Um guys, A5/3 is completely broken. According to Wikipedia: "In 2010, Dunkelman, Keller and Shamir published a new attack that allows an adversary to recover a full A5/3 key by related-key attack.[5] The time and space complexities of the attack are low enough that the authors carried out the attack in two hours on an Intel Core 2 Duo desktop computer even using the unoptimized reference KASUMI implementation. The authors note that this attack may not be applicable to the way A5/3 is used in 3G systems; their main purpose was to discredit 3GPP's assurances that their changes to MISTY wouldn't significantly impact the security of the algorithm."

Even if A5/3 weren't broken, there are still tower dumps and IMSI catchers, which are a whole lot easier to use than breaking encryption. Yes A5/3 is better than A5/1, but I call bullshit on this whole article.

>Yes A5/3 is better than A5/1, but I call bullshit on this whole article.

Super pendantic, but the title is 'hardens' not 'makes hard'. If it's better, than it's been hardened. Might not be the best thing available, but that's the meaning of a comparative.

Super-duper pedantic, but I would say broken can be considered neither hard nor hardened.

> The authors note that this attack may not be applicable to the way A5/3 is used in 3G systems.

This is a pretty big conditional. But I still imagine intel agencies have broken KASUMI.

Yea, it is a related-key attack.

Indeed, I would not personally consider using KASUMI to be a positive change worthy of the title of "hardening".

The undo functionality is actually almost there, I think we just don't have an interface for it. The arrow keys thing should be easy to implement. I will add it to github, and of course accept a pull request! We definitely need to accept ranges for formulas or some up with some easy way to do that.

The project as a whole is definitely unfinished but we have been working on it for so long and lost some motivation. I wanted to just get it out there to at least re-motivate myself to work on it.

Please note, I don't think all the features I listed are desirable. (Pie menus for dragging gestures? Pretty hokey.)

Ethersheet allows typing ranges in formulas and that is possibly enough. Dragging to choose a range can be nice but it has to be implemented carefully. When to stop acquiring cells for the range and return focus to the formula editor? Some versions of Excel do this badly. For example, after dragging to enter a range you use arrow keys to balance parens. Excel thinks you are still selecting cells and inserts multiple ranges into your formula.

I swear this person was not coached in any way ... :)

The same way as you would in excel `=SUM(A1:A11)`

Interesting, it looks like it's based of ethercalc which is also a great project. We were trying to go in a different direction though and be a little bit more user friendly and also work on mobile.

I specifically addressed this in the article. The feds located freedom hosting by using an exploit in Firefox which was able to deanonymize users. I don't know enough about the silk road case, but it seems probable that traffic correlation was used in that case. I agree that things can change in a year, but the essential point that Tor is not cryptographically broken is still true, IMO.

>The feds located freedom hosting by using an exploit in Firefox

That doesn't even make sense. Firefox is client software. How would I locate a hidden service, a server, via an exploit in Firefox? Servers don't generate websites with Firefox.

Tor is not cryptographically broken, I agree. But see my post above about the number of nodes-- it is trivial for any nation-state to spend a small bit of money to completely compromise Tor.

If you enjoyed this article, there is a fantastic book explaining all of the problems with the American prison system and the 'Felon' Label that goes along with it. It's called "The New Jim Crow" by Michelle Alexander. Highly recommended reading.