Stenography layout

This is a very far fetched idea, but I absolutely believe it’s worth considering, and it fits the text blade with only minor modifications. A steno layout with possibly even plover compatibility would make it possible for court reporters and live captioning people to have a device with them at all times. They do 200+ wpm on a modified shorthand layout,helping the deaf and hard of hearing. This is impractical or even impossible to do on the qwerty layout.

It’s amazing what stenographers do, and we’d love to help.

TextBlade is re-programmable via a companion smartphone app. We’ll be offering a library of common layouts, and tools for the community to roll and share their own.

Would you need keycaps with steno legends, or would a virtual re-mapping be sufficient?


Wow, thanks for being enthusiastic about this! Great news on the software side.
If you ever decide to implement this, these are the obstacles I see:
it’s a chording layout, with dictionary end everything. This means it’s not going to be a clear cut case of remapping keys, because oftentimes steno defines a single chorde for one word. It also needs 4 separate positions on the space bar, that is two different keys for each thumb. On top of all this the keys need to be scanned independently, that is n-key rollover, or NKRO is needed.

This probably makes this idea feasible only when you are finished with the first version.

Yes, our focus right now is on the QWERTY version, followed by the most requested international + alternative layouts. We’re very passionate about keyboards, and are interested in how our technology can benefit different groups (including the world’s fastest typists).

A few thoughts…

N-key rollover is covered. TextBlade monitors all keys independently in real time.

The dictionary-based translation could be handled by a dedicated steno app.

Regarding the spacebar, would 4 touch-sensitive zones be sufficient?

4 touch-sensitive zones are good for first stage, but lack of haptic feedback is disorienting (on thouse 4 buttons).
I understand this is not a priority now. it shouldn’t be, first you have to nail the QWERTY version, be profitable as a company, etc. I have to say it all looks promising so far. it’s such a simple and brilliant idea to make a separable layout that snaps together.

couple other thoughts:
N key rollover: check, that’s great.
the software side/dictionary capability: sounds good.
middle row of the qwerty layout can be used as a place to rest your fingers(stenotype is two rows, not 3)
haptic feedback on top of the spacebar/vowels can be hacked with different texture tape on top of the keys, as a prototype thing using the qwerty version.

This is interesting. I’ve no experience with stenography machines, but poking around revealed that they go for about five thousand dollars. It would certainly require practice, but imagine the possibilities if the existing TextBlade layout can be repurposed to stenography! Certainly, as speech-to-text recognition advances, there may come a day when stenography is rendered moot. I’m over sixty years old and retired. I have the time and patience to practice cursive with fountain pens, but find this stenography ‘thing’ intriguing. Using hands for sign language doesn’t require technology, but it does require skilled correspondents. Ach! Plover…thank you; time to poke around some more…

Electromagnetically activated keycap ‘domes’ or ‘ridges’ or braille ‘dots’ as haptic feedback?

NKRO is a requirement for steno as it is massively chorded. Plover is the software of choice for qwery based stenography. I’ve dabbled with steno on my poker 2 and the textblade is extremely interesting from the perspective of minimal finger travel. It will depend a lot on whether it can support the appropriate chording and whether this little gem is rugged enough to be an everyday driver rather than an occasional toy for your android device.

Blanks, dvorak, colemak would be nice keysets for these if they are a success. If there are small runs in the store or via massdrop they will sell if the device is practical.

Robert - yes, NKRO (n-key rollover acronym) is indispensable for steno.

This is where TextBlade shines. Every input area has its own dedicated port to a computer (there are 4 computers inside TextBlade).

This means there are no rows or columns, and no possiibility of obscuring any entry by any neighboring key inputs, ever.

It can literally recognize every key down at once - though we don’t assign that to anything, (yet).

There is no other keyboard built with this architecture.

Further, it has global timing precision to 500 microseconds. Notwithstanding the three separate blades, the mutual precision of each event on each disparate blade is all synced to a global network time reference.

It is the most temporally precise keyboard made.

All of this means serious chording is fully supported by this hardware.

A Steno template is totally doable. We will add chord mapping to our MultiMap editor in a subsequent release, and then steno will simply be a template.

This is a very cool area to extend the utility of TextBlade. We’ll address this after ramp up, but from the beginning of our work, we intended TextBlade to be a fully software-malleable development platform.


Nice to hear. Looking forward to a review or two on build quality but it sounds like I will be placing an order as well.
If the api is open source I’m sure there will be plenty of interest in the device from geekhack and the opensteno project. Autohotkey (remaps colemak/dvorak or anything else) and Plover (stenotype + dictionary) both map against the nkro usb keyboards but it is much nicer when you can program your layout directly at the keyboard layers for ease of use and portability (corporate customers often don’t like people installing key loggers on their hardware). Note: 6kro is not adequate to steno.

Can’t believe I missed this thread. Check out this talk from Pycon 2013

1 Like

Mirabai posts regularly on the plover group.

I believe that the textblade fundamentally cannot support stenography within the context of the standard format for chording as taught. This means that you would have to build your dictionary and associated chords from scratch.

Most vocabulary is phonetic in steno systems and the inability to chord the letters phonetically (textblade has many letters to a key) makes developing a vocabulary difficult. Stenographers remember how to chord a word at it sounds so it is not purely memory based. Chording on the textblade is obviously beneficial but it’s use as a stenographer’s tool seems unlikely.n

I know this is a very old post (and I only noticed it!). Curious to know why you think using the textblade within the context of current standard stenography would not be likely possible. I think you explained it, but I don’t think I get it :slight_smile:

I understand that stenographers “type” phonetics and not actual spellings of words, for the most part. As I understand it, it is a transcription of the sounds made, not of the spellings of the words uttered. Why wouldn’t the TB be able to handle this?

(I’ll understand if no one responds; this is an old thread, after all). Cheers!

I don’t think there is anything from a hardware point of view that would prevent this as chords are used extensively right now to trigger jumps, modifier keys and modes. So, the limiting factor would be software. Currently the software is configured to identify specific cords and take some sort of action based on them. A new software mode would be needed for Stenography where the existing chord identification would be suppressed and Stenographic chords could be identified and translated into characters/phonetic syllables. I think it is definitely doable and I believe WT has said as much, though I don’t see it getting any attention until after general release.

I can see WayTools adding this to their product roadmap to work on once all the main stuff has been addressed and TextBlades have shipped at production levels.

My implementation thought would be to add a specific modifier to the Jumps chords that would activate the special Steno mode. Possibly holding the space as you would to access the green layer. Then, the row of keys closest to the SpaceBlade that is normally used for modifier key chords could be used for Steno chording, leaving the middle row of keys for edit chords and cursor movement and the top row for Jumps. Issuing a Jumps chord without the modifier would restore normal operation.

1 Like

I imagine that to be useful to stenograpgers, the style and form of input must be similar to existing standards, otherwise it would be difficult to catch on in an established but small and specialised market.

Agree. I think what I proposed would be a very similar experience to the specialized input devices used for steno.