* PGP matches
    * Authenticode matches
    * SourceForge data was modified
    * DNS records were modified

It's quite possible this came from 1 big developer hack, but considering how the release was done, with full source and everything for every supported platform... if it was a hack, it's a very, very good one. They've also decided to modify the license terms, perhaps bringing it into compatibility with more common FOSS licenses.

I think it's far more likely at this point that the devs, who had not updated their software in years, finally decided to call the project over and have marked it insecure because the codebase is now unmaintained and should be assumed insecure.

This does not explicitly suggest vulnerabilities exist in older versions, but rather the latest version with these changes is very explicitly insecure and should not be used. This does leave room for issues to have been discovered in old versions - maybe rather than fix, they are throwing in the towel.

It would be so easy for the person(s) in question to just say that, though.

I don't know what to think.

So they decided to end things with such an extremely juvenile behavior devaluating the years they have invested in this project even if not recently?

Unless the responsible one fell into clinical depression it's a pretty strange reason.

I'd hardly call the behavior "juvenile" nor would i call it "devaluating". They've simply abandoned it and are offering alternatives.

(Devaluating? sic? Or is that actually a word?)

If it's the actual TC crew, the explanations toward an NSL or a new vuln that an NSL or similar applies to seem to be about the only rationale - though even then, it seems to me it'd be possible to steer the code audit in the right direction.

Given the churn of new keys today, I'm more inclined to think that the comms of TC have been broken and the breach is being used to drive people away from the releases of the tool for which source is available.

As I said even if not recently they still invested many years into that project.

Of course it is juvenile to senselessly ruin the code and suddenly advertising a very different commercial product, especially without proper scientific reason.

Not to mention that precisely because they haven't done much work lately I don't see any reason why the developers would disfigure their project like that.

They're not "advertising a very different commercial product". They're recommending you actually use a maintained alternative. Bitlocker is probably the most viable alternative on Windows.

It's also possible that Bitlocker solved the problem well enough for them so they saw no gain in maintain TC any further.

How would a software with closed source from a company that has been gladly working with the US government in the past be a "viable alternative" to TrueCrypt? Really, please try to read the comments above before you enter the discussion.

BitLocker lacks another one of TrueCrypt's most important features: open-source code, readable, verifiable, and compilable by any interested user. BitLocker is almost definitely already backdoored, so encouraging people to switch to that makes all of that data accessible by the powers that be.

I think it's silly to pretend like no authorities would have an interest in promoting the use of closed-source encryption techs. Apple and Microsoft were both willful participants in the PRISM program. TrueCrypt was the only open-source FDE software that had widespread adoption on non-Linux systems. After this, no major corporation or group is going to use TrueCrypt to encrypt anything anymore and will rely entirely on the backdoored encryption solutions provided by the NSA's known and confirmed compatriots.

Though I suppose that's not the best legal rationale, now is it?

https://gist.github.com/anonymous/e5791d5703325b9cf6d1

The entire source has been modified to reflect the Sourceforge page its contents. Encryption process is disabled. The current binaries can only be used to "migrate". You can deface a webpage but the effort it takes to rewrite the entire source code, compromise the GPG, compromise domain, compromise mail servers et cetera is not minimal.

It is happening. Whether they are being forced to do so is a whole different story.

I dunno. We saw the legend version happening with a lot of people recently. Yep, we still mostly don't belive it, but...

What's the oposite to The Boy that Cried Wolf?

While this might be reading a tad much into it, the language of the announcement, specifically the "...as it may contain unfixed security issues" bit, sounds like what somebody who just came out on the losing end of a heated debate about whether some bug-feature should or should not be considered the former would say. Knowing the vitriol and determination with which software developer arguments are often carried out, this would explain the observed combination of remarkable dedication and haphazard execution.

I realize we are firmly in conspiracy theory territory here, but perhaps the suggestion that users switch to Bitlocker is intended to be so patently absurd as to be a signal that the developers are under duress?

My very first thought upon hearing this news was that just maybe the trucrypt devs are secretly cheering for others to come and replace the hole they fill in this world. Maybe this is the best possible thing they could do to inspire the next generation of truecrypt. They've laid a good foundation but they know they can't win the war for free information security on their own. So what better good could they do than to inspire the mother of invention; necessity.

The Bitlocker thing seems strange until you realize that it probably really IS the best alternative for most users. The users who are paranoid enough to not trust Bitlocker can probably look out for their own security, so it makes sense to give instructions for the rest.

None of the explanations people have came up with regarding intelligence agencies and conspiracies really hold water. If the devs were compromised, I doubt they would be given the freedom to post something like they did today. If they were under some kind of NSL or gag order, this would almost certainly violate it.

It seems odd to me that security specialists would recommend a flawed tool to 'the masses'. It just doesn't fit with what I usually read from people in this line of work.

[1] - Whether it's flawed or not depends on your needs. Not everyone cares about defending their data against an attacker with the resources of a nation-state - many just want their laptop's hard drives to be effectively inaccessible if the device is stolen or lost.

Because while specific orders can't be disclosed if they'd give away information to the actual target, the general nature of such orders is well known - information can be demanded if held. You can't be compelled to engage in subterfuge though, and since such an order would be illegal, you could freely disclose it and let the civilian courts strike it down.

As one might note from the Lavabit fiasco, things only got weird because Lavabit decided to screw around being non-compliant, while also always having the technical capacity to decrypt everyone's email (and thus opening up the legal doorway to just seize the keys and all the data, rather then the tiny chunk that was wanted).

If this is true, then perhaps such listlessness was also catalysed by the ongoing audit. Maybe seeing such a mass of crowdfunding income towards a project to pick Truecrypt apart, in contrast to the scant donations to its development, disheartened the authors towards further work?

Abandoning it in this rather dramatic way ensures that Truecrypt's users are warned against using unsupported software where any bugs will remain unfixed. This is especially important when such bugs revealed in the future (and maybe ones already known) have the possibility of being deleterious for security.

https://code.google.com/p/cryptsetup/wiki/Cryptsetup160

So at least Linux users should be covered.

Edit: The following clause was deleted in the 7.2 release.

- c. Phrase "Based on TrueCrypt, freely available at - http://www.truecrypt.org/" must be displayed by Your Product - (if technically feasible) and contained in its - documentation. Alternatively, if This Product or its portion - You included in Your Product constitutes only a minor - portion of Your Product, phrase "Portions of this product - are based in part on TrueCrypt, freely available at - http://www.truecrypt.org/" may be displayed instead. In each - of the cases mentioned above in this paragraph, - "http://www.truecrypt.org/" must be a hyperlink (if - technically feasible) pointing to http://www.truecrypt.org/ - and You may freely choose the location within the user - interface (if there is any) of Your Product (e.g., an - "About" window, etc.) and the way in which Your Product will - display the respective phrase.

As was its ability to encrypt, apparently. I don't exactly count those as one in the same.

Perhaps ownership does run deep, even if you've never really released a product for money and it's always been free. Still, it's something I don't understand: If I had a tool that I released for free and finally gave up on it, I'd like to think I'd open it up under an exceptionally liberal license or just dump it in the public domain. That'd be especially true if it had a lot of users.

That's what makes me think that if this isn't some elaborate scheme it's likely the result of some sort of legal requirement or action (e.g. Lavabit) which would preclude the author(s) from doing anything else with the software. It's a shame they couldn't take a scorched earth-esque approach of dumping everything in the public domain, including notes on why this was happening, everyone else be damned, but I'd imagine their entire career might be in jeopardy at that point (and possibly their freedom).

Given that border searches of electronic hardware are becoming more commonplace in the US, I should think that something like this is important. I know when I was driving back and forth to university, the thought crossed my mind when I had my laptop with me as I went through the border patrol checkpoint that there wasn't anything much I could do (outside lawyering up) if they took it upon themselves to grab it and search. Sure, the worst they could have done was read my email (and maybe clear out the junk folder for me while they're at it), but it was the principle of feeling so violated by the act itself that drove me to stuff all my school work into a TrueCrypt volume.

(This was years ago, and TC was the best option for XP. Though I later switched the laptop over to Linux.)

However it's unclear if this change is only for Truecrypt 7.2 or can be applied retrospectively to previous versions. As the authors have deleted sizeable portions of the encryption code in this final version, such ambiguity could be problematic.

no one is going to come after anyone for licensing fees

It was developed in almost total secrecy, with binaries and code being tossed over the wall once in a while, provided under a non-OSI license.

I built it from source a few times, but Truecrypt was used by a lot of people, far beyond the developer community, people who totally lacked the technical skill needed to compile it. SO it's a safe bet the majority just used those binaries. And those binaries actually changed without warning more than once, independent of the version changing. That alone bred suspicion and fueled the demand for an audit.

The behavior of the developers (and their total lack of a public presence, regardless of the reason) may have also discouraged donations. Same with the behavior of the forum moderators, whoever they were.

While this move seems odd, the new binaries are properly signed and the domains have been updated accordingly. If this was another project, like Rails, the maintainer could come out and say they were hacked and the last good version was X. Otherwise, the project would likely die off.

But since we know so little about the TrueCrypt maintainers, there's little way for us to hear that this isn't legitimate. In order to keep the project from dying (if this is a hoax), they would have to prove that they are the maintainers, because any plausible deniability would undermine their claim that the change was not legitimate.

If two groups with opposing messages control the key, it's pretty clear that the key is compromised in some manner.

It doesn't appear related to the audit

* Possible, but once again I see no motive that would produce this brand of outburst.

* And it's unfixable? That would be a world first.

I think it's much more plausible this is some powerful entity forcing a hand. We know by now there's plenty of motive and candidates to fit that shoe.

Edit: with a new, compromised key.

http://sourceforge.net/projects/truecrypt/files/TrueCrypt/

http://sourceforge.net/projects/truecrypt/?source=navbar

http://sourceforge.net/p/truecrypt/activity/?page=0&limit=10...

Odd, 6 hours ago someone updated the TruCrypt-key.asc files, then 3 hours later posted all the new binaries.

Also odd is whoever posted the new binaries completely yanked all the previous ones, leaving only the new and questionable binary available for download.

hmm...

And the new key can be found on the SF site.

I'm chalking this up to, after all that work, having come so far, kind of wanting to say you did it, and providing a few (IMO) hand-wavy arguments why those last impossible chunks of bits probably don't matter, instead of having to admit defeat after being able to account for 99.9% of all bytes.

I can understand that, I've been wanting to type "but it's probably not backdoored" three times already typing this post, and catching myself because really I have nothing to go on to make or back that statement.

edit: Apparently not, according to the link @Alupis posted.

I concur -- likely an elaborate website deface.

> WARNING: Using TrueCrypt is not secure as it may contain unfixed security issues

That is a perfectly reasonable thing to say if you are abandoning security software. Any issues discovered will not be fixed, so you should stop relying on this software for security.

I don't think any legit organization would do that, but what if it's maintained by a small team or even individual -- I don't think I've ever seen a single face of TrueCrypt developers out there...

If true, I'd much rather have them post a countdown clock and say "If we don't reach X funding goal in donations by date Y, then we will be forced to close the project". Funds would come in then... a lot of people depend on truecrypt.

In fact, BitLocker would be the first thing I'd weaken. Because

1) It's closed source, hard to externally audit.

2) It's one of the most used encryption packages in the world.

3) Microsoft's poor security track record provides excellent cover if the weakness is ever found.

Closed-source security software is a recipe for disaster.

1) Obviously I was joking in my comment above.

2) That post is dated well before common knowledge of how "in-bed" Microsoft is with the USA spy agencies (at least management is).

This is a topic for another discussion, but I just want to point out that, of course someone being coerced under gag order to install less-than above-water "features" and/or purposefully weaken the product would say exactly what is in the blog post.

> Well, maybe not literally---I’m not ready to be a martyr quite yet

So, in short he's written 8 years ago that in principle he was totally against backdoors. Times have changed so much that a statement of that tone is almost entirely useless IMHO.