This is a staging forum for AgileBits, not an official support forum. Visit http://discussions.agilebits.com instead.
Flame & Collisions
jpgoldberg
Agile Customer Care 
This topic is for discussion of the recent posting on our blog about [url="http://blog.agilebits.com/2012/06/07/flames-and-collisions"]Flame and Collisions[/url].
Flag
0
Comments
-
Let me add a little bit more about Patty and Molly. Molly is young and strong and, well, not the brightest dog we've encountered. She is also a bit of a bully to Patty. Patty is old and small.
Some times when Molly and Patty are helping us pre-rinse the dishes after dinner, Molly may finish rinsing earlier and will push Patty away from her plate. Patty will then run out through the dog door into the back yard and bark at something imaginary. Molly will insist and participating in this defense of the back yard, and so she will run out back as well. Molly will continue barking and defending the back yard from imaginary foes while Patty slips back inside and finishes up with the dishes.
The only thing implausible about the Molly and Patty digital signature story is that Molly wouldn't know how to sign a document.
Cheers,
-jFlag 0 -
I knew that my post would be out of date the moment it was published, but I hadn't expected this:
[url="http://arstechnica.com/security/2012/06/flame-crypto-breakthrough/"]http://arstechnica.com/security/2012/06/flame-crypto-breakthrough/[/url]
The cryptanalytic technique used to create the MD5 collision is new. It isn't radically different than previous known techniques, but this is using a technique that would have taken a great deal of expertise to develop.
People have always speculated about how far ahead in cryptanalysis agencies like the NSA or GCHQ are compared to what is known by the academic community. The assumption is that the gap has been narrowing over the decades as there is more open work in cryptography done outside of intelligence agencies. We don't often get data to help pin anything down with our speculation, but this definitely is interesting.
The Ars Technica article linked to above has a terrific diagram outlining the nature of the general technique used to create MD5 collisions.
Cheers,
-j
Cheers,
-jFlag 0 -
[quote name='Fooligan' timestamp='1339117157' post='60154']
Dude, your blog post was so 9:30AM. Get with the times.
[/quote]
Thanks! I needed that.
It could be worse (and is). I just started looking at reviving something I started on "when (not) to change your password". The first paragraph of my draft was "February 1 has been declared in some corners of the Internet to be "Change Your Password Day". So what better time for me to address the question of how often to change passwords."
Cheers,
-jFlag 0 -
Joking aside, the security blog posts are great. Keep them going. I typically send them to my friends and family that I know don't stay on top of this stuff.
One friend thought it was no big deal about his LinkedIn account until I asked if he used the same password for his bank or other sensitive sites. That got his attention. Like you have said before: your password is only as secure as the weakest site that it is used on.Flag 0 -
Thanks!
There is an outstanding analysis how this MD5 collision was generated and used.
[url="http://blog.trailofbits.com/2012/06/11/analyzing-the-md5-collision-in-flame/"]http://blog.trailofbits.com/2012/06/11/analyzing-the-md5-collision-in-flame/[/url]
This is by some of the people who were involved in the 2007/2008 demonstration of how MD5 collisions could be used to create a bogus certificate.
They include lots of parts of the story that I left out, but they don't provide much in the way of explaining those to the general public. If you have specific questions about any of the pieces, I'll see if I can answer them.
They also have estimates of the computing resources needed to construct a suitable collision ($20,000 in 2008), and confirm that the creation of it involved "world class cryptography".
Cheers,
-jFlag 0
