served a warrant sometime between Feb 1 and Feb 28. 159.153.4.50 (talk) 02:15, 6 June 2014 (UTC)


 * That said, once the first warrant canary is killed, the law can prevent you bringing in another one, as at this point the law has attached, and the gag order is in place. The only reason the first one can be put up is that the law has yet to attach if you've NEVER received one. While removing the first one is not speech, putting the next one back in IS speech in violation of the gag order.  — Preceding unsigned comment added by 50.21.206.23 (talk) 18:42, 30 December 2017 (UTC)

TextSecure
"The US security researcher Moxie Marlinspike states that 'every lawyer we've spoken to has confirmed that [a warrant canary] would not work' for the TextSecure server.[15]" The same GitHub page links to an [EFF blog post](https://www.eff.org/deeplinks/2014/04/warrant-canary-faq) that contradicts the conclusion in the page. — Preceding unsigned comment added by 188.24.181.238 (talk) 15:01, 14 December 2014 (UTC)

Differentiating between subpoenas and orders
There are a bunch of different types of methods that the government can use to obtain this type of information. Here is a rundown of what I understand about the types of orders available:


 * Subpoenas. These are what everyone is used to. They themselves have a weird provenance and a multitude of subtypes, but its simple enough.
 * Pen register and trap and trace device orders (aka wiretaps). Authorized by et seq. I don't know if these are subpoenas or not.
 * National security letters/national security gag orders. Authorized by, part of the Stored Communications Act, itself part of the Electronic Communications Privacy Act of 1986, which provides that "no wire or electronic communications service provider, or officer, employee, or agent thereof, shall disclose to any person" etc. etc. The Wikipedia article identifies them as a type of subpoena.
 * 2703(d) orders. So-called because they are authorized by, also part of the Stored Communications Act, which clearly differentiates such orders from "an administrative subpoena authorized by a Federal or State statute or a Federal or State grand jury or trial subpoena". It may be, as a matter of simple taxonomy, that they are a type of subpoena, but I don't see them referred to as such. For example, Orin Kerr (in the Volokh Conspiracy) differentiates between subpoenas and 2703(d) orders. These were served on Twitter in relation to WikiLeaks.
 * 2705(b) gag orders. So-called because they are authorized by, also part of the Stored Communications Act, which can also apply to subjects receiving a warrant or subpoena. The persons receiving the subpoena, warrant, or 2703(d) order "is directed, for such period as the court deems appropriate, not to notify any other person of the existence of the warrant, subpoena, or court order." Even here 2703(d) orders are differentiated from subpoenas.
 * 3123(d)(2) gag orders. I don't know if that's what they're called, but they are authorized by, part of the Electronic Communications Privacy Act of 1986 but not part of the Stored Communications Act. Recipients of wiretaps are required to be ordered by the court that they not disclose the existence of the devices or the investigations.
 * 5318(g)(2) gag orders. Authorized by, as far as I can tell part of the Annunzio-Wylie Anti-Money Laundering Act, itself part of the Housing and Community Development Act of 1992, which authorizes suspicious activity reports and that "neither ... may notify any person involved in the transaction that the transaction has been reported". (Not so relevant, but just to drive home how fragmented the exceptions to freedom of speech are.)

I think this article equally applies to these other (than subpoena) mechanisms, and we should consider a more expansive description of the covered mechanisms to include more than just subpoenas and NSLs. Int21h (talk) 03:01, 4 February 2015 (UTC) Int21h (talk) 03:12, 4 February 2015 (UTC)

Warrant Canary List?
Would it be appropriate to provide a new section at the bottom of the page titled something along the lines of "Company's with Warrant Canaries"? My thoughts here are that there are not very many companies that do it as of now and I know several firms have reliable sources backing this. I'm sure most, if not all, have sourcing.DaltonCastle (talk) 04:48, 6 March 2015 (UTC)
 * Yes. Int21h (talk) 02:27, 7 March 2015 (UTC)

@User:Dodi 8238, lets discuss the Canary list. I understand the concerns here. There is a source that published that list but I would consider it primary. (https://canarywatch.org/). I know policies on primary sources well: its generally best to avoid them but there are times when they are acceptable. This could be a helpful information section to a reader on the page however. Im sure some of the firms and organizations on this list have reliable sources for their individual warrant canary. What are your thoughts here? DaltonCastle (talk) 21:26, 20 June 2015 (UTC)


 * The edit in question. I removed the list because it appeared to be original research. Why wasn't Canarywatch cited? Also, many of the companies and organizations that were listed are probably not notable enough to have articles, so they shouldn't be redlinked. --Dodi 8238 (talk) 22:00, 20 June 2015 (UTC)


 * Well, not sure why it wasnt added before. As far as notability of the pages, I dont doubt some of them are not notable, but Im sure some are. And I do a lot of article creation, especially with tech companies. I can do my best to create some of these pages in the near future. Do you think we can add this info back then? DaltonCastle (talk) 22:33, 20 June 2015 (UTC)


 * I restored the list. I also added citations to Canarywatch and a few entries that were on Canarywatch but that weren't included in the list when I removed it. Feel free to start articles if you think the subjects meet WP:GNG. --Dodi 8238 (talk) 22:50, 20 June 2015 (UTC)


 * Great! Thank you! Ill create some of these pages soon. DaltonCastle (talk) 22:59, 20 June 2015 (UTC)

Aim of a warrant canary

 * A warrant canary is a method by which a communications service provider aims to inform its users that the provider has not been served with a secret government subpoena.

I'm not sure if that really gets the idea across. The canary does inform its user that the provider has NOT been served, true - but is that really the aim of a warrant canary? Isn't the aim to inform the user - passively - when they HAVE been served? — Preceding unsigned comment added by 5.42.129.104 (talk) 22:28, 13 March 2016 (UTC)

The words "aims to" should be deleted, and inform changed to informs. Because that's what the canary does. It aims to inform it's users that a secret subpoena HAS been issued without violating the gag order. 50.21.206.23 (talk) 18:57, 30 December 2017 (UTC)

Warrant canaries found to be legal by US Justice Department?
I read the articles linked in the citations for the statement "Warrant canaries have been found to be legal by the United States Justice Department, so long as they are passive in their notifications.[3][4][5]" None of these articles state that warrant canaries have been found to be legal, only that companies are arguing that they should be and that said companies believe the would be found to be if tested in court. But as far as I can tell no such test has yet happened, and in fact, the articles state the opposite in that some companies have been instructed in a letter by the government they may NOT use canaries, and Apple's removal of their canary indicates that one of two possibilities is that they are abiding by that letter. Vertigo220 (talk) 22:29, 10 April 2016 (UTC)

Yeah, as near a I can tell, it's impossible to legally prevent the removal of a warrant canary. But a law could be made to disallow putting them up in the first place (the law could murder the canary first). Additionally, common law allows a judge to attempt to rule on the spirit of the law and close loopholes. A warrant canary clearly violates the spirit of the law, without violating it's letter. Therefore, a prosecution and conviction is definitely possible, and would be a case of judicial legislation, effectively enacting a law against putting up the canary in the first place. Sich things should be tested in court. 74.211.58.163 (talk) 17:52, 20 September 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 1 one external link on Warrant canary. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Corrected formatting/usage for http://www.librarian.net/technicality.html

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.—cyberbot II  Talk to my owner :Online 12:51, 16 April 2016 (UTC)

US Law
The paragraph covering US case law under the usage section seems like original research, or at best unattributed from a comment on the Bruce Schneier blog previously cited https://www.schneier.com/blog/archives/2015/03/australia_outla.html#c6692535

5.151.178.168 (talk) 18:31, 17 June 2018 (UTC)

For all paranoid Wikipedians out here, a convenient Userbox
-- Wesha (talk) 03:45, 21 April 2022 (UTC)

Why and how?
This article might benefit from discussion of why the communication providers would do this. Is it out of loyalty to their users, or as an act of civil disobedience? Also, how is this information "published"? On a dedicated page of their website or in communications? Cerulean Depths (talk) 21:21, 9 December 2022 (UTC)