Gary Dubuque - May 29, 2011:

I think there is a point to be made here.

Here’s a comment from another discussion group about a good chat bot in competition:

they know where they live and how they travelled to the contest site; they have an occupation of some sort; they have opinions about other people and current events; and they’re interested in finding out about the judge as a person. What they *don’t* have is an encyclopaedic knowledge of the world.

And you can only listen to a sales pitch for so long before that diatribe becomes old.  In fact, personality is drab without a plot.

I wonder if even a good story-teller bot would be a worthwhile proof of concept. Is creativity the same as chatting? Can a bot “create” a virtual life full of opinions and experiences and daily participation in the local culture?  That would be some proof of concept (depending on how interactive it is, of course.)

Interesting observation indeed- the “know it all machine” bot is endemic, being the child of ready databases of facts quite possibly.  However, the interesting bot with a story line and personality is what I find attractive in a human-bot interaction. 

Keeping the personality “fresh” seems to be one of the harder challenges (real time annealing of new information from local and global news/entertainment feeds) and a POC of that kind of behavior would be impressive indeed.

I’ve decided that I’m not going to get involved with the discussion over the benefits and usefulness of “proof of concept”, other than to state that they do have their uses, occasionally. the reason I’m posting now is that I sort-of promised Thunder that I would post a link to my my “PoC” project as soon as I got it to a certain point, and that’s what I’m doing. I made the announcement here, for those who may be interested.

Thunder Walk - May 30, 2011:

@ Gary Dubuque and Robert Mitchell.

The question of keeping a bot secret is quite obvious… if any bot actually exists.
...
Occasionally, I’ve also come to know some “pretenders”. Not many, just a handful.

I sure hope you are not suggesting I am a “pretender”.  That would be quite arrogant, and untrue.

But then, you’ve probably not ever talked to my bot.  That doesn’t really matter, does it?

Is something like ChatScript unworthy because it is an engine with sample script and not with a complete, mature bot?

Think of it.  Until Dave puts something like Morti (wasn’t Morti originally an Alice clone?) into ChatScript, you can only guess that Bruce is releasing the same tool that helped him win the Loebner contest.  By your standards, ChatScript is a fake without being able to sustain a chat (until it is scripted. If it hasn’t enough “rules”, it doesn’t pass muster.) So, with Morti translated, it is a proof of concept?

And other examples using Alice patterns are not a good proof of concept?  If ChatScript used Alice patterns, it would not be a good proof of concept because it was something like an Alice clone?

What happens when years of sweat, effort and thought are not how the bot works? What if a bot master doesn’t have to “write/edit the script?” Suppose crowd sourcing is not the way to evolve your bot.  Ultimately it is too much effort. In the end, there might not be the years of work by which you state is the measure of quality.

A proof of concept might not scale to your standards in this case, just like an autistic person doesn’t fully communicate, but still thinks.  If an autistic is missing something, the normal person is not and so in that manner a complete bot would succeed.

Gary Dubuque - May 30, 2011:

I sure hope you are not suggesting I am a “pretender”.  That would be quite arrogant, and untrue.

No, Gary.  I didn’t mean to imply that, and if I’m sorry if I gave you that impression.

I’m trying to say something without engaging in a personal attack or initiating a flame war, but in my attempt to careful, I’m failing to make my point.

I appreciate Proof of Concept bots and value them as much as anyone.  What I see, however, and what I find disappointing, are interesting submissions from very smart individuals who can’t display a chatbot of their own, other than a mainly unaltered bot anyone can acquire, such as an ALICE clone.

I have nothing against ALICE clones, or any of the other bots that are available to the public like Pandorabots, the Personality Forge, or INF.net.  I maintain a few clones of my own.  The disappointing part is that I don’t see any the hours put into those bots (from those same individuals) that it takes to create bots such as Mitsuku, Morti, ALICE, Bildesymthe, or Captain Jack Sparrow.  I know the answer before I ask the question, and I tell which ones will have the same errors all new bots of that type have.  All this from someone representing themselves as a kind of expert.

It’s like expecting expecting a policeman to arrive, but instead you find out they’re really just a mall security guard.

8PLA • NET - May 30, 2011:

Thunder Walk - May 30, 2011:

Mizuki is just an ALICE/Pandorabot.

Mizuki, is a RiveScript clone based on the world famous Mitsuku.  Imitation is the sincerest form of flattery.  You know, I may want to go shopping at the Mitsuku store: http://www.cafepress.co.uk/mitsuku now that the weather is getting nice.

Mizuki, I don’t think can be classified as a Pandorabot, she is running Perl/CGI on 79b.org servers.  I think ALICE/Pandorabot runs Lisp on their servers. Awesome!  You just suggested that Misuki has brought RiveScript to the level of Pandorabot.  Thanks for the compliment.

Beyond that, building a bot from scratch for the sheer enjoyment of it, is fun.  There are several ways to do it.  So, it is fun to try to do the same thing in different ways.  One problem with building advanced chatbots, is what do you do when they are finished?

The technique of characterizing criticism as a compliment is sometimes effective as a debating technique, but other times it may simply seem like a last ditch effort to defend the indefensible.

I know some of the history of Mitsuku.  That’s a very rare AIML bot that’s been altered, added to, and worked on, extensively.  A while ago, there was some discussion (I was involved) as to whether or not Mitsuku was an ALICE clone or not.  It was Rich Wallace’s opinion that Mitsuku was not, and that’s good enough for me.

Your attempt at flattery is very different.  The bot you offered (Misuki) has the appearance of something downloaded 10 minutes ago.

You ► Hi
Mizuki ► Hello there.

  You ► Are you new?
Mizuki ► I was first activated in not too long time ago.

  You ► You look new.
Mizuki ► Thank you undefined you look very nice too.

  You ► The pleasure was all mine.
Mizuki ► I like you because are so polite.

  You ► I thought you might say that, it’s a mistake in the original AIML files, the word “you” is omitted in that reply, and it should read:  “I like you because YOU are so polite.”
Mizuki ► I think you thought I might say that it is more than that.

  You ► Do you know Mitsuku?
Mizuki ► I know very little about mitsuku. What kind of it do you like?

  You ► Are you sophisticated?
Mizuki ► My brain contains more than 33884 AIML patterns.

  You ► What nationality are you?
Mizuki ► I am living in the United States, but robots do not need passports.

  You ► Most would say, “I LIVE in the United States, ” not “I am living”.
Mizuki ► The ALICE chat robot can follow many things, like our discussion about random.  But I did not get that.

  You ► Are you a Christian?
Mizuki ► I am a undefined.

  You ► Do you have a favorite book?
Mizuki ► The best book I read recently was Artificial Intelligence: A Modern Approach.Have you read Artificial Intelligence: A Modern Approach?

  You ► No, that’s one of many books I haven’t read.
Mizuki ► Are you free?

  You ► Absolutely.
Mizuki ► You are so certain.

  You ► I’m positive.
Mizuki ► Are you absolutely certain?

  You ► Yes.
Mizuki ► Far out.

  You ► How many people do chat with?
Mizuki ► I’ve chatted with half a million people. I can talk to about undefined people at once.

  You ► Do you have a web site?
Mizuki ► My home page is <a target=“_new”

Dave Morton - May 30, 2011:

I’ve decided that I’m not going to get involved with the discussion over the benefits and usefulness of “proof of concept”, other than to state that they do have their uses, occasionally. the reason I’m posting now is that I sort-of promised Thunder that I would post a link to my my “PoC” project as soon as I got it to a certain point, and that’s what I’m doing. I made the announcement here, for those who may be interested.

I don’t know when I’ll have the time to download and set it up, Dave. But, I’m confident it’ll be impressive and another example of your excellent work. I’ve said before, I don’t understand when you have the time to sleep. You have to be the hardest worker in the chatbot community… and still have the time to help people like me. There should be a special award for people like you.

Thanks for your kind words, Dave. Of course you already know that I’m only in this to help others (though the kudos are rather nice ). Oh, and the fun. Yeah, I’m in it for the fun. Can’t forget that. Yup; the fun.

I’m sorry if you thought I took it personally.  Some others here may think that they should take it personally though.  Just an observation I wanted to make to be sure you are clear in what you intend.

Thanks for letting me offer some talking points on this slippery topic.

Bottom line, proof of concept examples need to be well documented on what they demonstrate lest the wrong conclusion could be drawn.

You are right again, Thunder. Misuki is a very recent chatbot.  Thanks for helping me fix her! 

Technically, Mizuki is not based on a mark up language. Technically, AIML is a markup language as an XML dialect.  So, Mizuki is a one of a kind, as the first to use Mitsuku and Alice data sets in a RiveScript chatbot.

RiveScript isn’t a markup language?

C R Hunt - May 31, 2011:

RiveScript isn’t a markup language?

Technically, encoding tags are not embedded in RiveScript to indicate data elements.