Hacker News Reader

Comment by Adaptive | Hacker News Reader

In order of likelihood:

    * Defaced site, timed to screw up a big announcement
    * Rogue content maintainer
    * Phase II of audit turned up something rather bad
      (edit: NO - see tptacek below)

edit: Variations on "developer forced to do this" (cf simmerian's comment):

    * Developer was big brother all along and they are shutting it down
    * Security vuln about to be disclosed, dev scrambles to inform (albeit poorly)
    * Legally or otherwise compelled to compromise source code,
      dev complies and/or nukes project from orbit

The last alternative would be suggested in part by the strange content of the page, assuming it is legit from the developer: Normally I'd expect at least something like "there's a major vuln that is unfixable and we'll disclose formally in a week/two, migrate now.".

View on HN

After examining all the facts, I think it's most likely they just didn't want to develop it anymore:

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

And to top it off, let's put ourselves in the theoretical attacker's shoes, the binaries when run make no unexpected connection attempts or write to any unexpected places and don't appear to contain any unexpected imports, so if this was a hack, it's a very stealthy and very boring one. The most they achieved would be uninteresting to most attackers. It would only really be an effective attack against people who had TrueCrypt volumes but not a current copy of TrueCrypt as there's no compelling reason for anyone to upgrade to 7.2 and certainly they'd be skeptical after this. Any attacker with the intelligence and patience for such an attack would surely realize how poor an execution this would be. A better attack would be "here, it's TrueCrypt 8, it has loads of EFI support and mad security, everyone should install it, it's the best!". There's simply no reason to shut it down like this, unless the attack is just an elaborate practical joke.

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.

I guess the new license only applies to the release it is distributed with and this latest version removed encryption features so I doubt it will make the truecrypt project more compatible with FOSS licenses.

True, but they likely intended it for all releases and I highly doubt the dev(s) are going to burn their anonymity to go after you even if they didn't.

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

>>After examining all the facts, I think it's most likely they just didn't want to develop it anymore:

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.

Does National Security Letter sound depressing enough?

They haven't updated it for years.

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

I think suggesting BitLocker as a viable alternative to their work is "devaluing" their work. Who would trust BitLocker not to be vastly more compromised that TrueCrypt?

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

You're right - few would trust it more. However, it is the closest viable alternative to TrueCrypt on the windows platform today.

>> They haven't updated it for years.

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.

I think it makes sense, it can wear on a developer fairly heavily to be burdened by the user community of a cryptography product. Just look at this very thread, so many paranoid theories about NSLs and such, where they don't even make remote sense. They most likely just wanted to stop needing to work on the project or respond to comments.

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.

"Bitlocker is probably the most viable alternative on Windows."

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.

Easily. It performs the same task, is actively maintained and supports things like EFI. If you know of a better alternative on the windows platform, please do tell, because I'm not aware of one and neither are the TC devs it appears. Just because an alternative is not as perfect as you would like does not invalidate it completely. If you try actually reading, you'll note that I am the poster of many of the comments above. Please avoid making an ass out of yourself in the future. And keep your retarded paranoia out of legitimate discussion.

One of TrueCrypt's best features was hidden volumes. As far as I know, BitLocker doesn't haven anything like that. I can definitely see that someone would have an interest in getting that shut down.

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.

They don't offer alternatives. Please take a moment to read the second page - the instructions for "other operating systems." The linux instructions make it damned obvious that someone is playing a game. The question is who and why.

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.

They're offering an encryption product, not a service, an NSL would seem pointless unless they want to slap one at everyone who implements cryptography. and there's plenty of better ways for 3 letter agencies to obtain FDE keys. The Linux instructions just make it clear that your options and process vary per distro, if they're ditching the project I doubt they felt like writing piles more docs to describe every possible combination of loop-aes, dm-crypt, ecryptfs, encfs, luks, etc.

Could you post the PGP & Authenticode details? I was unable to verify the 7.1 releases.

I think it's most likely they just didn't want to develop it anymore...

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

I don't know what to think.

After the years of silence and the previously infrequent updates, I don't consider it farfetched at all.

I agree that it appears the developers are calling it quits. If you look at the diff itself, they have replaced large chunks of logic with an abort routine, passing the message "Insecure App".

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.

This is legit and I am willing to bet.

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 agree that this really looks like it is legit. It looks like, for some reason (we don't believe in the legend version, do we?) they abruptly (the diff contains many normal changes also) stopped the development and are burning all the bridges. They also slightly changed the license so now the forks are free to not mention that they are based on TrueCrypt, they are not allowed to link to truecrypt.org site or mention the TrueCrypt name in their product's domain name. They also removed all the links to their site from the source code (even the donation page).

> we don't believe in the legend version, do we?

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?

The element that does not square with any theories that suggest benevolent intent behind the change is the recommendation that users switch to Bitlocker. Surely, a Truecrypt developer who got served a gagging order to build in a backdoor would realise that a big and compliant target such as Microsoft would have been subject to the same measure long ago, and likewise that if a pre-existing vulnerability on a sufficient scale to justify this went unnoticed in an open-source project even after a proper audit, the situation is unlikely to look that much better in a closed-source solution (that only the makers and government agencies have access to).

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.

> The element that does not square with any theories that suggest benevolent intent behind the change is the recommendation that users switch to Bitlocker.

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?

I'd agree, that seems more believable than some of the other theories I've read.

That's my take on it as well, even though it fails the Occam's razor test. This all sounds like a very understated way of saying "we can no longer develop truecrypt with impunity, and the only other options are closed source, and highly likely to be compromised out of the gate".

This is currently my favorite theory, for lack of a better word. It's also potentially pretty disturbing news. I saw another user mention the term warrant canary. I'd call it a warrant canary for the cautious/well informed. Unless there is some serious action on the horizon, no one well informed will ever use a truecrypt release past 7.2 ever again. https://en.wikipedia.org/wiki/Warrant_canary

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.

Agreed on the language/vitriol interpretation. I was thinking the same when I listed rogue content maintainer.

As crazy as it sounds, I think you're right and it's just the developer(s) quitting (rage-quitting?) the project. Nothing else makes sense.

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.

No more than shutting down Lavabit violated whatever NSLs/court orders were directed at it.

> 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.

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.

It's pragmatic. BitLocker may or may not be flawed[1], but I don't see a lot of better options for disk encryption on Windows being presented.

[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.

Another possibility - the author was required by a court order to provide a backdoor for unfettered access to truecrypt disk, and to not disclose the existence of the order. The solution was to modify the code so that everyone has unfettered access (i.e. disable encryption entirely) and make the recommendation that everyone switch to something else.

Question: has any such court order, ever, in the history of the United States, actually been given? Can it be given? Because that would be news to me.

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).

Maybe while looking at the code themselves they found a very bad bug which would make previously made encrypted partitions easily crackable, and fixing it would obviously make the world aware to this, and they don't want to endanger or ruin the lives of everybody who has had a truecrypt container with sensitive data taken from them (for example to a malicious government), so the only way to go for them is to tell people their product should not be used any more and is bad.

That's rather radical move in the face of the audit currenlty being conducted. If the bug is so dramatic, it will be revealed by the audit with high probability rendering this move practically useless.

It could be that they've simply lost interest in developing it. It's quite the ongoing responsibility, and they may well be tired of working on it - a decade is a long time in anyone's life.

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.

If you're developing a free product and you're going to throw in the towel anyway, why not just open up the sources with a liberal license and/or hand the project over to someone else who's willing to carry the torch.

It may be the same strong sense of ownership and control that precluded the liberalisation of Truecrypt's license during its lifetime in the past decade.

Your reply seems to be the most sensible out of the lot. (Side note: I presumed there'd be a couple other ones that suggested it's open source--never mind that prior to the removal of some of the license text it wasn't "free as in beer" open. To be fair, I had forgotten myself that TrueCrypt wasn't exactly open source.)

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).

The one thing I have to add. Maybe by taking this route, the author(s) hope to inspire others to see the obvious need for a project like truecryp more dramatically. It's not like old bin/source repos aren't available in 100 different locations across the net; if you need the program you can get it. But this action does send a strong message and it's that there is no adequate existing cross-platform solution to this problem outside of trusting your os or using (not so plausibly deniable) file archives. I'd like to think that the author(s) are trying to highlight the necessity while secretly cheering on the next generation of trucrypt.

That's a good point, and it's certainly true on many fronts. Commercial alternatives are questionable at best (particularly given the NSA revelations), but the F/OSS community (and others) relied perhaps too heavily on popular software like TrueCrypt to fill the void. Now that it's gone and the other alternatives aren't quite as cross-platform as one might like, it does seem to illustrate a dearth of cryptographic software available for the general public.

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.)

It's already open-source.

As far as I know, its license is incompatible with other open source licenses due to an advertising clause (all derivative works have to state "based on Truecrypt" somewhere in the documentation or via use of the software). The old four clause BSD license had a similar issue.

One of the changes in the newly-uploaded version is indeed a change in the license.

It does appear that the authors have removed the advertising clause in this latest license version. Also the section on commercial licensing, some mentions of registered trademarks, and all specific references to the truecrypt.org domain, including email addresses.

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.

the devs themselves are anon

no one is going to come after anyone for licensing fees

The license was changed with the new release.

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.

Indeed it was.

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

Cryptsetup 1.6 supports Truecrypt volumes now, using its own reimplementation:

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

So at least Linux users should be covered.

The donations thing could be true, but Truecrypt simply wasn't a donation-friendly project.

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.

What if this is an attempt to smoke out the TrueCrypt devs?

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.

Wouldn't they just have to published a signed message stating that the change was not theirs and the key is compromised? Or better yet, revoke the key?

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

No, because the suggested "hackers" have published a signed message.

Doesn't matter. Publish another message signed with the same key saying "this key is compromised."

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

If 7.2 is part of the hoax, then they would be signing with a compromised key. This would be evidence, but would not be conclusive.

If the audit turned up something bad, the obvious step to take would be to publish it in all detail, fix the flaw, and then tell users to upgrade as soon as possible. Not go "OK SHOW'S OVER, USE PROPRIETY SOFTWARE FROM NOW ON".

Yes. This would be just about the worst way to announce and make recommendations. Looks like defacement to me.

https://twitter.com/matthew_d_green/status/47174183672207360...

It doesn't appear related to the audit

Nitpick: Truecrypt is proprietary (it's source is viewable, but you aren't licensed to distribute modifications of it).

Open(ish) source but non-free.

Unless some kind of backdoor was about to be discovered...and they'd rather close it down before it gets discovered.

My order of likelihood is #1. This is a canary. https://en.wikipedia.org/wiki/Warrant_canary

Do agencies like the NSA/FBI/etc have the power to make a company publicly lie against their will?

if you count lies of omission, yes. you can be required as part of an ongoing investigation or court case to not talk about it. The legal term is "gag order".

If you are ordered not to communicate something, you are ordered not to communicate it. Doing a "I'm not not telling you something, ha ha" might work on the playground, but not in real life.

I meant can the NSA force you to keep publishing "we have not worked with the NSA" when you have worked with them.

* That's a lot of wasted effort for a defacement with seemingly no motive except some (uncredited) lulz.

* 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.

It is a lot of effort, and you make good points. I'll add a note to my original post.

Phase 2 of the audit hasn't started yet.

Seems to point towards compromised SF account.

There's a new binary that recommends moving to BitLocker during install, and the signature matches.

Edit: with a new, compromised key.

Is source still available? Can we check the commit tree for anything suspicious lately? Can someone compile it and check the hash against the 7.2 binary being offered?

It's not that simple. It won't match anyway. Signatures, compiler versions, SDK versions, etc.

This guy managed to compile a previous version and have it match the released binaries https://madiba.encs.concordia.ca/~x_decarn/truecrypt-binarie...

Not quite. First, it was a ton of work to do so. Second, in the end he in fact only managed to match most of the released binary. Then, for most of the unmatched parts he came up with some solid arguments why they would be different (timestamps, pathnames and such). However, in the end there were still a couple of chunks of unmatched data that he couldn't explain why it has the value it does. Finally he argues that these chunks probably can't contain any hidden backdoors. This last bit I have a bit of a problem with because it's just speculation, all the rest of the research is very solid and without guesswork, and the conclusion of the article IMO incorrectly states he "has matched the binaries".

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.

Yes, but did they sign it using the same key they were using before?

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

Project on SF is still available if you have a direct link:

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...

Looks like it's the same key as before (F0D6B1E0)

On a day like this, compare the entire key.

It's the same. Original key: http://pgp.mit.edu/pks/lookup?op=get&search=0xE3BA73CAF0D6B1...

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

Their mail server is on the same IP as truecrypt.org, and it's now rejecting mail.

Also might be coming from lack of donations. I remember that button becoming more and more prominent lately...

That would not result in a message of "True Crypt Is Not Secure!!!!" in bold red. Seems to be geared towards frightening people.

I concur -- likely an elaborate website deface.

As irrational it may be, I've seen people writing something like that out of frustrations...

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...

Good point.

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.

That's not what the message says, though.

> 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.

Scroll all the way to the bottom

* SourceForge account compromised and developers unwisely stored private keys and other information somewhere inside this account, permitting the attacker to compromise lots of other stuff and generally have a blast.

NSA is obviously in on it. Who else would recommend using holy-bug-riddled proprietary-back-doored-on-purpose encryption software? ;-P

I choose to believe Niels Ferguson when he says "Over my dead body.": http://blogs.msdn.com/b/si_team/archive/2006/03/02/542590.as...

A NSL is powerful enough to render all this moot. Therein lies the great danger of NSLs.

2 things.

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.

If I were the NSA, I'd try to get one of my hundreds of world-class cryptographers a job on the BitLocker team.

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.

> Over my dead body.

> 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.

Reply on news.ycombinator.com