Captcha identify

[Ian Kelly]

On Thu, Aug 14, 2014 at 12:37 PM, Peter Pearson
<ppearson at nowhere.invalid> wrote:
> On Wed, 13 Aug 2014 14:16:02 -0600, Ian Kelly <ian.g.kelly at gmail.com> wrote:
>>
>> . . . and as computers get more powerful the intersection
>> of {problems machines can't solve} and {problems humans can reliably
>> solve} grows ever smaller.
>
> "Which of the following eight sentences are sarcastic in tone?"

People are actually pretty bad at identifying sarcasm, or at least
agreeing on what it is [1] and there's a segment of the population who
simply don't understand it at all, so you'd be trading one kind of
inaccessibility for another.

And you might be surprised at how good machines can be at identifying
sarcasm [2].

Besides, this suffers from the problem of a limited pool of questions,
in that the spambot could simply build up a database of which
sentences that are used by the system are sarcastic and which are not,
as determined by their human controllers.

[1] http://psychcentral.com/blog/archives/2009/08/18/would-you-even-recognize-sarcasm/
[2] http://www.bbc.com/news/technology-23160583