[DemonRaven]

I have to share this story as it is kinda cute and shows how much the general public is lacking in knowledge of  the vast amount of time and difficulty is involved in creating a plain chatbot. Virtual assistants are easier as most of the responses are in regards to specific enquiries. Unlike some programmers I do differentiate between chatbots and virtual assistants. 

http://money.cnn.com/2016/06/15/technology/facebook-messenger-bot/

I built a Facebook chatbot, but it was awful at chatting

 I just wanted a bot that could talk to my friends, but the bot I built couldn't say much.
The plug-and-play software came pre-programmed with a few keyword responses, such as "Awh, shucks! I love you too!" if someone messaged me "I love you."
I added more replies so the bot would sound like me and hold a basic conversation about the stories I enjoy covering. For example, if someone told the bot they were my friend, it would trigger this response: "Hello, friend! What's goin on."
Despite creating over two dozen more responses, I couldn't customize my bot to run smoothly.

[Art]

Good example, Raven. Yes, I make a distinction between a chatbot and an assistant. Most people don't know (or care) what actually goes in to creating them.

There are some "boiler-plate" ones that let you build your own bot and we'll host it for you (and capture everything you say and do while your bot is hosted by us). Not for me!

Regardless of the type or language in which your bot is programed it still comes down to how it "thinks" and reacts.
Like Steve Worswick once said something to the effect that, "Programming a chatbot is 90% creative writing and 10% coding."  Hopefully close but you get the overall message. It's even better to me if the chatbot is able to "learn" from past conversations and create its own sentences based on what it has learned or knows.

So for those people wanting a "turn-key" ready to talk chatbot with any similarity of intelligence, invest some time learning about what makes them tick and what can be done to make your bot a better bot.

[8pla.net]

Hey Raven,

According to Dr. Wallace, who confirmed it on the Robitron
list, I was the first to put an AIML chatbot on facebook, as
an officially approved facebook app.

There is a possibility, it may be the first ever facebook chatbot.
However, it has been offline for about five years now:


https : //apps . facebook . com / eight-pla-dot-net


EDIT: Dead link deactivated.

[Art]

Why provide a dead link?

[squarebear]

Despite creating over two dozen more responses, I couldn't customize my bot to run smoothly.

OMG! I nearly choked when I read that! Mitsuku has 300,000+ responses and I still find things that need correcting.

[8pla.net]

Why provide a dead link?

Well,  I do remain an approved facebook developer at that URL.
The project is just offline.  However, I agree with you, Art. 
You're right. So, I edited that link into plain text.

[8pla.net]

Squarebear,

Would you share some of the early days of Mitsuku?   

How many responses did Mitsuku start out with in the very, very beginning?

[squarebear]

When I started Mitsuku, I used some AIML files from ALICE, so probably around 20,000 or so responses.. I saw no point in reinventing the wheel so based Mitsuku on the pre written files. I changed many of them to suit Mitsuku's personality but left many of the answers the same.

[8pla.net]

Thank you for your response. 20,000 records to start, and now Mitsuku is 15 times that.  That is useful information in terms of expectations for recently developed chatbots.  Not too long ago, I heard that 10,000 records was the threshold for convincibility in chatbots.

Not to suggest any criticism for being based on ALICE, but have you ever considered a comparison between ALICE and Mitsuku to determine how or whether they differ, in order to upgrade Mitsuku to become more unique?

[DemonRaven]

Despite creating over two dozen more responses, I couldn't customize my bot to run smoothly.

OMG! I nearly choked when I read that! Mitsuku has 300,000+ responses and I still find things that need correcting.

Ya i agree i thought it was pretty funny as they so clueless.