Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Fake Linus Torvalds' Key Found in the Wild, No More Short-IDs

lkml.org

Post reply on HN
Print view View original

31–40 of 131 posts

Re: Fake Linus Torvalds' Key Found in the Wild, No More Short-IDs

#31

Post by masklinn »

sleepychu wrote: post #26
lucaspiller wrote: post #21

Earlier quoted context omitted.

UTF-8 v9.0 contains 1085 emoji, that should be even easier to compare than random symbols.

I'm not sure if you're sincere but I don't think emoji would be easier to compare. +Might be less friendly to screen reader users?

> I'm not sure if you're sincere but I don't think emoji would be easier to compare.

The point is to leverage human pattern matching so you want short-ish figures with large differences.

Each hex digits is 4 bits, but each emoji is 10 bits, a 128 bits key is 13 emoji which is significantly more eyeballable than 32 hex digits, and chances are you'll notice EGGPLANT being replaced by CAMERA in 13 pictures easier than you'd notice B being replaced by 8 in 32 characters.

Re: Fake Linus Torvalds' Key Found in the Wild, No More Short-IDs

#32

Post by cesarb »

forgotpwtomain wrote: post #4

32 bit seems so obviously bad I'm not sure why we still have it..

We have 8-digit PGP short IDs for the same reason we have abbreviated 7-digit hashes for git commits (`git rev-parse --short`): it's short enough to keep in one's head (see https://en.wikipedia.org/wiki/Seven_plus_or_minus_two).

Re: Fake Linus Torvalds' Key Found in the Wild, No More Short-IDs

#33

Post by AdamN »

avisser wrote: post #27
vesinisa wrote: post #22

Earlier quoted context omitted.

Pretty sure most of the time people only read the first few and last words and e.g. don't concern if the words in the middle are in correct order. Not sure about the size of that dictionary, but it would seem to greatly diminish the entropy.

It's asking for a solution with more gestalt impact. Turn the key into a painting or some sort of visualization.

Ah, you mean randomart:

http://superuser.com/questions/22535/what-is-randomart-produ...

Re: Fake Linus Torvalds' Key Found in the Wild, No More Short-IDs

#34

Post by JulianMorrison »

lucaspiller wrote: post #21
JulianMorrison wrote: post #12

Take a leaf out of urbit's book, and convert hex strings into readable nonsense syllables that are a lot easier for humans to compare.

UTF-8 v9.0 contains 1085 emoji, that should be even easier to compare than random symbols.

I think emoji would be hard to compare. A lot of very similar little faces. Is that a wink or a blink or a frown?

An example of urbit's rendering of a 128-bit number into textual form is "racmus-mollen-fallyt-linpex--watres-sibbur-modlux-rinmex". While it might be gibberish, it's gibberish that even a screen-reader program could take a swing at, and humans can easily read.

Re: Fake Linus Torvalds' Key Found in the Wild, No More Short-IDs

#35

Post by d33 »

ajdlinux wrote: post #5

All the fake keys that I've seen mentioned are from the data set at https://evil32.com . It appears a couple of researchers decided, back in 2014, to demonstrate this issue by cloning the entire strong set of the PGP web of trust (not just Linus' key, but basically everyone who uses PGP/GPG for Free Software development - myself included). It would appear that sometime quite recently, someone decided it would be fun…

At this point this is INSANE that GnuPG still defaults to short IDs...

Re: Fake Linus Torvalds' Key Found in the Wild, No More Short-IDs

#36

Post by thomasahle »

avisser wrote: post #27
vesinisa wrote: post #22

Earlier quoted context omitted.

Pretty sure most of the time people only read the first few and last words and e.g. don't concern if the words in the middle are in correct order. Not sure about the size of that dictionary, but it would seem to greatly diminish the entropy.

It's asking for a solution with more gestalt impact. Turn the key into a painting or some sort of visualization.

It has to be a visualization in which changing a few pixels make it look significantly different. Otherwise we can still make a 'low distance' brute force attack.

Re: Fake Linus Torvalds' Key Found in the Wild, No More Short-IDs

#37

Post by JulianMorrison »

avisser wrote: post #30
sleepychu wrote: post #26

Earlier quoted context omitted.

I'm not sure if you're sincere but I don't think emoji would be easier to compare. +Might be less friendly to screen reader users?

I hate to say it, but comparing two, long hash values via a screen reader doesn't seem viable for humans, regardless of emoji. Maybe an auralizer to turn the hash into a short piece of music?

See my above example from urbit. They are something a screen reader would clearly read differently for different values, if not comprehensibly.

Re: Fake Linus Torvalds' Key Found in the Wild, No More Short-IDs

#38

Post by cooper12 »

_wmd wrote: post #10
wtbob wrote: post #6

I don't know why more folks don't display keys and fingerprints as Base64; it seems to me that "q68RxlopcLEwq+PEeb4+QwBBGIY=" (Linus's real key) and "D2oUZTLYaa7kOPdLYhGqOwBBGIY=" (Linus's fake key) are pretty easilt-distinguishable, and not terribly verbose.

You can do what PGPfone did and encode the fingerprint as a series of dictionary words: Real gregkh: 647F28654894E3BD457199BE38DBBDC8 = style tactful newcomers file gallows adored insist flags athletics Fake gregkh: 497C48CE16B926E93F49630127365DEA = jukebox governor fashionable mahogany prepares gobble surprised martha apostles There's even an Internet standard for this: http://tools.ietf.org/html/rfc1751 , although…

Regarding the SSH randomart, it's enabled by adding `VisualHostKey=yes` to your ssh config or adding the flag `-o VisualHostKey=yes` on the command line. It was announced with OpenSSH 5.1 (http://lists.mindrot.org/pipermail/openssh-unix-dev/2008-Jul...) and there's also a paper on it (http://www.dirk-loss.de/sshvis/drunken_bishop.pdf). You can find the code/comments in `key.c` under `key_fingerprint_randomart()`. (http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin/ssh/key...)

Re: Fake Linus Torvalds' Key Found in the Wild, No More Short-IDs

#39

Post by mtgx »

wtbob wrote: post #6

I don't know why more folks don't display keys and fingerprints as Base64; it seems to me that "q68RxlopcLEwq+PEeb4+QwBBGIY=" (Linus's real key) and "D2oUZTLYaa7kOPdLYhGqOwBBGIY=" (Linus's fake key) are pretty easilt-distinguishable, and not terribly verbose.

I believe miniLock also does that. Any reason why miniLock can't be used as an alternative to PGP? (for email/file encryption)

https://minilock.io/

Re: Fake Linus Torvalds' Key Found in the Wild, No More Short-IDs

#40

Post by VMG »

cesarb wrote: post #32
forgotpwtomain wrote: post #4

32 bit seems so obviously bad I'm not sure why we still have it..

We have 8-digit PGP short IDs for the same reason we have abbreviated 7-digit hashes for git commits (`git rev-parse --short`): it's short enough to keep in one's head (see https://en.wikipedia.org/wiki/Seven_plus_or_minus_two ).

But you usually don't have to worry about adversarial input when you check out a local git branch.
Post reply on HN