Opensource maskerkeying project proposal - restarted

[32768]

According to Matt's paper, nearly every master keyed system in use is designed according to either the TPP or the RC schemes.

And with either of these schemes, you can guarantee to read the top master bitting for the entire system using just one lock, the correct change key for that lock, and some key blanks.

Thanks for the clarification! No matter how the system is done, you'll always get a key with at least the lowest level of master privileges. The basic problem Matt points out is that you can interrogate one pin of the lock at a time, reducing the number of keys you have to try to a manageable number.

I understand now (thanks PickPick!) that TPP or RC uses a combination of change and top level master bittings for sub-masters. This means that you would only need three pins (two cuts) per stack. Is there a system where you could have completely different bittings for master, sub-master, and change keys? This could be done by having more than three pins in each stack. Or would this just lead to having too many very thin pins that would jam all the time?

With TPP or RC, it seems that all change keys in the same sub-master group would need to have some cuts in common, namely the cuts they share with the sub-master key. I've made a small example for a system with two four pin locks with change keys, a sub master and a top level master. The notation I'm using is that a pin stack with cuts at 2 and 3 is denoted as "[2, 3]".

Core A:
[1, 4] [2, 5] [3, 6] [4, 7]

Core B:
[1, 4] [2, 5] [5, 6] [6, 7]

Change Key A:
1 2 3 4

Change Key B:
1 2 5 6

Top Level Master:
4 5 6 7

Sub-Master
1 2 6 7

Does this accurately depict how this system would work?

I hope this isn't sidetracking too much from the programming project, but I'm guessing (hoping) that I'm not the only one with some level of confusion.

[Jarod]

i think i understand that, and if i understand right then it should work, but don't take my word for it, btw let me know when the site is up on sourceforge, i tryed searching for omk but sourceforge is acting funny, because i also searched for torcs (i KNOW that one still exists) and got nothing, so anyhow tell me when its up

[WhiteHat]

yep - all volunteers, sign up on sourceforge. I think if you search for whitehatnetizen, then you can watch my diary or something similar and it will send out an e-mail when the project gets approved I'll also post here I guess.

32768 is hereby appointed requirements custodian...

varjeal (or any one else really) do you have the specifics of those masterkeying standards handy?

[Hojo]

Ok guys I finished my master key decoder, there is some problems with it I know, I will eventually get it better, so far it can only handle 1 master pin and has no limit on the numbers ie, you could say you have a 9 bottom pin and a 9 master on top, that is WAY to big and is imposible to have in any lockwood, It is made for the members here and hope you like it...just need a place to host it, it will be about 50k at the most

Its called MasterKD (Master Key Decoder)
V 1.0.0

Anyone who wants it email me and Ill send it untill I find a server...

[WhiteHat]

sourceforge unix name OMK rejected.

changed unix name to OSMK and re-applied...

[Hojo]

MasterKD Download

Enjoy

[WhiteHat]

yes - testing is a Good Thing

[Jarod]

ok, i'm registered as "jarodthelinux" (jarod was taken and jarodthelinuxguy was too long) so i'll do anything in my power to help,
i've got an
athlon xp 2500 with an biostar m7ncd mb, nvidia tnt2 video card, and windows 98,xp and of course Gentoo Linux if you want me to do linux teesting i'll do that, or if you want me to do windows testing i'll do that too (i'd rather do linux though)

[Varjeal]

WhiteHat: If by handy you mean in an 1 1/2" thick book....yeah..hehe...

[32768]

sourceforge unix name OMK rejected.

changed unix name to OSMK and re-applied...

Sounds good. My sourceforge ID is d32768 (some silly rule about not starting with a number..) Sourceforge was having trouble with one of their servers yesterday, so that may slow down their response. They're usually pretty fast though.

How about twisting the acronym a bit more to get COSMK, which is pronounceable? Not that I should be the one to suggest pronounceable names.. :wink

The C could, in the best GNU tradition, stand for COSMK. Yeah, recursion!

( 2004-11-23 08:07:38 - Project CVS Service ) On 2004-11-23 a disk
failure on the anonymous CVS server for projects starting with the
letters: e, h, i, j, o and v is causing anonymous CVS, ViewCVS and
tarballs for these projects to be unavailable. We are currently working
on this issue and do not have an ETA for resolution.

[32768]

Here's the folew belsaw rules in a nutshell. Varjeal- do you think you could distill something similar for other systems from your 3.8 cm of books?

FB1: No cut in the change key is to be the same depth as the corresponding cut in the master key.

FB2: There should be at least two depths between each cut. This eliminates very thin wafers [i.e. master pins] which otherwise would have a tendency to wedge between the shell and the plug of a worn lock.

FB3: There should be no more than 4 depths between successive cuts.

FB4: There should be no keys or locks set up with all the same cuts.

Here's my first shot at writing these in a formal notation. If we're going to allow users to formulate their own rules, then we'll have to define such a notation and the users will have to use this notation to enter new rules. I think this will be a very uncommon activity for most users so I'm not too worried about making this friendly. As PickPick suggested, there can be an easy way to modify existing rules (for instance, changing the 4 in rule FB3 to 5) and an advanced interface for actually entering new rulesets. I don't expect that users will type rules exactly as I have below, this is more of an exercise to identify what constructs the language will need, like quantification and a way of summing (for FB4).

M represents the master key, K is the change key being considered. There are n pins and m cut depths. The variable i is used to range over pins, and j over depths. Thus, M and K are arrays of size n which hold values ranging over m. |x| represents the absolute value of x.

FB1: for all i, M[i] != K[i]
FB2: for all i, |M[i] - K[i]| >= 2
FB3: for all i from 1 to n-1, |K[i] - K[i+1]| <= 4
FB4: for all j, the number of positions i s.t. K[i] = j is less than n.

[Chucklz]

The FB2 and 3 rules seem to me to require some user input.

Code: Select all

print "What is the MACS?";
my $MACS = <STDIN>;
print "What is the smallest master pin you would like to use?
my $step = <STDIN>;



Oh wait... I just learned this holiday how terrifying command line is to most poeple...

[WhiteHat]

update:

SF project space still pending - probably won't be looked at until monday.

[32768]

The FB2 and 3 rules seem to me to require some user input.

Code: Select all

print "What is the MACS?";
my $MACS = <STDIN>;
print "What is the smallest master pin you would like to use?
my $step = <STDIN>;



Oh wait... I just learned this holiday how terrifying command line is to most poeple...

Yeah- we'll definitely want to allow users to set this somehow. The rules are presented with MACS and step hard coded with those values, so we should leave those as the default and have an option to modify.

[Jarod]

chucklz, that makes perfect sense, i like that ui there too