Posey's Tips & Tricks
The Practicality of Conversing with Copilot, Part 2: Navigation
Drawing on lessons learned from Dragon NaturallySpeaking, this second installment examines where voice-based navigation and dictation in Microsoft Copilot may streamline workflows -- and where traditional input methods are likely to remain more practical.
In my first post in this series, I explained that there are quite a few parallels between Microsoft copilot's speech capabilities, and the old Dragon Naturally Speaking software that I used to use to write most of my articles. Being that I used Dragon for so many years, I wanted to share some lessons learned and how I think those lessons will apply to Copilot.
One of Dragon's lesser publicized capabilities was that the software could be used to verbally navigate your computer. You could for example, launch programs or select menu items by using your voice in place of a mouse. Honestly, I never used these capabilities in production. Sure, I tried them out, but found that it was just quicker and easier to use a mouse or a touchscreen than to try to verbalize whatever it was that I was trying to do. I fully expect things to play out the same way with Copilot. Ultimately, it might not be a big deal to use your voice to initiate a workflow, but I just don't think that it is going to be practical to use a microphone as a replacement for a mouse.
Another big lesson that I learned while working with Dragon Naturally Speaking was that some types of documents were more practical to dictate than others. I found that Dragon worked really well for writing opinion type blog posts (something along the lines of the blog post that you are reading right now). It also worked well for explanatory pieces. However, there were two types of content that Dragon really did not work well with.
First, Dragon wasn't practical for writing documents that contained a lot of source code. Imagine trying to verbally write a PowerShell script for example. Sure, it can be done (with enough patience), but it's way faster to write the content the old fashion way than it is to try to verbalize all of the symbols used in the language's syntax.
The other type of content that Dragon didn't do so well with was anything involving step by step instructions. In all fairness though, I think that my own individual writing style was as much to blame as Dragon.
The problem stemmed from Dragon's built in editing commands. Suppose for example, that you spoke a phrase and Dragon typed something different from what you said. In such a situation, you could say, "correct that," followed by the text that should replace whatever it was that Dragon got wrong.
Some of these editing commands were obviously super useful. However, it was the "select" command that always caused problems. The idea was that you could tell Dragon to select a block of text and then perform some sort of action on the selected text, such as applying a bold font.
The reason why the Select command caused me so many problems is that I would sometimes speak phrases such as, "click on the file menu and select the Save As option." Dragon would often hear this phrase and interpret the word Select as a command rather than a word to be typed. It would inevitably find the word "the" somewhere in my document, select it and then replace that word with whatever I said next. Needless to say, those glitches were infuriating. It even happened without me realizing it on more than one occasion, leading to emails from confused editors.
This is one area where I think that Copilot is going to make a big difference. Copilot is based on GPT-5, meaning that it has a much better understanding of context than Dragon ever did. As such, I think that most of the problems associated with spoken words being incorrectly interpreted as commands or incorrect word usage (such as right vs write) probably aren't going to be an issue for Copilot.
The version of Dragon that I used had some minimal mobile support that was designed to help you to be productive while on the go. These capabilities largely involved dictating to a device and then using Dragon to transcribe the recording later on. I tried dictating in the car a couple of times (as a passenger), but when I had Dragon to transcribe my recordings the resulting documents required so much editing that it would have been faster to start from scratch rather than using my recordings.
My guess is that the mobile experience is one area where Copilot will really shine. I'm not sure how practical it is going to be to write lengthy documents while on the go, but I absolutely believe that Copilot on a mobile device will be useful as a personal assistant. I can just imagine being on my way somewhere and telling Copilot to add an appointment to my calendar or to send an email to someone.
Copilot for Microsoft 365 is a hugely capable and feature rich experience. Some of those features are going to be every bit as good as what Microsoft advertises. However, other features will inevitably prove to be impractical when used in the real world.
About the Author
Brien Posey is a 22-time Microsoft MVP with decades of IT experience. As a freelance writer, Posey has written thousands of articles and contributed to several dozen books on a wide variety of IT topics. Prior to going freelance, Posey was a CIO for a national chain of hospitals and health care facilities. He has also served as a network administrator for some of the country's largest insurance companies and for the Department of Defense at Fort Knox. In addition to his continued work in IT, Posey has spent the last several years actively training as a commercial scientist-astronaut candidate in preparation to fly on a mission to study polar mesospheric clouds from space. You can follow his spaceflight training on his Web site.