[OSM-talk] ReMAPTCHA Demo BETA 0.2 online! (Was: Hate captchas!!!!)

Stefan Keller sfkeller at gmail.com
Sat Mar 29 17:05:47 UTC 2014


Hi moltonel

You wrote:
> I'm worried about bots still having a very high chance of sucess. With
> two fairly-legible words in the image and a chalenge asking me to
> write either one of the words or both, a bot still has 33% chance of
> success if answering randomly

I can't follow what 33% means and why you are more worried than about the
usual CAPTCHAs: The known word (the one which is more scrambled) in our
ReMAPTCHA currently consists of 5 characters, and the word is placed
anywhere in the image - slightly tilted. This makes it at least as
difficult as usual CAPTCHAs.

Yours, Stefan


2014-03-29 14:21 GMT+01:00 moltonel 3x Combo <moltonel at gmail.com>:

> On 28/03/2014, Stefan Keller <sfkeller at gmail.com> wrote:
> > The ReMAPTCHA Demo BETA 0.2 finally is ready to be tested:
> > http://remaptcha.herokuapp.com/
>
> Great, looks promising :)
>
> > Please note that perfomance is slow (so hover over satellite icon taks
> > seconds), and that the data is restricted to Switzerland since that's the
> > imagery data I needed (if anyone knows freely accessible aerial/satellite
> > imagery of similar quality I'm interested).
> >
> > I'm open for hints and enhancement requests (excuse me that I can't
> answer
> > then all).
>
> I'm worried about bots still having a very high chance of sucess. With
> two fairly-legible words in the image and a chalenge asking me to
> write either one of the words or both, a bot still has 33% chance of
> success if answering randomly, wich is high enough that bot authors
> won't even bother trying to smartly interpret the map.
>
> The most direct way to reduce those odds is to put many words in the
> picture. But that'll quickly get messy and as annoying as traditional
> captchas. How about getting rid of words alltogether ?
>
> First of all, display only the satellite imagery, because the map
> rendering is too parseable for computers, and not really necessary for
> humans. Then ask the user to click on specific areas of the image,
> using your knowledge of the underlying map data. Some challenge
> examples :
>  * Click on the car park if you can reach it from the motorway,
> otherwise click the big building.
> * Click on the playground, then the fountain, then the lake
> * Click on the road that leads to the football pitch
> * Click on one of the park entrances
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20140329/bec39ba6/attachment.html>


More information about the talk mailing list