Outcome

Intention

I found the topic of the final project aligned with some of the directions I am exploring for my assistantship. My assistantship project New Metaphors is a research into how complex or abstract concepts can be explained through everyday objects or phenomenon. Currently, it's in a workshop format and I wanted to make it more approachable and accessible for individual use. Hence, I built Inspiro. Inspiro can help get designers or creatives get inspiration for their projects by getting metaphors which allows them to think about visualizing ideas  in different ways.

Context

I took inspiration from a couple of Twitter bots. These two twitter bots have been made by Dan Lockton from the School of Design. https://twitter.com/newmetaphors, (Links to an external site.)Links to an external site.https://twitter.com/SyllogismBot (Links to an external site.)Links to an external site.

(Links to an external site.)Links to an external site. These are other similar examples -- https://twitter.com/pentametron/ (Links to an external site.)Links to an external site.
https://twitter.com/metaphorminute (Links to an external site.)Links to an external site.
https://twitter.com/roofslappingbot (Links to an external site.)Links to an external site.

I intended for my bot will build on some of the existing work done for the new metaphors twitter bot. At the moment it just throws an unusual randomly generated metaphor every half an hour. I want to make this more conversational so that the users of the bot can contribute to making new metaphors. I was interested in how through my bot I could create serendipitous moments of inspiration and conversation between the bot and the user + the user with other people.

 I want to still try and create a semi-random mash-up of images and idea/concept words but unlike the twitter bot I wanted to make it more informed through user feedback 

Process

Outline your approach to the project? What ideas did you generate and how did you refine or reject them? What did you research and explore? what were the design choices? What challenges were encountered and how did you resolve them?

I successfully tried out the MiniMagick Ruby wrapper for ImageMagick that allows to read, write and modify bitmap images. One of the issues that I had is to figure out the container size for the text overlay on the image. Second, randomizing the background color and images. Third, how could I connect it database of images and then where do I export the resulting image to send it to the user using Twilio.

1. .MiniMagick is a ruby wrapper for ImageMagick or GraphicsMagick command line. ImageMagick can be used to read and write bitmap images in many formats.
   b. Dialog flow to build a conversational UI with SMS.
2. Minimagick was interesting for me because it allowed me to create composite images and in my case a visual "metaphor" and send it to the user. I also found a few code examples that allowed me to do a text overlay on the image.
3. MiniMagick and, specifically, ImageMagick is great if you have a visual/image component in the service. it can be used to create/select an existing image from a folder/, receive and modify an image and send it to the user.
4. For my project, as I mentioned earlier, I would be using it create "new metaphors" and share them with the user. The image will be created using a composite function in MiniMagick. It will take the user text input and select a random labelled image from a database. I will then create a composite larger image depicting the metaphor using the image and the text input like this.

I'm also including some code snippets that were essential to building Inspiro--

Composite Image

Image file name with time stamp

Returning multiple values from a method

Product

Inspiro is an SMS bot built on Ruby+Sinatra, deployed on Heroku and with Twilio. You can try it out by messaging to +1 2246287718. Inspiro is friendly and it helps you discover, contribute to and create interesting and fun metaphors to inspire new concepts for designers and other creatives or allow users, in general, to understand and visualize complex concepts in different ways.

Reflection

The whole process for this project was exhausting but extremely rewarding. I was able to achieve pulling in text and images from a folder and randomizing the inputs(color, image+image label) and compositing an image metaphor. I also feel satisfied with how I was able to bake it in smoothly with the conversation flow.

Future work—

• Smoothen out conversation flow: Currently, the interaction works smoothly for the first interaction but I need to set-up the session variable to clear out once the conversation is over.
• Allow for the bot to handle different kinds of responses using dialog flow
• Connect with a database to allow to record user responses on the metaphors which can be used for further research. 
• Users may not always intentionally engage with the bot so I may want to add scheduled tasks to keep users engaged.
• I also want to explore what form could it take on other platforms like Alexa show and perhaps integrating with dribbble.