AI Zone Admin Forum Add your forum

NEWS: Chatbots.org survey on 3000 US and UK consumers shows it is time for chatbot integration in customer service!read more..

My thoughts on the 2014 competitiion
 
 
  [ # 46 ]

I now understand the history and reasoning for LPP, thanks for the clarification and it makes perfect sense in that context.  Sorry for my less than measured initial comment on LPP.  smile  I would be very interested in the set of use cases any chat protocol for the loebner would need to meet.

As one possible alternative I would like people to consider XMPP.

XMPP (http://en.wikipedia.org/wiki/XMPP) is the widely used ( google talk, apple ichat, Facebook, etc ) and open ‘Jabber’ protocol.  It has libraries for any language you can think of: http://xmpp.org/xmpp-software/libraries/.  It’s whole purpose is to be a text chat protocol.  It’s an open standard that is maintained and evolved by IETF.  Using the libraries it’s trivial to hook a bot up to any XMPP network.

Judges could use one of the many out of the box jabber clients to interact with the various participants and it would be fairly simple to guarantee bots were doing the talking on a closed network controlled by the competition organizers ( ssl using competition supplied and installed certs for example ).

And most of all - people who could get their bots to communicate over XMPP would have something very very useful - the ability to be a jabber client on most chat networks in the world.

 

 
  [ # 47 ]
Dave Morton - Nov 21, 2014:

Hi, Rob, and welcome!

The problem with remote chatbots is that you can’t really guarantee that it’s not a human on the other end, and that negates the value and importance of the contest.

This problem got me thinking…  How about CAPTCHA for robots.  lol!  no, seriously, humans can’t answer certain kinds of questions quickly but robots can.  Could a system be devised that _could_ verify that it was a robot on the other end?

 

 
  [ # 48 ]

What would stop a human from exploiting such a system by taking over only after having a computer program answer the CAPTCHA-equivalent?

 

 
  [ # 49 ]
Jarrod Torriero - Nov 26, 2014:

What would stop a human from exploiting such a system by taking over only after having a computer program answer the CAPTCHA-equivalent?

True, and human captcha has the same problem of course.  I imagine spam bot networks are already using services like Amazon’s Mechanical Turk (https://www.mturk.com/mturk/welcome) to offload the ‘are you human’ (Human Intelligence Tasks - HITs) questions to real humans… wink

The internet in general is identity challenged, that isn’t likely to change until identity is embedded into the hardware, which of course I hope never happens ( big brother and all… ).

 

 

‹ First  < 2 3 4
4 of 4
 
  login or register to react