Hacker News Reader

Comment by Shank | Hacker News Reader

Seems to me like whichever "state party" this was didn't consider the implications behind their actions. Wikileaks has released insurance files in the past, and I wouldn't be surprised if this doesn't, automatically or not, trigger a dead man's switch for more leaks.

If you're a state party trying to prevent leaks, silencing Julian is probably the worst way to go about it. It would be arcane not to think of this scenario and plan for it.

View on HN

So long as we're all speculating, perhaps the state actor who leaked the documents to Wikileaks in the first place didn't agree with Assange's schedule and wants to trigger the dead man's switch.

Well, I'm off to read some classic Tom Clancy.

That sounds sensible, they say it's over 500GB of Documents. The release of such a flood of files would swamp all "public intelligence capacity" until after the election.

If you looked at how long it took to find the gold nuggets in the Podesta mails it sounds reasonable willing to force their hand.

I think that any intelligence community that couldn't digest 500GB in full text / keyword searching (keywords that are likely already configured/known) between now and the election would be a pretty sad intelligence community.

Apropos of the keyword list, dissecting the corpus could be done in a few hours with a promo-code worth of AWS time.

I think "Public Intelligence Capacity" was referring to the ability for use regular members of society to find and verify meaningful information from the leaks.

Yes, i thought more of "people on 4chan" than actually any Intelligence Service of any nation state. I'm pretty sure they are mining it, but there would be political fallout from publishing their findings.

> It would be arcane not to think of this scenario and plan for it.

And following on from this line of thought, it is therefore not unreasonable to conclude that they did think of this scenario, but feel that the risk of calling Assange's bluff (see the speculation that the insurance files are just random data meant to seed misinformation) is less than the risk of whatever they think he knows about them being leaked.

Which makes you wonder, what are they afraid of to consider such action?

Don't underestimate the ability of intelligence agencies to lack it ...

Or there's no insurance files and Assange is bluffing. He won't sit on some scoop harming the United States for long, his ego and power trip would force his hand.

The insurance files are public. They just have to release the keys.

3.6GB - http://wlstorage.net/torrent/wlinsurance-20130815-A.aes256.t...

49GB - http://wlstorage.net/torrent/wlinsurance-20130815-B.aes256.t...

349GB - http://wlstorage.net/torrent/wlinsurance-20130815-C.aes256.t...

94GB - https://file.wikileaks.org/torrent/2016-06-03_insurance.aes2...

I don't know why they would put out a few hundred gigabytes of fake files. The contents might not be as interesting as people hope, though.

"Bluffing." It takes zero effort to distribute a a few hundred gigabytes of encrypted random data, relative to the amount of investigation and speculation as to what if anything is in those files. They are called "insurance." Assange/Wikileaks wanted to deter actions being taken against him.

Seems to me as likely as not that there is nothing really there.

the good news is that you don't even need to encrypt your data, if they are random to start with.

That might be tricky. An encrypted file may have some properties a random file does not have in general.

No, unless the algorithm is broken.

Ciphertext indistiguishability[1] is a basic requirement for any encryption scheme. All modern ciphers are designed to make that hard.

[1] https://en.wikipedia.org/wiki/Distinguishing_attack

Yes, but you remove that attack vector simply by encrypting a random file instead.

There was a link on HN not too long ago to a quote from Assange, stating that part of the strategy is to sow uncertainty and doubt among the players to increase the secrecy tax. A multigig encrypted dump could have that effect.

That said he can't really play that card too often.

>I don't know why they would put out a few hundred gigabytes of fake files.

Looks like you can check it now. The keys went public:

pre-commitment 3: UK FCO f33a6de5c627e3270ed3e02f62cd0c857467a780cf6123d2172d80d02a072f74

pre-commitment 2: Ecuador eae5c9b064ed649ba468f0800abf8b56ae5cfe355b93b1ce90a1b92a48a9ab72

pre-commitment 1: John Kerry 4bb96075acadc3d80b5ac872874c3037a386f4f595fe99e687439aabd0219809

https://twitter.com/wikileaks

Those tweets came out 7 hours before the Internet link message.

I speculate that these are only SHA256 hashes of keys or documents that will be released in the future. It's pre-commitment. I didn't try decrypting one of the files with a key yet, though. I'm busy with other things (like browsing HN, aaah).

The tweets are just hashes of the insurance files to confirm their validity, not the decryption keys (yet).

> I don't know why they would put out a few hundred gigabytes of fake files.

To trick people like you? `dd if=/dev/urandom` is much easier than finding a comparable quantity of actual info.

I don't know enough to have much opinion on the odds that's the case, vs them being real, though.

> I don't know why they would put out a few hundred gigabytes of fake files.

The same reason a stick-up man might put their hand in their pocket to mimic the shape of a gun. If you have no weapon, it's still often useful to make people think you have one.

Seems overblown. Do they really have enough data to dead man switch? I'm unconvinced. Sure they may have something but if it was that good they would have released it by now. Even if whatever they have (if anything) is actually good I would imagine risking it leaking is much better than leaving Julian in position to constantly leak things for a lifetime.

Reply on news.ycombinator.com