"although we have no reason to believe password information ever left Twitter’s systems or was misused by anyone" for twitter and "GitHub said it discovered its error during a routine audit and made it clear its servers weren't hacked.". How can they be sure that there havn't been hacked? does enetering a server always leave traces?
@Kepotx probably mroe that they came across an error during their own audits rather than finding the passwords scattered all over the internet like dog toys.
Yesterday, Twitter anounced that they recently identified a bug that stored passwords unmasked in an internal log.
Recently, Github also had a similar bug
In both cases, they pretend that nobody had acces to the file
Twitter:
We have fixed the bug, and our investigation shows no indication...
a possibility though: if the passwords were saved indeed accidentally in that internal log (in Twitter's case) they might not have full file access logs for that
The newer picture-based ones are only needed if I do the initial account sign-up on my phone (where they can't detect my humanlike hovering pattern), but I can usually get through them after one or two screens.
Spammers suck because they made captchas necessary.
I never know what Google thinks is part of a street sign. Do I still check a square if it contains just an edge? Do I include the pole? Is a velomobile a car?
@Arminius because it is AI-based, it isn't about what Google thinks is part of a street sign, it is what we think is part of a street sign.
(maybe the majority of the people?)
@TomK. what do you mean by fail? I didn't know it was possible to outright fail (I thought it just kept presenting screens to you until it had determined you were human).
the worst one I had recently was hills. One picture had nothing but grass in it. How am I supposed to know if that grass is part of a hill or not?
@NH. Well, when I already know which object I want to frame, I'm not sure why Google would be interested in keeping the instructions for indicating that object ambiguous.
@NH. Basically captcha had 2 words, 1 known and 1 unknown, so if they manually had a large number of people type the known word correctly, but put "penis" for the unknown word, a significant percentage of "known" words will incorrectly be "penis", at which point they can automate it.
@JourneymanGeek basically the first picture (of a word) was already known by the algorithm, but the 2nd wasn't. so when the captcha was implemented at first you only had to "guess" the first word correctly
Yesterday, Twitter anounced that they recently identified a bug that stored passwords unmasked in an internal log.
Recently, Github also had a similar bug
In both cases, they pretend that nobody had acces to the file
Twitter:
We have fixed the bug, and our investigation shows no indication...
