Odd password bug...

[IBMMuseum]

I stumbled upon this creating accounts. In theory both of these passwords should be the same strength, but are not:

MikE321

RyaN321

Apparently all UC/lc combinations having "MIKE" in the password are not the same strength as another four-letter name with the same UC/lc parameters.

Can someone confirm this and explain why (a bugfix would be nice too)?...

 

[Miraenda]

I stumbled upon this creating accounts. In theory both of these passwords should be the same strength, but are not:

MikE1321

RyaN321

I believe you have a typo and meant this:

MikE1321

RyaN1321

You forgot the 1 in 1321. Right now as you have it, they definitely wouldn't be the same strength.

 

[IBMMuseum]

I believe you have a typo and meant this:

MikE1321

RyaN1321

You forgot the 1 in 1321. Right now as you have it, they definitely wouldn't be the same strength.

Yes, a typo. I am comparing "MikE321" and "RyaN321". Those passwords are not the same strength (the password with "mike" is below 50, the password with "ryan" is above 50), even though the same convention is used in both.

 

[IBMMuseum]

I've accepted a reply elsewhere that it is probably an additional dictionary reference when the name denotes another item as well. Certainly "Jack" and "Dick" seems to be the same strengths as "Mike" when they have the same conventions applied, however "Will" does not. So maybe not truly a bug, but it does seem to be a little inconsistant in dictionary references.

I consider the topic now closed, however it might get everyone thinking of names that denote other items as a side effect...