Talking is different than writing: the effects of voice-recognition software on writing

Talking is different than writing.  Plain and simple. 

I have been using the online tool Jott, which allows you to phone in up to 30 seconds of spoken words and then Jott transcribes it into text.  You can send Jotts to other people (who receive both audio files and the transcribed text) or just to yourself.  It's a great time-saver when you have lots of short emails to send or other short blurbs to write and also have a lot of time when you are somewhere computer-less (read: in a car commuting 90 minutes each way). 

I've been using Jott for emails, blog posts, and reminders for myself for some time now.  I found that how I talk is a lot difference than how I write.  Not only the tone, but the content, the word choice... everything is different when I am speaking instead of writing.   Content is a lot more superficial, a lot less coherent, less formal, and a lot less structured...so it ends up being a little bit rambley and miscellaneous (kind of like this post, which was written with Jott).

I think that as voice recognition software takes off, becomes commonplace, and accurate, we will need to think about this overall issue of speaking vs. writing.  Here are things I've been thinking about in relation to the subject:

• Am I actually performing and speaking as well I would be if I were writing?
• Am I spending a lot of time correcting mistakes the software made?
• Am I saving time and pretty soon producing the same quality of material?
• Or am I am saving time and producing junk?
• Am I even saving time at all?

I think these are important questions that need to be considered when considering using tools like this that are supposed to enhance productivity. I think that tools like Jott, and voice recognition software in general, have huge power as tools, but I think it really needs to be something that people think about while using.  Assuming that tools like this enhance productivity is a big assumption so early in the game.