Dubfire

I am massive fan of Dropbox and like to see it doing well. I have talked about its virtues on many Quora posts. I have ussually recommended using a Truecrypt volume within the Dropbox folder to guarantee encryption. Now reading findings like this and the authentication issues exposed in the last few weeks I'm getting quite worried about the security of the whole solution. Going to have to change a lot of my posts to say do not store anything sensitive without Truecrypt: http://paranoia.dubfire.net/2011/04/how-dropbox-sacrifices-user-privac…

The overreach is in expanding the original mission of DNT, which was to request that sites not harvest a user’s data for targeted advertising: http://paranoia.dubfire.net/2011/01/history-of-do-not-track-header.htm… Expanding that to mean “do not reveal the user’s IP address” is a significant expansion of the original meaning, having little to do with the protection against being individually tracked that the DNT header was designed for. Asking Homebrew to disable analytics has nothing to do with targeted advertising opt-outs, but supports the ideological goal of the authors in promoting their new idea that revealing the existence of the user is “tracking”. That expansion exceeds the scope of the definition of “tracking” as used to reference targeted advertising and marketing, something that is not applicable to the Homebrew project’s existing flag in any way. Attempting to promote their beliefs by aliasing those two concepts together - tracking protection against targeted advertising, versus disabling telemetry and auto update - is therefore overreach to me, especially when presented without this context to projects such as Homebrew and others in pull requests, and I think it will ultimately lead to their effort’s failure. To me, this is all a direct consequence of their attempt to ride the coattails of the failed DNT effort by reusing its key phrase “do not track”. They should have used a different name. In a few seconds, I can think of a pretty great one that still…

http://www.law.cornell.edu/uscode/text/47/1002 "(3) Encryption A telecommunications carrier shall not be responsible for decrypting, or ensuring the government’s ability to decrypt, any communication encrypted by a subscriber or customer, unless the encryption was provided by the carrier and the carrier possesses the information necessary to decrypt the communication." More info: http://paranoia.dubfire.net/2011/02/deconstructing-calea-hearing.html Also, declan, I was in the room during meetings with the FBI when I worked for a large telecommunications carrier. I integrated the CALEA mediation platform with the IP network and I'm well aware of what is required and not required to be present in the network regarding CALEA. IANAL, but you would be wise to consider that the FBI considers the former "information service providers" to be "telecommunications providers" to the extent that they can convince a judge that they are acting as one. I wouldn't think I was safe because I was using any kind of messaging where I couldn't control the keys, the software that uses them, and the distribution and verification of said keys.

I suspect it was just the conservative choice, mirroring the behavior of the predecessor browser extension (which also required deliberate user action to activate): http://paranoia.dubfire.net/2011/01/history-of-do-not-track-header.htm… I couldn't find any explicit discussion of that aspect of the spec (https://datatracker.ietf.org/doc/html/draft-mayer-do-not-track-00 is an early version), but I suspect that as noirscape wrote, it made consensus-building easier. Not that it mattered in the end, of course.

> Assuming any image or audio stream is available on the internet (not true but very close), you can have a super large compression by replacing the media by their url. That reminds me of the "Dropship" utility: to save server storage and reduce file upload time, Dropbox used to deduplicate uploaded files globally in a way that files would have the same hash regardless of who uploaded them. Anyone who knew the hash of an uploaded file could download it to their Dropbox folder, so people could share large files by just sharing the Dropbox hash. http://paranoia.dubfire.net/2011/04/how-dropbox-sacrifices-user-privac…

To be fair, the DNT launch was botched from the beginning, starting more as hack than an industry-wide consensus [1]. While it eventually got implemented by browsers, it lacked adoption, and had risks with fingerprinting [2]. The nail in the coffin was when Internet Explorer 10 decided to enable it by default [3], completely disregarding user intent. [1] http://paranoia.dubfire.net/2011/01/history-of-do-not-track-header.htm… [2] https://www.macworld.com/article/232426/apple-safari-removing-do-not-t… [3] https://en.wikipedia.org/wiki/Do_Not_Track#Internet_Explorer_10_defaul…

I'm not sure how it overlaps with other laws, but at least there's this about CALEA: > not only does the law speak about encryption, but it specifically protects the right of companies to build strong encryption for which only the customer has the decryption key into their products. http://paranoia.dubfire.net/2010/09/calea-and-encryption.html If that's true, then companies could actually use CALEA on their side to implement true end to end encryption, that they can't decrypt themselves, for their customers.

> "DuckDuckGo can easily be compelled either under the Communications Assistance for Law Enforcement Act (CALEA), standard court orders, or by secret orders from the Foreign Intelligence Surveillance Court (FISA) to provide tap-on-demand" Can they actually do that? I mean it's one thing to just "hand over the data" you already have about the user, and maybe even compel the company to decrypt it (although I still think that's BS [1] and companies should fight against it), but can they actually force a company to spy for them, and change their service in such a way that makes it possible? Does FISA and the Patriot Acts actually imply that? Or does he mean it might be yet another one of their "interpretations" of the laws? Either way, if that's possible, just start using StartPage.com. They're based in Norway. [1] http://paranoia.dubfire.net/2010/09/calea-and-encryption.html

CALEA does NOT require Microsoft to provide decryptable communications services; in fact it ensures Microsoft can do exactly the opposite. http://paranoia.dubfire.net/2010/09/calea-and-encryption.html

For those interested, there is a dissection of Yahoo!'s statement by Chris Soghoian (privacy and security research, policy expert/technologist @ the ACLU): http://paranoia.dubfire.net/2013/06/analyzing-yahoos-prism-non-denial.… Like the other official/legal postings so far, the verbiage in this post is carefully crafted to mean the opposite of what it says.

The government's theory is that a national security letter is sufficient to get access to your data. No warrant required. And Dropbox is not allowed to tell you that it happened. And yes, they can give your data to the government. Communications to/from Dropbox are encrypted. But it is unencrypted on the back end. See http://paranoia.dubfire.net/2011/04/how-dropbox-sacrifices-user-privac… for how we can know that.

I didn't know this. Although as a tech reporter I should have, I would argue that typical people are unaware of these kinds of architectural implications and that without obvious disclosure this design decision sort of takes advantage of users. Customers are paying twice, for the service and for other people's data, and it shouldn't take a Hacker News-level understanding of technology for them to be aware of this fact. It feels icky and reminds me of when I found out that Dropbox was "sharing" my files with other users: http://paranoia.dubfire.net/2011/04/how-dropbox-sacrifices-user-privac…

Dropbox used to be vulnerable to this because of their back-end deduplication. Users could know if a file had been uploaded already by seeing if the dropbox client had to upload it, or if the service "recognized" the hash. http://paranoia.dubfire.net/2011/04/how-dropbox-sacrifices-user-privac… Actually an FTC complaint was filed and Dropbox had to give a up a few of the benefits of deduplication.

Direct link to security researchers blog. (http://paranoia.dubfire.net/2012/03/fbi-seeks-warrant-to-force-google-… PDF of the application. (http://www.archive.org/download/gov.uscourts.casd.378626/gov.uscourts.…

The first time someone pointed this out, the FBI raided his house[1] and sparked a Senate investigation. This was four years ago. I did this to one of my Southwest tickets recently, though didn't use the forged copy. Honestly, it's like they think HTML is unreadable, or, more likely, that it's security theatre designed to make everyone feel safe. I would be okay with that if it wasn't taken so seriously. [1] http://arstechnica.com/security/news/2008/06/tsa-defiant-passengers-wo… Edit: The Soghoian blog post about the raid: http://paranoia.dubfire.net/2006/10/fbi-visit-2.html

Couple this with the unencrypted metadata on mobile problem: https://grepular.com/Dropbox_Mobile_Less_Secure_Than_Dropbox_Desktop And how their "encryption" on the server side is basically a lie, as they do dedupe on data: http://paranoia.dubfire.net/2011/04/how-dropbox-sacrifices-user-privac… I'm stunned that anyone would use them for anything for ephemeral data you wouldn't mind posting in public.

T-Mobile was the only telco that has avoided logging the goods that the feds want: http://paranoia.dubfire.net/2011/03/negative-impact-of-at-purchase-of-…

The speculation I'm seeing elsewhere is that this was 8 million pings sent by a much smaller number of survielled phones, with each sending pings every 30 minutes or whatever. Anyway, the original blog post has much other interesting info, despite the probably inflated numbers in its title. <http://paranoia.dubfire.net/2009/12/8-million-reasons-for-real-surveil…

Gosh, if DP wasn't so convenient and worked so well. They've always given off a "not quite right" vibe from the corporate level that I can't quite put my finger on ever since the TC Cribs video [1]. The weird public responses surrounding privacy and security issues since then have kind of reinforced this feeling. [2][3] But it's so convenient and works so well you kind of quickly forget about all the fuss. And even this [4] just sort of blew by like a summer storm. It goes to show how offering a really good service can make all kinds of principles take a back seat I guess. Let's see how many people offering to boycott DP are still on the service in 3 or 4 months now. 1 - https://www.youtube.com/watch?v=RAfIDUTPJJM 2 - http://venturebeat.com/2012/08/01/dropbox-has-become-problem-child-of-… 3 - http://paranoia.dubfire.net/2011/04/how-dropbox-sacrifices-user-privac… 4 - http://arstechnica.com/tech-policy/2014/03/dropbox-clarifies-its-polic…

> i don't think this is the correct interpretation. in a court of laws, acquiring evidence is something procedural and governed by rules and regulations. having a third party (lavabit) acquire the evidence and then turn it over to the government is probably something that wouldn't pass muster in court due to chain of custody and other rules. Search for "$" on this page: http://paranoia.dubfire.net/2009/12/8-million-reasons-for-real-surveil… It is very common for third party service providers to search records themselves on behalf of law enforcement, and law enforcement has historically trusted that and even compensated them. It's the only way that really makes sense IMO. Can the FBI really bust down, say, Verizon's doors, seize all their hard drives, and correctly generate evidence based on systems that aren't theirs, and may in fact be unique among the phone systems of the entire world? Maybe an email system is simpler, but still. Also, I think it's inappropriate to seize 400,000 users' data just to (ostensibly) get to one person. But I'm not a lawyer, maybe it is arguably legal. P.S. Also see https://www.eff.org/files/filenode/social_network/Yahoo_SN_LEG-DOJ.pdf for all the things Yahoo does on behalf of LE.

Apple makes money by selling hardware and software, whereas Google and Twitter make money selling ads, which incentivizes them not to encrypt our data so that it can more easily be used to target users. Or at least that's what this post seems to imply: http://paranoia.dubfire.net/2011/11/two-honest-google-employees-our.ht… I'm not convinced that this story is necessarily reliable, but it's better than the reverse: http://news.cnet.com/8301-13578_3-57577887-38/apples-imessage-encrypti… And in Apple's statement on privacy: https://www.apple.com/apples-commitment-to-customer-privacy/. "For example, conversations which take place over iMessage and FaceTime are protected by end-to-end encryption so no one but the sender and receiver can see or read them. Apple cannot decrypt that data. Similarly, we do not store data related to customers’ location, Map searches or Siri requests in any identifiable form." As John Gruber says, "That last sentence separates Apple from many other companies." (http://daringfireball.net/linked/2013/06/17/apple-privacy)

> If they're going to supply access via the website, or allow you to share with others then they need to be able to decrypt the file. They need to be able to decrypt the file right then to give you access via the website. They don't need to be able to decrypt the file the rest of the time. If the encryption key depends on your password, this is easy to implement. In modern browsers, this could actually be done in the browser, so that the keys for decrypting the file never reach their servers. To allow you to share individual files with others, they could encrypt each file with a randomly-generated session key, which is encrypted with your user key. When you give access to the file to a recipient, you would be encrypting a copy of that session key with the recipient's public key. In actual fact, it appears that anyone who knows the secure hash of the file's content is able to discover that Dropbox is storing it: http://paranoia.dubfire.net/2011/04/how-dropbox-sacrifices-user-privac… — and there's at least well-founded speculation, if not proof, that this means that knowing the secure hash is sufficient to read the file: http://news.ycombinator.com/item?id=2440074 For their previous statements to be true — that their employees have no access to your files — they would have to be doing something like this.

[0] He heard about this in 2005 (when the government was run by a different administration) but couldn't find anything. Eventually he filed a FOIA request. Silence, until spring of last year, when he was told there was some some information but it was redacted. So he filed an appeal, and it was granted. I suspect the lack of DoJ comment is because they're still trying to find out whether it was a fully formed program or just junior official's bid for career visibility - I suspect the latter, given the extremely basic level of information on these slides and lack of other documentation. It's unclear whether the question left with the DoJ was about this 'hotwatch' project or about administrative subpoenas in general. An AS is a request for infor information issued by a government agency under statutory authority (ie Congress passed a law specifically saying an agency may issue one in an investigation and it must be complied with). About 300 government agencies are empowered to issue them altogether, although usually within a narrowly defined scope. The obvious constitutional problems with administrative subpoenas have been known for some time [1]. Misuse of such procedures has resulted in disciplinary proceedings at the FBI, among other agencies [2]. The basic thing to remember is that legally speaking warrants are not required for this kind of information gathering because it does not constitute a search [3] - as a legal term, this may not always mean what you think it…

This is most disconcerting. Here's a bit of background: The original NYTimes Charlie Savage article from September breaking the news the that "crypto wars" may be heating up again: http://www.nytimes.com/2010/09/27/us/27wiretap.html Blog pointing out that current law specifically allows for encryption, and that what the Feds are after could be a significant change: http://paranoia.dubfire.net/2010/09/calea-and-encryption.html Two good articles from networkworld: http://www.networkworld.com/columnists/2010/100610antonopoulos.html?fs… http://www.networkworld.com/newsletters/sec/2010/101810sec1.html The EFF's position: https://www.eff.org/deeplinks/2010/09/government-seeks

Quinn Norton wrote an article for Wired describing cryptocat and headlined "This Cute Chat Site Could Save Your Life and Help Overthrow Your Government" http://www.wired.com/threatlevel/2012/07/crypto-cat-encryption-for-all… Chris Soghoian criticized that article and articles like it here that overly hype their subjects. He compared it to the unwarranted hype for Haystack. http://paranoia.dubfire.net/2012/07/tech-journalists-stop-hyping-unpro… Ryan Singel, editor of Wired's Threat Level, called Soghoian's article a sexist attack on Norton here: http://www.wired.com/threatlevel/2012/08/security-researchers/all/ Soghoian replied here: http://paranoia.dubfire.net/2012/08/responding-to-wireds-ad-hominem-ha… This is the news.ycombinator thread that discussed Soghoian and Wired's responses: https://news.ycombinator.com/item?id=4354959 This is Quinn Norton's, the original reporter's, response. I disagree with Norton (and Singel) and believe if Soghoian' original response had been sexist, Norton would have been able to quote from it. As it is, she can only make a tone argument.