Key material

[FiloSottile]

I think that how much bytes are taken out of the key strengthening function and how they are separated should be better explained.

(There is also a typo "seperate")

[maxtaco]

thanks, the typo is fixed.

i will tighten this when we upgrade the docs to reflect v3

[FiloSottile]

Yeah, I had a bit of problems figuring out which cipher key is taken out first of the Scrypt stream

[maxtaco]

You are right, this is a bug. It comes from this line:

https://github.com/keybase/triplesec/blob/master/src/enc.iced#L81-L85

but that's relying on the fact that keys come out of a hash the same order they go in, and it's best not to rely on that. Thanks for pointing that out, it should be fixed.

[FiloSottile]

Good, it is also a bit confusing that the order is the opposite of the one in which ciphers are listed in https://keybase.io/triplesec/

[maxtaco]

Ok, I will fix it in header_v3. Which order makes the most sense to you?

On Sun, Nov 3, 2013 at 2:53 PM, Filippo Valsorda
[email protected]:

Good, it is also a bit confusing that the order is the opposite of the one
in which ciphers are listed in https://keybase.io/triplesec/

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/22#issuecomment-27652499
.

[maxtaco]

Actually, I'll keep it as is. but it's hmac, aes, twofish, salsa....

On Sun, Nov 3, 2013 at 3:05 PM, Maxwell Krohn [email protected] wrote:

Ok, I will fix it in header_v3. Which order makes the most sense to you?

On Sun, Nov 3, 2013 at 2:53 PM, Filippo Valsorda <[email protected]

wrote:

Good, it is also a bit confusing that the order is the opposite of the
one in which ciphers are listed in https://keybase.io/triplesec/

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/22#issuecomment-27652499
.

[FiloSottile]

Yeah, I don't have any preference for an order over the other, but it needs
to be explicitly documented.

Probably is better to keep it like it is for backwards compatibility of the
code.

On Sunday, November 3, 2013, Maxwell Krohn wrote:

Actually, I'll keep it as is. but it's hmac, aes, twofish, salsa....

On Sun, Nov 3, 2013 at 3:05 PM, Maxwell Krohn <[email protected]<javascript:_e({}, 'cvml', '[email protected]');>>
wrote:

Ok, I will fix it in header_v3. Which order makes the most sense to you?

On Sun, Nov 3, 2013 at 2:53 PM, Filippo Valsorda <
[email protected] <javascript:_e({}, 'cvml',
'[email protected]');>

wrote:

Good, it is also a bit confusing that the order is the opposite of the
one in which ciphers are listed in https://keybase.io/triplesec/

—
Reply to this email directly or view it on GitHub<
https://github.com/keybase/triplesec/issues/22#issuecomment-27652499>
.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/22#issuecomment-27653004
.

Filippo Valsorda

[SparkDustJoe]

This is a stale issue as this has been better explained on the website, and now we're in Version 4. This can probably be closed?