Blog Home  Sign In RSS 2.0 Atom 1.0 CDF  

  def Softwaremaker() :
         return "William Tay", "<Challenging Conventions />"

  knownType_Serialize, about = Softwaremaker()

 Saturday, May 31, 2008
« Too many books - too little time and spa... | Main | Hi-Def AVCHD Video Format Processing for mere mortals ... »

I was recently pointed to this post that highlights a "successful attempt" by some students in Germany to crack Microsoft Cardspace.After reading through the post several times, I became convinced that it is NOT what it seems it is and that if the "breach" is what it says it is, there must be some pre-conditions that must be satisfied before it can happen and these criteria are not going to be easy...

Just as I was putting some of my thoughts down that relates to why I think the attempt is somehow "inappropriately glorified":

  1. If an end-user would be stupid enough to put and store his/her passwords, credit card information on his PC
  2. There must be some sort of DNS compromise on the end-user side, which also means successfully hacking into his/her router
  3. There must be some sort of Digital Certificate Store compromise on the end-user side, which also means successfully hacking into his machine with highly-elevated priviledges or saying, the user's machine password has been stolen

Points [2] and [3] relates to the statements from the attempt and I quote from the above post:

openquotes.png To reproduce the demonstration, you should change your own DNS settings and install an untrusted certificate closequotes.png

If I can do both those points sucessfully, to be honest, I already have control over what the user does on his machine, stealing his Infocard is probably of low priority at that point in time.

Then, the brains behind Cardspace, Kim Cameron, himself, wrote a comprehensive reply, which basically was a detailed answer to my brief thoughts above, to counter the students' attempt and should really put any doubts in anyone's mind to rest.

[Added 02 June 2008]: In this video on his blog, Kim demonstrates how YOU, the end-user, must FIRST POISON your own machine first before the attack can happen:

Some comments standout and I quote:

openquotes.png The demonstrator shows that if you are willing to compromise enough parts of your system using elevated access, you can render your system attackable. This aspect of the students’ attack is not noteworthy.


openquotes.png There is, however, one interesting aspect to their attack.  It doesn’t concern CardSpace, but rather the way intermittent web site behavior can be combined with DNS to confuse the browser.  The student’s paper proposes implementing a stronger “Same Origin Policy” to deal with this (and other) possible attacks.  I wish they had concentrated on this positive contribution rather than making claims that require suspension of disbelief. closequotes.png

openquotes.png However, the students propose equipping browsers with end user certificates so the browsers would be authenticated, rather than the sites they are visiting.  This represents a significant privacy problem in that a single tracking key would be used at all the sites the user visits.  It also doesn’t solve the problem of knowning whether I am at a “good” site or not.  The problem here is that if duped, I might provide an illegitimate site with information which seriously damages me.


While I know the ignorant media will find some ways to sensationalize this unworthy episode, especially when Microsoft is such a big target, this brings to mind a popular joke which I think can be used as an anology:

Q: How do you make 1 million dollars ?
A: Start with 2.

Saturday, May 31, 2008 8:45:51 AM (Malay Peninsula Standard Time, UTC+08:00)  #    Disclaimer 
  • Blog reactions