[alicebot-archcomm] Loose ends

[Ernest Lergon]

Hi all,

as you might have noticed, I often become fundamental (no, Kim, I'm not
a Taliban ;-).

So I like to bother you with some statistics:

Out of more than 40.000 Alice cats

	    0	uses <javascript>
	    0	uses <gender>
	    2	uses <person2>
	    2	uses <gossip>
	    8	uses <condition>
	   53	uses <topic> (46 different topics)
	   75	uses <input index...>
	  352	uses <that index...>
	 1117	uses <random>
	 1267	uses <that> (pattern-side)
	 1758	uses <person>
	 9946	uses <star index...>
	16771	uses <srai>

Don't take those numbers absolut, there are cats doubled in French or
just test cats. But they show a trend.

According to NYT Alice is the 'best artificial-intelligence program on
the planet'.

She is that without having the abillity to i.e. compare two values.

The statistics show, where the main focus has to be seen: On the
stimulus-response mechanism using symbolic reduction - just following
Zipf's law. Let's think about refining this instead of implementing just
new functions.

Or to say it in another way: Is there really a need for tags, which
mutate AIML into a real programming language? Are there other usage
statistics? I have the impression - mainly by looking at the style list
- that many things, even shuffle, array pop etc. - can be done in pure
AIML, aka "programmed" ;-)

On the other hand: If we take AIML serious as Artificial Intelligence
Markup Language in general, why shouldn't we think about a <gsrai> tag
for instance? Maybe I come back to this while trying to teach Alice
German - a kind of 'grammar srai' might be helpful there to reduce words
to their basic forms, work with these and create syntactical correct
sentences from them. Don't forget the different use of personal pronouns
in German or other languages compared to English ('du', 'Sie' and 'ihr'
are all 'you'). Is somebody out there who worked already on this topic?

And - just to play advocatus diaboli - should a general AIML not be more
extensive? What about neural networks, knowledge webs, reasoning etc.
defined by AIML tags? All that stuff normally subsumed under AI.

Don't crucify me, but we must be aware of this - maybe just in order to
clarify for us, how far AIML should go.

But back from visions to reality:

Browsing the archcomm mailinglist I found many loose ends:

- patternside disjunction
- 'predicate' as predicate part of a sentence
- meaning of 'equal' (touches my <condition> proposal as well)
- <shuffle>
- <set var=...> (despite overnight voting)
- <add ..>
- <script...> aka extension mechanism

There might be more. In my impression some of the above threads were
just fizzled and neither brought to a decision nor discussed in depth.

Who let the worms out?

Ernest


-- 
              ProgramV - Alice on Perl - available at
               http://www.virtualitas.net/perl/aiml/

      VIRTUALITAS - Manufacturer of fine OOPPS - since 1996
*********************************************************************
* VIRTUALITAS Inc.               *      http://www.virtualitas.net  *
*                                *                                  *
* European Consultant Office     *                                  *
* Internationales Handelszentrum *                                  *
* Friedrichstraße 95             *   contact:Ernest Lergon          *
* 10117 Berlin / Germany         *    mailto:Ernest@virtualitas.net *
*********************************************************************
       PGP-Key http://www.virtualitas.net/Ernest_Lergon.asc