It looks like Dragon users have gotten an excellent gift this year. James Stout, who explores the state of speech input via his blog Hands-Free Coding, is working on fixing something that has dogged Dragon users for years.

In many programs, Dragon can’t track words, and therefore the most intuitive commands for editing after-the-fact aren’t available everywhere. Dragon has at times referred to these as Select-and-Say commands. This partial availability is terrible for speech input users, because once you get used to using a command and it becomes habit, you naturally apply it everywhere where it makes logical sense without even thinking about it.

So the spotty coverage increases cognitive load and leads to inefficient disappointments. Before you can think, you’ve tried to select and delete text saying a Select-and-Say command that works well in Word but doesn’t work in Google Docs, and instead of neatly selecting and deleting a phrase, you’ve got some extra dictation at a random spot in your text that you have to clean up before figuring out how to do the original task a different way.

In his blog post Enhanced Text Manipulation Using Accessibility API’s, James details the technical aspects of the problem and how he’s fixing it. If you’re comfortable installing Python and testing beta software, check it out and let him know what you think. I think this is the best Dragon news of the year, and I’m hoping to hear much more about it next year!