What else is out there like it?

Existing products:

• Voice assistants (Siri/Alexa/Ok Google), VoiceOver, smart watches, all of which can be operated without staring too hard at a screen.
• The snooze button on an alarm — specifically designed to be interacted with with our eyes closed! (Btw, how easy is it to hit snooze on a phone? Not as easy)
• Chorded keyboards
• Stenography keyboards
• One-handed keyers (example, Twiddler)
• There’s an app that just got released called Voiceliner and it allows you to talk into your phone to dictate a hierarchical outline. (Writing this blog post partially using it!)

Existing research:

• Textile interfaces (took class this semester)
• Project Jacquard: digital controls woven into jeans
• Haptics (taking class next semester)

Inspiration by analogy:

• Looking at the evolution of HCI up until present day. In particular, the keyboard -> touchscreen evolution allowed us to interact more directly with a computer. What’s the next step?

How is yours different?

Voice assistants like Siri and Ok Google can’t handle fine-grained controls. If you make a mistake or need to specify any details outside of the expected flows, you have to interact with the screen to complete it.

VoiceOver is too verbose. It is designed for translating a visual interface into an audio interface. The problem is that the visual interface is noisy; therefore, the audio interface is noisy as well. In particular, current UIs throw a lot at the user and expect the user to only interact with it in certain ways. In real life, in a conversation between two people, it’s actually a more two-sided conversation.

Here’s an analogy. Imagine telling a server your order at a restaurant. The equivalent of voice assistants is as if the server read the full menu first before allowing you to say anything, and then expecting you to place your entire order in one go, without pausing or asking for recommendations in between. The equivalent of voiceover is the server monotonously saying “menu”, “section”, “item” repeatedly as you inquired about different items on the menu. In both cases you were forced to prioritize how to give your input rather than letting you say what you really wanted to say.

Project Jacquard seems expensive, because it requires integration into the existing clothing. Also, functionality seems basic — it seems to be mostly for lifestyle-related tasks, and not productivity tasks.

How does it improve what exists?

• Unlike voice assistants, fine-grained controls. (The control is in your hand and has the precision of typing.)
• Unlike VoiceOver, not verbose — gives you just the right amount of feedback, which you can control with your hand as well. Ideally not having a lot of audio feedback at all.

What audience is it for?

• It’s for people who want to be productive.
• It’s for people whose eyes are distracted by the screen all the time.
• It’s for regular, non-technical people.

I think it’s helpful to also describe who I am not trying to target:

• I’m not trying to target just people who can afford expensive technology. Unlike to VR, smart home stuff, etc.
• I’m not trying to target just people who are proficient in coding.
• I’m not trying to target just people with disabilities.

It’s not that I don’t want to serve these groups, it’s that I want to serve not just these groups. Ideally I can make a product that all 3 groups (and many other people) can use as well.

What do you need from a tech standpoint?

• I’ll probably need two Arduinos for each of the two pocket keyboards and fabrication enclosures and as switches or keyboard caps to be able to build the two pocket keyboards.
• I need to write some kind of software for these devices to connect to, to be able to process the events.

If it’s too much for the time, are there any discrete parts of it you could accomplish?

• I want to get the project out quickly. The minimum viable product would be just a small Bluetooth keyboard that can fit into a users pocket while they’re using it.