So. It’s just after the holidays, and a whole slew of people opened a new Amazon Echo or Google Home brought by a jolly elf as a gift. Now, they are rolling their eyes or scratching their foreheads over the responses they get from their supposedly smart home assistants, or (if they are like one of my office colleagues) they are claiming the thing is trash, but perhaps better, they are praising the heavens and enjoying the heck out of artificial intelligence and Voice-User-Interface. Hallelujah for converts!

Many along any point on that spectrum are wondering how the heck it could be so difficult to get that stuff right, and many of you, dear readers, are wondering what this has to do with the price of peas and technical writing.

Allow me.

You see, a few weeks ago, my development team here at my office had what we all an “innovation sprint.” That is where we get to blow off some steam and quit working on our traditional run of the mill development tasks and think, instead, about the kinds of things we would do if we didn’t have customer deadlines and goals. We get to open up our imaginations and play around a bit – so what we did was to imagine what would happen if we could get something like the Amazon Echo to talk to mainframe computers and therefore get mainframe computers to talk back to the Amazon Echo – the ultimate in automation processes, as it were. Now mind you, the entire team does not have to participate. This process is totally voluntary. Team members can catch up on overdue work or sketch out other projects that they may want to work on in the future. They can conceive of ways to make their workdays easier or to otherwise improve the customer experience, that sort of thing.

But I was sticking with the Echo. I did not think it would work, but I am a glutton for impossible tasks.

I was right, and we failed fast. Too many security issues, turns out. But what did work, and what tickled me to no end, was that it IS possible to ask the adorable Voice User Interface to search my well-crafted user documentation (a tome that would be some 6,000 pages if it were printed, mind you) for any string of words or numbers that you wish to find, and the marvelous Alexa can be taught to return results with alarming accuracy. Voila!

Just like Alexa can learn movie trivia or what time the bus will arrive at my stop, just like this model of artificial intelligence can archive my weekly grocery list and take on my daily calendar or read each day’s news from a variety of prescribed sources, this cylindrical wonder will root through those words upon words upon numbers and more words, and even acronyms and reveal the right answer time after time. Once we program her.

Eureka!

In addition to these innovation sprints, my company also supports a project called the Accelerator. Think “incubator” but with less risk. A gargantuan company helping to nurture fledgling startups. It’s pretty great. First there is a rigorous application process and proving ground, naturally, but once you’ve run the gauntlet, it’s quite the ride – a supported startup, if you will. And that is where I am headed, my Alexa by my side.

A technical writer’s dream, and maybe nightmare, all at once.

It seems I have opened an oyster of sorts.

I found a path to documentation using VUIs, a search tool for the largest documentation set with which I have worked. And after all, the Alexa was named after the Library at Alexandria, so it stands to reason. I am pleased as punch. Let the innovation continue, and the documentation shall “speak” for itself.