Hacker News Reader

Comment by jaymzcampbell | Hacker News Reader

Creating the rules for it reminds me a little bit of writing a text adventure game from days gone by:

    ...
    name: {
      onEnter: "'What do you want to name your kitten?'",
      exits: "'*' ->respond_to_name name=INPUT",
    },
    respond_to_name: {
      onEnterSay: "The kitten purrs happily, I guess it likes the name #/name#!",
    },
    ...

The readme for it is nice and detailed which I always appreciate and it's turned me onto http://tracery.io/, which I'd not heard of but looks like a lot of fun ("A super-simple tool and language to generate text").

View on HN

> Creating the rules for it reminds me a little bit of writing a text adventure game from days gone by

Interactive fiction has long been the playground for conversation UIs since its inception (Zork's development is even directly entwined in the early history of MIT's AI lab). It's also a useful thing to ask someone starting a bot project: what was the last IF game you played? Have you written one?

There's a lot of advances in conversational UI in the IF space that are worth exploring when writing bots, too. It's definitely worth spending some time exploring recent and modern IF games for ideas. I recommend Emily Short's blog [1] for a lot of discussions and examinations of especially conversational UI in IF over the last few decades (that's one of Emily's particular passions).

[1] https://emshort.blog/

I have been building/running a pretty successful chatbot for a large corporate for over a year now. We use Inkles Ink to script the conversation flows and maintain state. Works very very well. https://github.com/inkle/ink

Inkle/Ink looks awesome and a lot of fun - thanks for sharing that! Mutters looks like a really cool use of the tech btw.

Thanks!

Neat, so I just read through ink and its tutorial.

Can you briefly describe the kinds of things you use it for in a large corporation? Thanks!

Sure.. it is used to manage all the conversational flows and state for a chatbot. We use it for both branching conversations flows and goal based flows. Ink is simple enough that our content writers can work directly with it and we get clean separation of content and code.

The bot itself is pretty interesting. It has 350+ conversation flows, largely dealing with customer service for a business with fairly complicated products and business rules.

The bot is plumbed into 7 odd channels and has to deal with channel specific answers (eg if you are in the mobile app or the self-service portal then the answer on how to do something is different from how to do it if you are on the website's home page).

Additionally it has to handle regional specific answers, so the bots answers are country specific as well.

And it deals with logged in and not logged in states too :)

This all makes for an interesting content management problem (40k words) that our use of Ink readily solves.

Currently bot has 68% conversation success rate. E.g. 68% of the time he is able to resolve the customers issue which we think is pretty good considering the wide domain we are dealing with and the very strange edge cases the industry we are in has.

Ok cool, that maps to what I briefly read about ink much better. I had in mind some kind of 'devops chatbot'.

68% conversion rate is pretty impressive, nicely done! And thanks for the description, it's quite fascinating.

Thanks. The really fascinating thing is have a direct "stream of conscience" from our customers. With a 1000+ conversation a day you really get to see the problems they are having with our products, processes, UX etc As well as their responses to things in the news etc. Our product teams love it.

That's a lot of conversations! Have you considered using some machine learning techniques?

We use ML for Intent matching and NER and Ink to control the conversation state and flow. The core of this has been open sourced so you can see how we will the two together in practice here https://github.com/rabidgremlin/Mutters

Hey - @joethephish here from inkle! Really cool to to hear that you're using ink for that :) Would love to know more detail about where this is being used if you're able to share!

Hi! Thanks for creating Ink it has been awesome for us. Let me talk to the bossman to see what I can share :)

Hiya. So the Chatbot is Oscar, Air New Zealand's chat bot. Here is a recent press release about him: https://www.airnewzealand.co.nz/press-release-2017-air-nz-ch...

Well, the README directly mentions Twine, which is an interactive fiction authoring system. Although I believe it is further to the “branching story” end of the spectrum than things like Inform and TADS.

I'm starting to introduce someone to programming and am putting together a MUD-style game in Typescript for them to hack on. I kind of wish that instead of magic strings in JSON this was a JS object where the values could be functions. Introducing more than one syntax for variables, etc., could be confusing at the beginning.

Reply on news.ycombinator.com