Are you going to use chatbots in the future?

Now that nobody’s interested in smartwatches anymore, we’ve moved on to cloud-based chatbots that reply to your instant messages right away as the latest progression in human-computer-interaction design. The idea is that you simply type what you want to do in a message to a specific chatbot and they’ll reply to help. Microsoft, Facebook, and Google have announced a number of these chatbots that you can add as friends in each company’s respective instant messaging program.

They’re not new

People are acting like this is something new and exciting, but it isn’t. Chatbots like this have been around for decades. I remember having a list of commands that I could use in order to request information via SMS text messaging on my Voicestream mobile phone service. Banking chatbots for SMS are still around.

When instant messaging became popular in the late 90’s, of course we had chatbots then as well. Scammers use chatbots ALL THE TIME to try to get real people to reveal certain information or sign up for services over instant messengers. It’s usually pretty obvious when you get a robot since they’ll ignore actual questions, but some are pretty sophisticated.

The difference with the scammers and their “SPIM” (Spam over instant messaging) is that they’re contacting you whereas legitimate chatbot services have always been something that you can subscribe to and request information from whenever you feel like it or on a schedule.

You know you want to use DOS on your phone.

Even before instant messaging existed, we had a command line interface for human-computer interaction. You had to learn the syntax or language required to interact with the computer. The only difference between that and chatbots are the chatbots should be capable of understanding and interpreting more human-like languages. I suppose that part is a pretty good thing though.

Is it a better touch screen UI?

The theory is that people spend so much time in these instant messaging and texting programs that it should be simple to be able to type your commands to a robotic friend to get information and perform functions the same way.

For some reason, somebody thinks typing text with a touch screen keyboard is easier than just tapping a couple of buttons in an app. Well, it turns out that mobile app designers are really so terrible at making a GUI that’s easy to understand (what with all these cryptic mystery meat icons and hamburger buttons), that maybe a command-line or chatbot interface actually would be easier to use. At least you’ll be able to understand what it’s saying as long as it outputs text in your language.

Sometimes chatbots completely ignore me though.

How about a speech UI?

Yes! This is where chatbots should be! Currently, companies are putting chatbots into things like Facebook Messenger, Slack, Skype, or other instant messengers where you have to launch the particular app, find the particular chatbot in your friends list, and then start typing to them. That sounds like a lot of work, and it is. Others have created completely separate apps that you have to launch and then start typing to interact with a chatbot.

The smart way to make use of chatbots is to build an extensible speech/text interface that’s part of the core operating system. Microsoft’s Cortana has had this capability for the longest time since it does accept both speech and text based commands in the same way AND it accepts extensions from 3rd party developers. Apple’s Siri and Google’s Assistant/Now generally require hard-coded extensions for 3rd party services, though at least Apple has changed that recently with the newest version of iOS and the SiriKit API.

Have you tried using new chatbots in any of your instant messaging apps? Are they worth interacting with?

Further reading

• Usability Heuristics for Bots
• Chatbots, UX, and Privacy
• Bots won’t replace apps. Better apps will replace apps.
• Are Chatbots Really The Future Of Web Design?
• How The New, Improved Chatbots Rewrite 50 Years Of Bot History

Discuss This Post