I just tested it on my site. It wouldn't accept garbage but it did accept a wrong spelling of both words.
--------------- Added [DATE]1202833409[/DATE] at [TIME]1202833409[/TIME] ---------------
Well, I wrote the reCAPTCHA people and this is the resaponce I got:
Quote:
Boofo:
well, I just tried one and I got both words mispelled and it accepted
it.
Support:
This is because on the known word we allow one letter typos -- we've found it makes the user experience better. We don't grade based on the word we are attempting to read, however we do offline studies on user accuracy on this word (based on other people's responses). This allows us to use the reading word to detect abuse.
And here is the expanation:
Quote:
From the description of how reCAPTCHA works:
"Each new word that cannot be read correctly by OCR is given to a user
in conjunction with another word for which the answer is already
known. The user is then asked to read both words. If they solve the
one for which the answer is known, the system assumes their answer is
correct for the new one."
So as long as you get the "known" word right, the other is assumed to
be correct. In your example one could surmise that 'Coat' is the known
word.
Having said this, when testing reCAPTCHA on my two sites that use it,
I have noticed that it accepts leading sub-strings of BOTH words as
correct. For example, for "surprise trouble", "surpris troubl" would
be deemed to be correct. This seems to apply if the sub-strings are
reasonably long, so "s t" would fail in this example. Has anyone else
noticed this behaviour? I am using a bespoke PHP implementation with
recaptchalib.php.
02-12-2008, 09:02 AM
. Stop Spam, Read Books!
.
Linear Mode
