plover.. steno on a qwerty keyboard

[Tiramisuu]

http://openstenoproject.org/

255 wpm is pretty interesting.  22 keys and chorded.
The ergo dox looks like a nice fit but any nkro works.

Would love to learn and have installed the keyboard but not sure where to find good tutorials yet.     

Colemak is keeping my brain baffled but the only real upside is less hand movement.   Doubling my effective typing speed would be very cool.

[Proword]

All this needs is WordPerfect software.  Can be used on any keyboard that the user is familiar with.

http://proword-transcription.blogspot.com.au/

I've been a court reporter since 1990, and I'm wary of the 99.9% accuracy claim, since the speed and accuracy of audio transcription depends upon the quality of the audio as well as the skill of the reporter.  I certainly work to 99.999% accuracy, but that involves proof reading the entire transcript against the audio recording which is usually not taken into account when calculating the speed.

[jacobolus]

Some other geekhack threads discussing plover:

http://geekhack.org/index.php?topic=4910
http://geekhack.org/index.php?topic=16927
http://geekhack.org/index.php?topic=16676
http://geekhack.org/index.php?topic=45396
http://geekhack.org/index.php?topic=53245
http://geekhack.org/index.php?topic=62302

One interesting thing for me about steno keyboards is that they require a substantially different type of hand/finger motion than a standard letter-by-letter keyboard. Instead of pressing each letter in turn by locating the keycap top and then pivoting the finger by flexing the metacarpal-phalangeal (MP) joint, you instead get several fingers ready above all the keys involved in a chord, slightly tense up those fingers so they form a rigid unit, and then perform the chord by moving the whole hand downward (the same as playing a chord on a piano). This is a substantially different motion, and I think it would be quite uncommon to suffer RSI on a steno keyboard (though existing RSI from other causes might still make typing on a steno machine difficult).

The big downside to chording systems, or other dictionary based systems like Proword’s, is that they require a pretty big time investment to really take full advantage of. The upside, of course, is that they can be amazingly efficient.

[Proword]

"The big downside to chording systems, or other dictionary based systems like Proword’s, is that they require a pretty big time investment to really take full advantage of"

Quite the opposite, at least with my own system.  Assuming that the operator can already touch type, then they are already being productive, and any additions  to the dictionary are based simply upon the operator's own view of what is needed ie a word or phrase is repeated often enough to warrant an abbreviation, so it's a "once only" creation, with no learning required.  On average to create an abbreviation in my system requires between  10-15 seconds.   

With steno there is a LOT to learn.



It involves learning somebody else's system, and learning it thoroughly, bearing in mind that shorthand of any sort, whether machine or manual, needed to be transcribable by a second party, possibly years after the writer has disappeared.  In using the word expansion system the abbreviation is only in existence for a short time, until the expansion key is executed.  I have over 7,000 abbreviations, some of which go back to the start of this century.  I think 15 seconds for 14 years is quite a good return on time invested.  Others, although they were created on a different machine and software, go back to the  very early '80s.  Nevertheless when I moved into the Windows environment, I was able to use the same abbreviations.

Of course, modern steno machines now  have the facility for creating a unique dictionary for each operator, that expands abbreviations in real time, which then means that that is the same as my system, which then returns us to chording, a la traditional stenotype, which, as the speaker in the video says, needs a specialised software that costs (from memory) $4,000,  whereas WordPerfect costs a hundred or so bucks AND does all the other things that good word processing software can do, such as tables, graphics, proofreading, typesetting and editing et al.   AND a computer comes standard with a keyboard, which can be used straight away.  She spoke of developing software.  Nothing  needs to be altered in WP, except the creation of a macro and some minor alterations to the functions, which can be swapped in or out, returning to the "standard" software in just a couple of mouse clicks. 

If necessary, any number of people can use the same computer, provided they have a unique log in, and create their own dictionary(s).  And of course, this bulletin board being about what it is about, it probably goes without saying that keyboards can be swapped immediately, so each operator can have it set up exactly as they like. This matches the "open source" criterion.

[jacobolus]

"The big downside to chording systems, or other dictionary based systems like Proword’s, is that they require a pretty big time investment to really take full advantage of"

Quite the opposite, at least with my own system.  Assuming that the operator can already touch type, then they are already being productive, and any additions  to the dictionary are based simply upon the operator's own view of what is needed ie a word or phrase is repeated often enough to warrant an abbreviation, so it's a "once only" creation, with no learning required.  On average to create an abbreviation in my system requires between  10-15 seconds.

Yeah, but someone’s not really close to “taking full advantage” of the system until they have many thousands of words defined, which is going to take a lot of time investment (and more importantly than time per se, a big investment of mental energy). Your system is limited as far as I can tell to WordPerfect itself (though something similar could probably be made to work systemwide, the way Plover does). So there’s a pretty big commitment involved: someone needs to anticipate using WordPerfect very seriously for lots of typing to recoup the costs of switching word processors and spending time and mental energy on creating new dictionary entries, and then is somewhat stuck with WordPerfect. That’s not going to help someone who for instance needs to write computer code instead of prose.

It’s true that your system has a big advantage over steno systems of being somewhat usable right out of the gate, so it’s not quite as big a commitment.

[...] a la traditional stenotype, which, as the speaker in the video says, needs a specialised software that costs (from memory) $4,000,  whereas WordPerfect costs a hundred or so bucks AND does all the other things that good word processing software can do, such as tables, graphics, proofreading, typesetting and editing et al.  AND a computer comes standard with a keyboard, which can be used straight away.

Well the whole point of Plover is to reduce the extra cost of using a steno system to effectively free. It just takes a keyboard with NKRO, and a software download/install. Then the system works system-wide, in whatever text editor or word processor you prefer (or in browser text boxes, etc.).

* * *

Anyway, I think your technique is ingenious and great! I’m just pointing out that it’s not totally trivial, and not 100% upside.

With the falling price of electronics, I’m sure a keyboard could be developed that stores a bunch of custom macros like yours directly on the keyboard, drops the time for creating new definitions down to 3–4 seconds, and works across operating systems and across programs. That would be the best of both worlds, seems to me.

(Though the actual use of chords instead of sequences of keys is a pretty interesting feature too IMO.)

[Proword]

"Your system is limited as far as I can tell to WordPerfect itself (though something similar could probably be made to work systemwide"

The reason(s) why it is restricted to WordPerfect is that, compared to MS Word, macros can be attached to almost ALL keystroke combinations in WP, and in early versions of MSW (at least before 2010), if the AutoCorrect file is bigger than 64kB this can result in "out of memory error".  Further the QuickCorrect and AutoCorrect function is executed by hitting the space bar, and various punctuation keys.  In WordPerfect this function can be moved elsewhere to another key, in MS Word I have not been able to do this.  Lastly, as I showed in my blog article, the WP keyboard map shows all available key stroke combos, whether connected or not, whereas MS Word only shows individual hot keys, and doesn't let the operator know which ones are available.  This makes the creation of meaningful hot keys almost impossible, at least quickly.


You seem to be overcomplicating the concept.  As for the "mental energy" expenditure, it just does not exist.   It is exactly the wrong thing to do to sit down and work out abbreviations.  It is done  as required.  There is a single question which the operator asks, having come across a word or phrase they think needs an abbreviation.   "What is a good abbreviation for this?"  Simply, there should not be any time spent "working out".  Take the first thing that comes into your mind, and put that on the keyboard and hit the execute key.  If nothing happens, there is your abbreviation  (I think I've probably had 10 duplicate abbreviations happen - ie two different phrases come up with the same abbreviation-  it is much more common to have two (or more) different abbreviations for the same full text.  This is not a problem.  It simply increases your chance of getting the right expansion later.)

(Bearing mind that I have 7,000 abbreviations, and I started creating them in 1982 or thereabouts, that averages out to  0.6 of an abbreviation every day.  Of course I don't work that way, because different topics need batches of abbreviations, so I might create 15 in a single session, but the same problem applies to steno.)

When you need to have a phrase expanded, it is simply a matter of asking exactly the same question, since you are working with the same mind (your own) you will almost invariably get the same answer.  That is ALL that is required.  But if you get the wrong abbreviation and nothing comes up, then simply make another abbreviation using the same  as you just thought might be the one.


I'm not sure that creating a programmable keyboard will make it any easier, since it is already down to a single mouse click to execute the creation function, the number of keystrokes for the abbreviation and the expanded text will be exactly the same, and a single mouse click will conclude the creation function.  In changing machines, as I used to do when I was a temp legal secretary, (the legal field still largely clings to WordPerfect because of its greatly superior functionality) I simply copy two classes of file, or three if you count the execute macro, but that can be written in a matter of seconds.  These files can be stored on a thumb drive, and unless I use a specialised keyboard (which I do), there's nothing to carry, and I could go straight into a workstation, copy the files and be functioning within a couple of minutes.

Your  point about "taking full advantage" is not close since the time investment, as I have said, is miniscule when compared to traditional steno, and from what I've seen, plover.


"That’s not going to help someone who for instance needs to write computer code instead of prose."

I'm not sure that is accurate, at least if the code is written on a program which has access to the MS clipboard, since the code can be written in WP, using all the shorthand functions and copied to the clipboard, then pasted.  Of course, if there is no clipboard, then that's a different kettle of fish, but we are getting into an extremely specialised area and a programmable keyboard may be the way to go, but that's not something I'm familiar with, so therefore I won't make any comment.

[jacobolus]

I think we’re talking past each-other.

You seem to be overcomplicating the concept.  As for the "mental energy" expenditure, it just does not exist.

If there were zero mental energy or up-front time/research/effort investment, then there would be thousands (or more like hundreds of thousands) of people using this system heavily. As far as I can tell there is at least one (you), and perhaps a handful of others.

That’s not at all to take away from the system you’ve invented/found/developed, which I think has a bunch of advantages and is very neat and elegant, but there is absolutely effort involved for someone to adopt your system.

I'm not sure that is accurate, at least if the code is written on a program which has access to the MS clipboard, since the code can be written in WP, using all the shorthand functions and copied to the clipboard, then pasted.

No, no one is going to do this. Existing text editors and IDEs have numerous very useful code authoring/editing tools which cannot be replicated in Word Perfect, and which are more important than shortcut expansion, per se.

Fortunately most serious text editors have similar kinds of mechanisms to the one you use from Word Perfect. Unfortunately, most developers only use them in a very limited fashion (similar to the way most WordPerfect users don’t use the shortcut-expansion mechanism you use very much or at all).

[Proword]

"If there were zero mental energy or up-front time/research/effort investment, then there would be thousands (or more like hundreds of thousands) of people using this system heavily. As far as I can tell there is at least one (you), and perhaps a handful of others."

People would rather con-form than per-form.  I've been putting this idea around  for many years now, even when I was doing my diploma of teaching, and just like when I talk about the Maltron keyboard, it's almost exactly the same answer - sorry, I'll stick with what I know!!!

It's like being in the public service.  Never do anything for the first time.
hahah

I'd be interested if you could give a detailed analysis of where you think there is time or energy being wasted/used.  I've never seen anything like that in all my years with it.  I didn't really develop it, it developed itself.  It has an air of inevitability.

[jacobolus]

People would rather con-form than per-form.  I've been putting this idea around  for many years now, even when I was doing my diploma of teaching, and just like when I talk about the Maltron keyboard, it's almost exactly the same answer - sorry, I'll stick with what I know!!! [...]

I'd be interested if you could give a detailed analysis of where you think there is time or energy being wasted/used.  I've never seen anything like that in all my years with it.  I didn't really develop it, it developed itself.  It has an air of inevitability.

I didn’t say the energy was being wasted! I just think there’s a big commitment up front. If you think about it, conformity is very low energy: it doesn’t take thinking, researching, or making decisions. If you just go along with the flow, you don’t have to worry nearly so much about the personal risks of what you’re doing, because those risks are broadly shared in the community.

Where the time and energy are in adopting your system is mostly in the decision to do it. It’s like any good habit (say, snacking on apples instead of potato chips): easy any individual time, but requiring a commitment to adopt and stick with.

In particular though, your system changes the type of thinking required for typing. Instead of just straight passively doing the job – putting the fingers on the letters and pressing them in the proper order – now typing requires active meta–thinking: not just “how do I always do this”, but now “what can I do now to best improve future efficiency”. I’m convinced that most people are deathly afraid of incorporating this kind of thinking into their routines, or at best indifferent about it. It’s the tool-maker’s thinking, not just the tool-user’s thinking.

[Tiramisuu]

From what I can see plover has the ability to provide real time expansion and a quite large  initial library has been provided to fast track the process.   I  would be  surprised if a user was not able to best their qwerty speed significantly in a year.   250wpm with high accuracy likely takes natural talent but I  would be perfectly happy capturing business requirements at 165.

[Proword]

"In particular though, your system changes the type of thinking required for typing. Instead of just straight passively doing the job – putting the fingers on the letters and pressing them in the proper order "

http://en.wikipedia.org/wiki/Touch_typing

Sorry, the type of thinking required is EXACTLY the same thinking as touchtyping.  An efficient typist DOESN'T type the three individual letters "C" "A"and "T", he or she types the entire word "cat", using muscle memory. 

"From what I can see plover has the ability to provide real time expansion and a quite large  initial library has been provided to fast track the process"

Yes, and it's that initial library where the initial "commitment" comes in.  The operator has to learn that library, which is based upon somebody else's way of thinking, whereas with the WP shorthand, there's just no commitment, because ALL library items are entirely the operator's choice, so there's no learning involved. 

My blog article consists of either background explanation or examples, so ignoring that, the essential steps are:

(1)  Create the following macro.

Application (WordPerfect; "WordPerfect"; Default!; "EN")
QuickCorrect (State: On!)
Type (Text: " ")
QuickCorrect (State: Off!)
DeleteCharPrevious ()

NOTE: The expression "EN" may vary, depending on which language/ country you are using.

(2) Create separate menu buttons for QuickCorrect and Keyboard to go on the menu bar. 

(3) Create a keyboard map with an appropriate name.

(4) Attach the macro to the desired keystroke, which will vary depending on the keyboard layout.

(5) Go into the QuickCorrect menu and uncheck the box which says replace words as you type.

(6) Commence typing.  As your work progresses you will decide whether a new abbreviation is required, whether caused by repetition or difficulty in typing accurately.

That's all you have to do.


The basic problem is delineated in the initial post. 

"Would love to learn and have installed the keyboard but not sure where to find good tutorials yet. "

There's nothing to learn and no tutorial required.

[jacobolus]

Sorry, the type of thinking required is EXACTLY the same thinking as touchtyping.  An efficient typist DOESN'T type the three individual letters "C" "A"and "T", he or she types the entire word "cat", using muscle memory. 

But what I mean is, every time I type “cat” I don’t have to think: “wait a minute, could I come up with some kind of abbreviation for ‘cat’ that would be slightly faster, enough so that when I amortize a 10-second up-front cost of defining a shortcut for it (plus some space in my brain to remember the shortcut and time spent later trying to remember whether there was a shortcut for ‘cat’ and what that shortcut might be) over the course of 20 years of typing ‘cat’ some number of times, it comes out to be a net positive?”

It’s not that going through such a thought process constantly for every word is hard, per se. But it takes strictly more mental energy than the default, which is to just type the word.

Once all the shortcuts have been defined and deeply memorized, I agree there’s a big efficiency bonus.

Does that make sense? I don’t think we really disagree all that much.

[Proword]

"plus some space in my brain to remember the shortcut and time spent later trying to remember whether there was a shortcut for ‘cat’ and what that shortcut might be"

That is where you and I appear to differ, in trying to "remember" the abbreviation.  There's no remembering involved, any more than remembering how to type "cat".   There's far more mental energy involved, for example, trying to differentiate homophones eg red/read, led/lead/, which/witch, canvas/canvass, no/know, be/bee etc.  If the operator can successfully negotiate those hurdles, keyboard shorthand should be a doddle.  If, on the other hand, the operator goes along thinking "I'll NEVER remember all those", then failure is a certainty.

I simply point to 30 plus years of success.

But yes, I'd agree with you entirely that packages like plover have a big upfront commitment, which is why back in 1979 I abandoned Pitman's pen writing when I got my first computer and saw its potential.

[Tiramisuu]

I am more than willing to believe that wordperfect expansions work.   Autohotkey could probably do something similar more portably but the lack of any community of interest to learn from or with makes it fairly unappealing.   

Steno is interesting and there are a small community of interested folk.   WordPerfect expansions would be a community of 2 with little hope of using the method across toolsets or operating systems without implementing it in a generic macro language.

To me this is a similar argument to that commonly made with alternate keyboard layouts.   Once you've picked one that doesn't suck, real value is gotten from learning it.   Picking one over the other doesn't provide a significant advantage once you have learned a technique. To me the fact that a prior body of knowledge exists with steno and that there is an OS and editor independent tool are pretty compelling.