Matt Kraning writes on LinkedIn:

“Fourteen months ago, I wrote that Menlo Ventures was betting Wispr Flow would kill the keyboard”

To which I say “Maybe?” I recently tried replacing part of my journaling workflow with voice and wound up falling back to an even older technology, paper.

Let’s get the truth out of the way up front. I’m not the target market. I’m 64. I’ve been touch typing for something like 58 years. I love the VT100 to this day. I’m on my fifth Kinensis Advantage 2 #keyboard which fixed my nascent carpal tunnel syndrome. I started using #Emacs around 1979 (before GNU) and it (and org mode) will be a core part of my #writing, programming and personal workflows until the day I die (or get senile and forget the keystrokes that drive it).

That said, here are a few of my recent experiences that inform my “Maybe?”

• Voice Journaling: I journal daily. On paper. I recently spent 5 weeks hiking the remote Kungsleden trail in northern Sweden. I wanted to journal. Before starting the hike I tried journaling with a voice recorder on my phone. I found it awkward. I like to think through what I’m writing. The “tabula rasa” of the voice recorder shut me down, the awkward pauses, wanting to go think through a word choice, wanting to revise a sentence. I wound up buying a small notepad in the Stockholm airport and taking daily notes on that.

• Writing on paper: In the 80s and early 90s I had a saying “If it requires paper, it’s the wrong interface”. I’ve done a 180 on that. I think better with pencil and eraser in hand. It’s mostly linear. It’s slow (that’s a feature). You have time to be careful with word choices, sentence structure and your thoughts. You can draw lines to things. You can include doodles. It’s just a different thought processes and the results are, I think, better, clearer, and at least for me as I write, much more satisfying.

  Incidentally, I’ve had “the AIs” transcribe some of my script. They do a good job. Then I can feed it as electronic text back to “the AIs” (or humans) for review, editing, feedback, etc.

• Writing at a keyboard: I’ve been writing most of my text at a computer keyboard since round 1981, shortly after my dad stopped typing my papers for me. White-out. Backspacing. Literal carriage returns. “Ka-chunk.” I’m writing this at a keyboard. I think it works pretty well. You be the judge.

  Many of the same arguments about paper hold (slower, more deliberate). Writing on a computer is different in many ways than writing on paper in that you are much more free to re-order chunks of text, you’re often on an Internet connected device with the distractions (and aid) of web browsers, AI, etc.

So, my arguments in favor of keeping the keyboard (and the pencil) largely come down to the effect it has on the thought process. In my experience you get qualitatively different results at a keyboard. I think it’s intrinsic to the medium.

Now, lest you say “those octopus-armed emacs using dinosaurs will just stay stuck in the 1970s” I will point you to Sacha Chua, whose organization of the annual-ish emacs-conf and whose weekly Emacs News are the heart of this antiquarian backwater. She’s been experimenting with driving Emacs with voice. So maybe voice is the future.

I don’t doubt that this voice thing might catch on. We’ve had Dictaphones since 1907 thanks to Alexander Graham Bell himself.

The uptake numbers, logic and financials that your post cites (Matt Kraning) are impressive. I will be giving Wispr Flow a try. Can I dump a small amount of capital in and get on the bandwagon?

I will close by taking issue with one statement in your LinkedIn post, you say:

“Typing into a text box is a tax on thought.”

No. Writing with a keyboard, writing on paper, and writing with voice all optimize different kinds of thought. Voice optimizes capture latency. Paper and keyboards (in my experience) optimize thoughtful composition. There’s a place for each.