>Even though contests like the Loebner prize may >not lead to "real" AI, they help develop useful >adjuncts to it, as well as tools for more >immediately useful machine interfaces. People >get all amped about speech interfaces and such >even for regular old computers; where do you >think that work is going to magically spring >from? It's a long hard incremental path.
Problem with text/speech interfaces (especially this keyword lookup thing) is that it is hard to find bugs, and you will never be sure that you have tested the system sufficiently.
i think that GUI's have less variations for combining things, so you can test them better.
So can you point to a niche for text based/speech based ui's?
The program enters the Loeber competition and it tries to disqualify the jury.
- hardwired dialogue. Totally unexpected. He must give us the price, just for the idea.
contest entry:
#include
main() {
printf("What are the exact rules of this contest?\n");
printf("What is the formal definition of the Turing test?\n");
printf("And you actually think i am DUMB?\n");
I have another one, it is geared towards tests with multithreaded environment.
here is the project page http://www.michaelmoser.org/jmtunit/
The library sets up a thread pool (n threads) and runs a number of identical test sequences; each test sequence is a virtual user (m users). All threads are waiting on one input queue for test requests. The threading modell is the same as in most of the popular application servers
The programming interface is very similar to JUnit - there is a JMTUnit.TestCase with some additions, there is JMTUnit.TestResult. The tool does print out performance statistics
i guess that the Chinese did have a strong reason not to use an alphabet... Language changes over space and time. (Chinese is a different language in every province). An alphabet reflects the current pronunciation. Hieroglyphs don't have this problem. For our civilization it is not a problem. So what ? You cant read a text written a 1000 years ago ? Big deal! China is different... It would make a big difference to them if they could no longer read those ancient stuff..
Civilizations that are build around a big stream (Ancient Egypt, China) have a longer life cycle...
Slashdot Mirror
... and the PC will be made obsolete by thin
clients, stupid boxes without a hard disk (predicted by Mr. Oracle from way back).
Does anybody remember those days?
>Even though contests like the Loebner prize may
>not lead to "real" AI, they help develop useful
>adjuncts to it, as well as tools for more
>immediately useful machine interfaces. People
>get all amped about speech interfaces and such
>even for regular old computers; where do you
>think that work is going to magically spring
>from? It's a long hard incremental path.
Problem with text/speech interfaces (especially this keyword lookup thing) is that it is hard to find bugs, and you will never be sure that you have tested the system sufficiently.
i think that GUI's have less variations for combining things, so you can test them better.
So can you point to a niche for text based/speech based ui's?
The program enters the Loeber competition and it tries to disqualify the jury.
- hardwired dialogue. Totally unexpected. He must give us the price, just for the idea.
contest entry:
#include
main()
{
printf("What are the exact rules of this contest?\n");
printf("What is the formal definition of the Turing test?\n");
printf("And you actually think i am DUMB?\n");
return 0;
}
I have another one,
it is geared towards tests with multithreaded environment.
here is the project page
http://www.michaelmoser.org/jmtunit/
The library sets up a thread pool (n threads) and runs a number of identical test sequences; each test sequence is a virtual user (m users). All threads are waiting on one input queue for test requests. The threading modell is the same as in most of the popular application servers
The programming interface is very similar to JUnit - there is a JMTUnit.TestCase with some additions, there is JMTUnit.TestResult.
The tool does print out performance statistics
This is a search form that contains very few graphics. http://www.deja.com/=dnc/home_ps.shtml
Civilizations that are build around a big stream (Ancient Egypt, China) have a longer life cycle...