Subject:
|
Re: Automated password appraisal (Re: New feature: Article rating)
|
Newsgroups:
|
lugnet.admin.general
|
Date:
|
Thu, 30 Mar 2000 18:02:46 GMT
|
Highlighted:
|
(details)
|
Viewed:
|
7763 times
|
| |
|
|
In lugnet.admin.general, Stephen F. Roberts writes:
> ...But anyway, I tried '4Gxc5t'... it came back failed but its reasoning
> was strange... (try it urself :-) the 'slight risk' ones just baffle me...
For one things, it's being a bit too harsh on numeric->alpha conversions like
4->A and 3->E...it should divide the intermediate results by 2 or something
internally while computing the score after a transformaion like that.
> what's a slight risk about the fragment 'xc'? I can see a slight risk for
> 'cst' (central standard time?), but 'agx'?
They might be host name fragments, for example xc.foo.bar.edu or
agx.plonk.com, or AGX might be some acronym.
> ...Most of the other ones I knew were weak came back failed, but it
> fequently chided me for 'slight risk' fragments that dont seem to make
> sense. The above was the only one that I would have suspected to get at
> least an 'adequate' :-/
I'll try to massage the number->letter conversions into being a little more
lenient in those cases. But false positives (positive in the medical sense
of testing for something bad) are OK here. What we don't want is missed
negatives.
> ...I noticed as I tinkered with some of the weak ones and watched the
> changing reasoning, perhaps an additional feature would be to list 'good
> features' of the password. such as saying "Good: 8 characters" or once it
> complained that 'unique charactes <75%' perhaps say 'good: >75% unique
> characters'... by watching the positive feedback, it may be possible to
> 'tune' bad passwords easier.
Ahh, OK.
> ...you can go back to ignoring me now...
No, thanks for the feedback!
--Todd
|
|
Message is in Reply To:
309 Messages in This Thread:
(Inline display suppressed due to large size. Click Dots below to view.)
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
