Any updates about exiting the alpha phase?

What is necessary to do for the first release? When shall it be? Who helps to do what? Discuss here.

Any updates about exiting the alpha phase?

Postby rotorcowboy » Sat Dec 27, 2014 6:40 pm

Hey guys!

I appreciate you putting so much time and effort into continuing the TrueCrypt project with CipherShed.

I'm trying to compare your new product with the other TrueCrypt fork, VeraCrypt. A major downside of so much security is convenience, and it takes an annoyingly long time for VeraCrypt to mount drives. I understand how only 1,000 iterations can be considered somewhat weak, but I'm not sure it was necessary to increase that by a factor of over 600. :|

But anyway! Do you have any updates on when the first release will be? I haven't really found any words past September and I'm just itching to try CipherShed on for size! I downloaded your alpha test version, but it refuses to co-exist with TrueCrypt, so I can't really test it yet.

Keep up the good work!
rotorcowboy
 
Posts: 1
Joined: Sat Dec 27, 2014 6:30 pm

Re: Any updates about exiting the alpha phase?

Postby compul » Sun Dec 28, 2014 6:53 am

Hi rotorcowboy,

CipherShed is not supposed to coexist with TrueCrypt, but to be an upgrade to TrueCrypt. That said, do not use it for critical purposes yet. Testing so far should be done in a testing environment only.

VeraCrypt has it's own issues. I have never used it. However, I have heard from someone on our team that VC's code-quality did not have the standard he would wish for / expect in such a critical application. That's just second-hand info though.

An obvious issue is that VC is being developed by a single person, [and that it is not compatible with TC containers] EDIT: apparently, VC's newest release apparently does support TC volumes .

CipherShed will also address the key derivation issue, as well as others mentioned by the audit (and more). How long does it take in VC to mount a volume? Typically a good design choice would be to have it last not over 1/2 second. But since PBKDF2 is easily parallelized this may still be not optimal. It would probably be better to switch to a memory-hard hashing algorithm like scrypt, where 1/2 second would likely be sufficient.

I do hope that, now that many commits "only" need to be audited, the first full release of CipherShed 0.7.3 will follow in the not too distant future.

Cheers ;)
Last edited by compul on Tue Jan 06, 2015 1:52 am, edited 2 times in total.
Reason: People appear to have taken offense due to a not-so-diplomatic choice of words. Also, VC is now apparently compatible with TC.
User avatar
compul
Site Admin
 
Posts: 69
Joined: Fri Jun 06, 2014 6:15 pm


Return to Short-Term Goals / First Release

Who is online

Users browsing this forum: No registered users and 0 guests

cron