CA2309868A1 - Cybernetic keyer for transmitting or entering symbols of a discrete alphabet into a device such as a wearable computer or portable information processor - Google Patents
Cybernetic keyer for transmitting or entering symbols of a discrete alphabet into a device such as a wearable computer or portable information processor Download PDFInfo
- Publication number
- CA2309868A1 CA2309868A1 CA 2309868 CA2309868A CA2309868A1 CA 2309868 A1 CA2309868 A1 CA 2309868A1 CA 2309868 CA2309868 CA 2309868 CA 2309868 A CA2309868 A CA 2309868A CA 2309868 A1 CA2309868 A1 CA 2309868A1
- Authority
- CA
- Canada
- Prior art keywords
- sensors
- keyer
- cybernetic
- output
- switch
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Landscapes
- Input From Keyboards Or The Like (AREA)
Abstract
A handheld keyer, suitable for sending code, transmitting messages, or entering data into a body-worn computer, or the like, is described. The keyer comprises a plurality of switches actuable in various combinations, such that, with a small number of switches, a large number of chords are possible, a chord being defined as a combination of closings and openings of one or more switches. Because of the numerous chords that can be generated, multiple chords can be assigned to the same symbol. Therefore, the choice of chord can be made based on a chord progression that makes data entry easier, more efficient, and less repetitious. The keyer, in the preferred embodiment, is operable while doing other things such as jogging, running up and down stairs, or the like. It can also be used to secretly type messages into a wearable computer, or to a remote entity, while standing and conversing with other people in a natural manner.
Description
iNT~.. , . . . . _;" f ;t'ilY4 ~r~ ~ a 200 Patent .Application of .. ~,, ~ ~~-~;~F~ i~vitLLtGTUFILE ' W. Steve G. Mann for CYBERNETIC KEYER FOR TRANSMITTING OR ENTERING
SYMBOLS OF A DISCRETE ALPHABET INTO A DEVICE SUCH AS
A WEARABLE COMPUTER OR PORTABLE INFORMATION
PROCESSOR
of which the following is a specification:
FIELD OF THE INVENTION
The present invention pertains generally to an apparatus for data entry, or for trans-mission of code, or the like. The field of the invention may be related to the fields of keyboards, keyers, input devices, human factors, mobile communication, cybernetic sciences, humanistic intelligence, and consumer electronics.
BACKGROUND OF THE INVENTION
Humanistic Intelligence is intelligence that arises in a natural cybernetic way, through having a constancy of user-interface, by way of an "always-ready"
computer system. A handheld keyer that can be easily custom built for each user, and can be used while doing other things such as jogging', running up and down stairs, or the like, is of great use in this context. Such a keyer c:an also be used to secretly type messages while standing and conversing with other people in a natural manner.
What is described; is a simple conformable keyboard, for use with computer sys-tems that vlork in extremely close synergy with the human user. This close synergy is achieved through a user-interface to signal processing hardware that is both in close physical proximity to the user, and is constant.
CA 02309868 2000-OS- .30 f ~~r~ ; - .~,.~~
PNO'r"~4_ ~ ~ ;,, ~ cLi~~ ('UELLE
The constancy of user-interface (interactional constancy) is what separates this signal processing architecture from other related devices such as pocket calculators and Personal Digital Assistants (PDAs).
Not only is the apparatus operationally constant, in the sense that although it.
nay have power saving (sleep) modes, it need not be completely shut down (dead as is typically a calculator worn in a shirt pocket but turned off roost of the time).
More important is the fact that it is also interactionally constant. By interaction-ally constant, what is meant is that the inputs and outputs of the device are always potentially active. Interactionally constant implies operationally constant, but op-erationally constant does not necessarily imply interactionally constant.
Thus, for example, a pocket calculator, worn in a shirt pocket, and left on all the time is still not interactionally constant, because it cannot be used in this state (e.g.
one still has to pull it out of the pocket to see the display or enter numbers). A wrist watch is a borderline case; although it operates constantly in order to continue to keep proper time, and it is conveniently worn on the body, one must make a conscious effort to orient it within one's field of vision in order to interact with it.
The invention can be applied to both traditional devices like calculators and v~rist watches, or to a new class of devices such as those that are interactionally constant.
The WearComp apparatus of the 1970s and early 1980s was an example of an interactionally constant wearable multimedia computer system for collaboration in computer mediated reality spaces.
Physical proximity and constancy were simultaneously realized by the 'WearComp' project. (For a detailed historical account of the WearCornp project, and other related projects, see http:~wvearcam.org and http:~~wearcomp.org~historical.) The W'earComp project of the 1970s and early 1980s comrpised Early embod-iments of the applicant's original "photographer's assistant" system and Personal Imaging systems. WearComp2, an early 1980s backpack-based signal processing and personal imaging system with right eye display comprised two antennas operating CA 02309868 2000-OS-30 ~ I[vj!'~~.; ~.v;w~~~: ~ c~~°,~~~~y N~A'~ 0 2000 z :s ~'HUt'r":. . ~ ., , ; ~s.a.i.~:l'UELLE
at different frequencies to facilitate wireless communications over a full-duplex radio link. WearComp4, a late 1980s clothing-based sigwal processing and personal imag-ing system had a left eye display and beam sputter. Separate antennas facilitated simultaneous voice, video, and data communication. W'earComp was an attempt at building an intelligent "photographer's assistant" around the body, and comprised a computer system attached to the body, a display means constantly visible to one or both eyes, and means of signal input including a series of pushbutton switches and a pointing device SUMMARY OF THE INVENTION
A preferred embodiment of the invention typically comprises an input device with pushbutton switches mounted to a wooden pushbroom hand-grip, or an input device comprising five microswitches mounted to the handle of an electronic flash. A
joystick (c:ontrolling two potentiometers), designed as a pointing device for use in conjunction with the ~~'earComp project, is also often present. The invention may also be used with various hand held devices, or may be mounted covertly to a belt, or in shoes, so that the user can key in data with the toes.
In the flashlamp embodiment, the user can hold the device in one hand to function as a keyboard and mouse do, but still be able to operate the device while walking around. In this way, the apparatus re-situated the functionality of a desktop multi-media computer with mouse, keyboard, and video screen, as a physical extension of the user's body.
An important aspect of the W'earComp is the keyer, which serves to enter com-wands into the apparatus. The keyer, in the preferred embodiment, is attached to some other apparatus, such as a flashlamp, or lightcornb, so that the effect is a hands free data entry device (hands free in the sense that one would need to hold onto the flashlamp, or the like, anyway, so no additional hand is needed to hold the keyer).
However, the keyer invention is useful by itself, or incorporated into various appliances m i t~t_tL ~ nAi.. wv~! ?rtn I r a MAY ~ ~ 2000 S
PROFrsa~ ~ :_ ._ ~~~al.E
such as toasters, kettles, pocket pagers, and the like, to allow a used to control the functionality of these devices quickly with a small number of switches, and without having to pay a lot of attention to a display system.
BRIEF DESCRIPTION OF THE DRAWINGS
The invention will now be described in more detail, by way of examples which in no way are meant to limit the scope of the invention, but, rather, these examples will serve to illustrate the invention with reference to the accompanying drawings, in which:
FIG. 1 illustrates hands free aspects of the keyer.
FIG. 2 Illustrates seven stages of a cybernetic keyer.
FIG. 3 Illustrates cybernetic keyer timing.
FIG. 4 Illustrates a two switch cybernetic keyer FIG. 5 Illustrates examples of some symbols output from a two switch cybernetic kever.
FIG. 6 Illustrates a timing graph for examples of some symbols and corresponding s«~itch timings.
FIG. 7 Illustrates keyer redundancy.
FIG. 8 Illustrates an ordinally conditional modifier example.
FIG. 9 Illustrates a. dual sensor.
FIG. 10 Illustrates a timing tolerance example.
FIG. 11 Illustrates traces of timing information for symbols of a two switch keyer incorporating timing tolerances.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
While the invention shall now be described with reference to the preferred em-bodiments shown in the drawings, it should be understood that the description is not CA 02309868 2000-OS-30 ...~.~,r !t~.I~FLI_.EC:T~3!~ ~=(~aFRTY
MAY ~ 0 2000 r,~
PROPHic i ~. ~, ~ ; ~:.~.tv i llF~ ~ F
to limit the invention only to the particular embodiments shown but"~"'~~'3~'f'' all alterations, modifications and equivalent arrangements possible within the scope of the appended claims.
In all aspects of the present invention, references to "sensor" mean any device comprising a device that can sense force, pressure, displacement, or the like, in either continuous or discrete steps.
References to "processor" , or "computer" shall include sequential instruction, par-allel instruction. and special purpose architectures such as digital signal processing hardware, Field Programmable Gate Arrays (FPGAs), programmable logic devices, Programmable Interface Controllers (PICs), as well as analog signal processing de-vices.
When it is said that object "A" is "borne" by object "B", this shall include the possibilities that A is attached to B, that A is bonded onto the surface of B, that A
is embedded inside B, that A is part of B, that A is built into B, or that A
is B.
V~ith reference to Fig 1; the keyer is preferably hands free. The hands free at-tribute comes from having the keyer borne by the handle of an object that 'would need to be carried by the user anyway. In this example, the instrument is a hand-held fiashlamp that is being used by the user for the painting with lightvectors ( "dusting" ) genre of photographic or experiential imaging. Since the flashlamp handle 120 must be held to direct the reflector 100 of lamp housing 110 at subject matter of interest, no additional hand is needed for the keyer. Alternatively the keyer may be built into the grip of a camera, if traditional photography were the goal. It could also be built into the grip of a ski pole, if, for example, a ski instructor was using the keyer to enter data, or to recall instructions into cybernetic Laser EyeTap (TM) eyeglasses as described in http://eyetap.org.
The original keyer had five keys, one for each of the four fingers, and a fifth one for the thumb; so that characters were formed by pressing the keys in different combinations. A computer or processor can read when each key is pressed, and when INTr~~ ..~ ,.., MAY .° (~ fQ00 PHU1'ri: e=~ 1..~ g each key is released, as well as how fast the key is depressed. A
satisfactor'~"~5'Pa~~'~8t' ' "'"' "
is the 6502 processor of Rockwell corporation. There are four large switches:
~ a thumb switch, SWt;
~ an index finger switch, SWi;
~ a middle finger switch, SWm;
~ a ring finger switch, S~~'r.
.-~ conditional modifier switch 190 has a long lever 199 that makes it. easy for the smallest finger to press it. In this embodiment, it is used less often than the other four switches.
A five switch keyer such as disclosed here, is called a pentakeyer. All five switches of this keyer are affixed to the handle 120 of the flashlarnp, which may be detached from the lamp housing 110 by way of a long screw through the entire handle from the bottom to the top, into a 1/4 20 screw thread in the housing 110. Housing 110 is made of metal to separate the 900 volts main supply in housing 110 from the handle and its associated low voltage switches.
The handle unscrews and is attached to any of a large number of different instru-menu, such as may be tapped with a 1/4 20 thread to accept the pentakeyer. The pentakeyer will screw onto the bottom of almost any camera, whether it be a 35mm still camera, a video camera, or the like. In this way the camera can be used while keying. Therefore, the result is an effective hands-free keyer. The fact that the keyer doubles as a handle for something makes it operationally hands free.
Lamp housing 110 has a cursor pointing device comrpised of housing 130 with four potentiometers 131, two of which are connected to a resistor capacitor tirrring circuit into a wearable computer for cursor control of the wearable computer. The control arm 140 is operable by the thumb of the user, so that the user can type, control the cursor, and aim the flashlamp with one hand.
Velocity sensing capability arises from using both the naturally closed (NC) and naturally open (NO) contacts, and measuring the time between when the common contact (C) leaves the NC contact and meets the NO contact. The velocity sensing timing circuit is similar to the potentiometer timing circuit, so that seven timing circuits in total are needed (five for the five switches, and two for the cursor, one for each of its x and y axes).
The operation of a keyer takes place over seven stages, as shown in Fig 2.
There are seven stages associated with pressing a combination of keys: A
Attack is the exact instant when the first switch is measurably pressed. (e.g. when its common moves away from its first throw if it is a double throw switch, or when the first switch is closed if it is a single throw switch). D Delay is the time betwTeen .Attack and when the last switch of a given chord has finished being pressed. Thus Delay corresponds to an arpeggiation interval (ARPA, from Old High German harpha, meaning harp, upon which strings were plucked in sequence but continued to sound together).
This Delay may be deliberate and expressive, or accidental. C Close is the exact instant at which the last. key of a desired chord is fully pressed. This Closure of the chord exists only in the mind (in the first brain) of the user, because the second brain (e.g. the computational apparatus, worn by; attached to, or implanted in the user) has no way of knowing whether there is a plan to, or plans to, continue the chord with more switch closures, unless all of the switches have been pressed. S
Sustain is the continued holding of the chord. Much as a piano has a sustain pedal, a chord on the keyboard can be sustained. Y Yaled is the opposite of delay (yaled is delay spelled backwards). Yaled is the time over which the user releases the components of a chord. Just as a piano is responsive to w hen keys are released, as well as when they are pressed, the keyboard can also be so responsive. The Yaled process is referred to as an APRA (OIGGEPRA), e.g. ARPA (or arpeggio) spelled backwards. O Open is the time at which the last key is fully (measurably) released. At this point the chord is completely open and no switches are measurably pressed.
It should be noted that the Close, Sustain, Release progression exists only in the firstbrain of the user, so any knowledge of the progression from within these three stages must be inferred, for example, by the time delays.
Arbitrary time constants can be used to make the keyboard very expressive, e.g.
characters can be formed by pressing keys for different lengths of time.
Indeed, a single key alone could be used to tap out l~Morse code, so that only one key would really be needed, if we were willing to use arbitrary timing information. Two keys would give the iambic paddle effect, similar to that described in a Jan. 12, '1972 publication, lay William F. Brown, U.S. Pat. No. 3757045, which was further developed in U.S.
Pat. No. 5773769, so that there would be no need for a heavy base (it could thus be further adapted to be used while worn).
Another example of time dependent keyboards include those with a Sustain fea tune, such as those with the well-known "Typematic'' (auto repeat) function of most modern keyboards. A key held down for a long time behaves differently than one pressed for a short time. The key held down for a short time produces a single char-acter, whereas the key held down for a. long time produces a plurality of the same character.
Some problems arise with time dependent keying, however. For example, a novice user typing very slowly may accidentally activates a timing feature. Although many input devices (e.g. ordinary keyboards and mice, as well as ordinary iambic paddles) have user adjustable timing constants, the need to adjust or adapt to these constants is an undesirable feature of these keyboards.
Moreover, there are problems and issues of velocity sensing, which is itself a timing matter. Sorne problems associated with velocity sensing include the necessity of selecting a switch with less deadband zone ( "snap" ) than desired, for the desirable amount of tactile feedback. There were some other undesirable attributes of the velocity sensing systems, so in this paper, the non-velocity sensing version will be described for simplicity.
Without using any timing constants whatsoever and without using any velocity sensing, nor Sustain, nor measurement of the timing in the Delay and Yaled stages), ( Fig 2) a very large number of possible keypresses can still be attained.
Consider first, for simplicity, a two key keyboard. There are four possible states:
00 when no keys are pressed, Ol when the least significant key (LSK) is pressed, 10 «-hen the most significant key (~MSK) is pressed, and 11, when both are pressed. It is desired to be able to have a rest position when no characters are sent, so 00 is preferably reserved for this rest state, otherwise the keyboard would be streaming out characters at all times, even when not in use.
In a dynamic situation, keys will be pressed and released. Both keys will be pressed at exactly the same time, only on a set of measure zero. In Fig 3, the time the LSK is pressed is plotted on the abscissa, and the time that the MSK is pressed on the ordinate, of a graph. Each keypress will be a point, and we will obtain a sc;atterplot of keypresses in the plane. Simultaneity exists along the line to = tl, a.nd the line has zero measure within the plane. Therefore, any symbol that requires simultaneous pressing of both keys (or simultaneous release of both) will be inherently unreliable, unless we build in some timing tolerance. Timing tolerances require tinning information, such as a timing constant or adaptation, so for no«-, for simplicity. let us assume that such timing tolerances are absent. Therefore, let us only concern ourselves with whether or not the key presses overlap, and if they do, let us only concern ourselves with 'which key was pressed first. and which was released first.
Two keys would be pressed or released at exactly the same time, only on a set, denoted by the line to = tl, which has measure zero in the (to, tl) plane, where to is the time of pressing or releasing of SWITCH 0, and tl is the tune of pressing or releasing of SWITCH 1. To overcome this uncertainty, the particular meaning of the chord is assigned based ordinally, rather than on using a timing threshold.
Here, for example, SWITCH 0 is pressed first and released after pressing SWITCH 1 but before releasing SWITCH 1. This situation is for one of the possible symbols that can be produced from this combination of two switches. This particular symbol will be numbered (4) and will be assigned the meaning of REW (Rewind).
This limitation greatly simplifies programming (e.g. for programming on a simple 602 microprocessor or the like), and greatly simplifies learning, as the pace from novice to expert does not involve continually changing timing parameters and various other subjectively determined timing constants.
Accordingly; without any timing constants or timing adaptation, v~e can, with only two switches, obtain six possible unique symbols, not including the Open chord (nothing pressed), as illustrated in Fig. 4.
In Fig. 4, timing information is depicted as dual traces: SWITCH 0 is depicted by the bottom trace and SWITCH 1 by the top trace. The zeroith symbol 00 depicts the open chord (no switches pressed). The first symbol O1 depicts the situation in which only SVG'ITCH 0 is pressed. The second symbol 10 depicts the situation in which only S~~- ITCH 1 is pressed. The third through sixth symbols 11 arise from situations in which both switches are pressed and then released, with overlap. The third symbol FLFL depicts the situation in which SWITCH 1 is pressed First, switch 0 is pressed Last, SWITCH 1 is released First, and switch 0 is released Last. Similarly LFLF
denotes Last First Last First (fourth symbol). FLLF denotes the situation in which SWITCH 1 is held down while SWITCH 0 is pressed and released (fifth symbol).
I:FFL denotes the situation in which SWITCH 0 is held down while SWITCH 1 is pressed and released (sixth symbol). The zeroith through sixth symbols are denoted by reference numerals 0 through 6, respectively. Each of the active ones (e.g.
other than the Open chord, 0) are given a meaning in operating a recording machine, with the functions PLAY, STOP, FastForward (FF), REWind, RECord, and PAUSE.
The operation of the cybernetic keyer is better understood by way of a simple example, illustrated in Fig 5. In this example, the top trace denotes SWITCH
1, and the bottom trace SWITCH 0. Initially, SWITCH 0 is pressed and then SWITCH 1 is pressed. However, because there is no overlap between these switch pressings, they are interpreted as separate symbols (e.g. this is not a chord). The separate symbols are 1 (PLAY) and 2 (STOP) respectively. This results in the playing of a short segment of video which is then stopped. Then, a little while later, SWITCH 0 is pressed and then SWITCH 1 is pressed. However; because there is now overlap, this action is considered to be a chord. Specifically it is an LFLF (Last First Last First) chord, which is interpreted as symbol number 4 (REW'IND). A REWIND operation on a stopped system is interpreted as high speed rewind. A short time later, SWITCH
is held down while SWITCH 1 is pressed briefly. This action is interpreted as symbol number 6 (PAUSE). Since PAUSE would normally be used only during PLAY or ILECORD, the meaning during RE~YIND is overloaded with a new meaning, namely slow down from high speed rewind to normal speed rewind. Thus we have full control of a recording system with only two switches, and without using any time constants a.s might arise from other interfaces such as the iambic ~~Iorse code keyers used by ham radio operators.
The timespace graph of Fig 3 is really just a four dimensional time space collapsed onto two dimensions of the page. Accordingly, we can view any combination of key presses that involves pressing both switches within a finite time, as a pair of ordered points on the graph. There are six possibilities. Examples of each are depicted in Fig 6. The symbol "X" denotes pressing of the two keys, and exists in the first pair of time dimensions, to and t1. The releasing of the two keys exists in the first second pair of dimensions, which, for simplicity (since it is difficult to draw the four dimensional space on the printed page), are also denoted to and tl, but with the symbol "O"
for Open. Examples of symbols 3 through 6 are realized. Two other examples, for when the switch closures do not overlap, are also depicted. These are depicted as 1.2 (symbol 1 followed by symbol 2) and 2,1 (symbol 2 followed by symbol 1).
With three switches instead of two, there are many more combinations possible.
Even if the three switches are not velocity sensing (e.g. if they are only single throw switches), there are still 51 combinations, which can be enumerated as follows:
~ Choose any one of the three switches (one symbol each) ~ Choose any pair of switches (e.g. omit any one of the three switches from the chord). For each of these three choices, there are four possible symbols (corresponding to the symbols 3 through 6 of Fig 4).
~ Using all three switches, at the ARPA (arpeggio, Fig 2) stage:
- there are three choices for First switch;
- once the first switch is chosen, there remains the question as to which of the remaining two will be pressed Next;
- then there is only one switch left, to press Last.
Thus at the ARPA stage, there are 3 * 2 * 1 = 6 different ways of pressing all three switches. At the APRA (oiggepra, Fig 2) stage, there are an equal number of ways of releasing these three switches that have all been pressed. Thus there are six ways of pressing, and six ways of releasing, which gives 6 * 6 = 36 symbols that involve all three switches.
Therefore, the total number of symbols on the three switch keyer is 3 + 12 +
36 = 51.
That's a sufficient number to generate the 26 letters of the alphabet, the numbers 0 through 9, the space character, and four additional symbols.
Uppercase and control characters are generated by using the four additional sym-bols for SHIFT, CONTROL. etc., of the letter or symbol that follows. Thus the multiplication sign is SHIFT followed by the number 8, and the at sign is SHIFT
followed by the number 2, and so on.
It is preferable to have all the characters be single chords, so that the user gets one character for every chord. Having a separate SHIFT chord would require the user to remember state (e.g. remember whether the SHIFT key was active), and would also slow down data entry.
Accordingly, if a fourth switch is added, we can:
~ Choose any one of the four switches (one symbol each);
~ Choose any pair of switches. For each of these z~44~z~, = 6 choices, there are four possible symbols (corresponding to the symbols 3 through 6 of Fig 4);
~ Choose any three switches (e.g. omit any one of the four switches from the chord). For each of these 4 choices, form the chord in any of the 3 * 2 * 1 =
possible ways, and unform the chord in any of six possible ways, giving 62 =
ways to create and uncreate the chord of the three chosen switches, as described in the three switch example above;
~ Using all four switches, at the ARPA (arpeggio) stage:
- there are four choices for First switch;
- once the first switch is chosen, there remains the question as to which of the remaining three switches will be pressed Second;
- once the second switch is chosen, there remains the question as to which of the remaining two switches will be pressed Third;
- then there is only one switch left, to press Last.
Thus at the ,~RPA stage, there are 4 * 3 * 2 * 1 = 4! = 24 different ways of pressing all four switches. At the APR.A (oiggepra) stage, there are an equal number of ways of releasing these four switches that: have all been pressed.
Thus there are twenty four ways of pressing, and twenty four ways of releasing, which gives 24 * 24 = 576 symbols that involve all four switches.
Therefore, the total number of symbols on the four switch keyer is 4~ (li)z + 4~ (2i)z + 4! (3~)z + 4~ ~4~)z 1!(4 _ 1)! 2!(4 _ 2)! 3!(4 _ 3)! 4!(4 _ 4)!
=4*lz+6*2z+4*6z+1*24z=748. (1) That's a sufficient number to generate the 256 ASCII symbols, along with 492 ad-ditional symbols which may be each assigned to entire words, or to commonly used phrases, such as a sig (signing off) message, a callsign, or commonly needed sequences of symbols. Thus a callsign like "N1NLF" is a. single chord. A commonly used se-<tuence of cozTZmands like ALT 192, ALT 255, ALT 192, is also a single chord.
Common words like ''the" , ''and" , etc., are also single chords.
The four switches can be, one each associated with the thumb, and three largest fingers, leaving out the srrzallest finger. Claude Shannon's information theory, how-ever, suggests that if we have a good strong clear channel, and a weaker channel, that we can get, additional error free communication by using both the strong and weak channels than we can by using only the strong channel. Therefore, we can and should use the weak (smallest) finger, for at least a small portion of the bandwidth, even though the other four will carry the majority of the load. Thus, referring to Fig l, we can see that. there are four strong double throw switches for the thumb and three largest fingers, and a fifth smaller switch having a. very long lever for the smallest finger. The long lever makes it easy to press this switch with the weak finger but at the expense of speed and response time. In fact, each of the five switches has been selected specifically knowing the strength and other attributes of what will press it.
This design gives rise to the pentakeyer.
The result in (1) can be generalized. The number of possible chords for a keyer with N switches, having only Single Throw (ST) switches, and not using any looping back at either the Delay or Yaled (Fig 2) stages of chord development, is:
_, A.
n! 2 2 n~(lV n)i ( ) ( ) TE= 1 Equation 2 simplifies to:
N!n!
), (3) (N-n .
Thus the pentakeyer gives us 5 + 40 + 360 + 2880 + 14400 = 17685 possible chords, without the use of any loopback, velocity sensing, or timing constants.
0.1 Redundancy The pentakeyer provides enough chords to use one to represent each of the most commonly used words in the English language. There are, for example, enough chords to represent more than half the words recognized by the UNIX ''spell" command with a typical /usr/share/lib/dict/words having 25143 words.
However, if all we want to represent is ASCII characters, the pentakeyer gives us 17685/256 > 69, e.g. more than 69 different ways to represent each letter.
This suggests that, for example. we can have 69 different ways of typing the letter ''a" , anc:l more than 69 different ways of typing the letter "b", and so on. In this way, we can choose whichever of these ways follows most conveniently in a given chord progression.
In using most musical instruments, there are multiple ways of generating each chord. For example, in playing the guitar, there are at least two commonly used "G"
chords, both of which sound quite similar. The choice of which ''G" to use depends on which one is easiest. to reach, based on what chord came before it, and what chord will come after it, etc.. Thus the freedom in having two different realizations of essentially the same chord makes playing the instrument easier.
Similarly, because there are so many different ways of typing the letter ''a"
, the user is free to select the particular realization of the letter "a" that's easiest to type when considering whatever came before it and whatever will come after it.
Having multiple realizations of the same chord is called chordic redundancy. Rather than distributing the chordic redundancy evenly across all letters, more redundancy is applied where it is needed more, so that there are more different ways of typing the letter ''a" than there are of typing the letter "q" or "u" . Part of this reasoning is based on the fact that there are a wide range of letters that can come before or after the letter ''a", whereas, for example, there is a smaller range of, and tighter distribution on, the letters that can follow "q", with the letter "u" being in the center of that rc>latively narrow distribution.
Redundancy need not be imposed on the novice, e.g. the first-time user can learn one way of forming each symbol, and then gradually learn a second way of forming some of the more commonly used symbols. Eventually; an experienced user will learn several ways of forming some of the more commonly used symbols.
Additionally, some chords are applied (in some cases even redundantly) to certain entire words, phrases, expressions, and the like. An example with timing diagrams fc~r a chordic redundancy based keyer is illustrated in Fig 7. Example of Keyer with a functional chordic redundancy generator or keyer having functional chordic redundancy. This keyer is used to type in or enter the numbers from 0 to 9 using three single throw switches. Each symbol (each number from 0 to 9) may be typed in various ways. Thus if we wish to type "001" , we can do this as follows:
~ first press and release switch SW'0, to obtain symbol Oo (the zeroith embodiment of symbol 0);
~ then to speed up the process (rather than press the same switch again) we press switch SWl while holding SW2, to obtain symbol O1 which is another realization of the symbol 0;
~ we then choose a realization of the symbol 1, namely 12, that does not.
begin with switch SV'2. Thus before the chord for symbol O1 is completely released (e.g. at the Yaled stage), we begin entering the chord for symbol 12, starting with the available switch S~VO.
This approach, of having multiple chords to choose from, in order to produce a given symbol, is the opposite of an approach taken with telephone touchpad style keyboards in which each number could mean different letters. In U.S. Pat. No.
6,011.554, issued January 4, 2000, assigned to Tegic Communications, Inc.
(Seattle.
W'A), King; Martin T. (Vashon, WA); Grower; Dale L. (Lansing, MI); Kushler;
Clifford A. (Vashon, WA); Grunbock; Cheryl A. (Vashon, WA) describe a disambiguating system in which an inference is made as to what the person might likely be trying to type. .4 drawback of this Tegic system is that the user must remain aware of what the machine thinks he or she is typing. There is an extra cognitive load imposed on the user, including the need to be constantly vigilant that errors are not being made. Using the Tegic system is a bit like using command line completion in Emacs.
~~'hile it allegedly purports to speed up the process, it can, in practice, slow down the process by imposing an additional burden on the user. In some sense, the Tegic system is a form of anti-redundancy, giving the user less flexibility. For example, forming new words (not in the dictionary) is quite difficult with the Tegic system, and when it does make mistakes, they are harder to detect because the mistakes get mapped onto the space of valid words.
Indeed, chordic redundancy (choice) is much more powerful than anti-redundancy (anti-choice) in how characters can be formed.
Ordinally conditional modifiers are a useful feature of some embodi-ments.
A modifier key is a key that alters the function of another key. On a standard key board, the SHIFT key modifies other letters by causing them to appear capitalized.
The SHIFT key modifies other keys between two states, namely a lowercase state and an uppercase state.
Another modifier key of the standard keyboard is the control key. A letter key pressed while the control key is held down is modified so that it becomes a control character. Thus the letter "a" gets changed to ''A" if it is pressed v~hile the control key is held down.
Fig. 8 illustrates an approach of having a modifier that is ordinally conditional, so that. its effect is responsive to where it is pressed in the chord. Fig 8 shows an example of how a four key ordinally conditional modifier is reduced to practice.
Letters are arranged in order of letter frequency starting with the letter "e"
which is the most commonly used letter of the alphabet.. Each of the 26 letters, the ten munbers, and some symbols are encoded with the 51 possible chords that can be formed from 3 switches, a middle finger switch, SW'm, an index finger switch, SWi, and a. thumb switch, SWt. (The more common letters such as e, t, a, etc., are also encoded redundantly so that there is more than one way to enter, for example, the letter "e".) A ring finger switch, SWr, is the ordinally conditional modifier.
If SWr is not pressed, an ordinary lowercase character is assumed. If a chord leads with SWr, the character is assumed to be an uppercase character. If the chord is entered while holding SWr the character is assumed to be a control character. If a chord trails with SV'r, it is assumed to represent a meta character. The ordinally conditional modifier is also applied to numbers to generate some of the symbols. For example, an exclamation mark is entered by leading with SWr, into the chord for the number 1.
One reason chording keyboards can be slow is that they often don't provide rollover. A regular ~W'ERTY... keyboard allows for rollover. A typical 1984 IBM
MODEL 1VI keyboard, for example, will be responsive to any key while the letter "q"
is held down. When the letters "q" and ''w" are held down, it is responsive to most keys (e.g. all those except keys in the q and w columns). When the letters "q", ''w" , and "e" are held down, it is responsive to other keys except from those three columns. When "q" , "w" , "e" , and "r" are held down, it is still responsive to keys in the right hand half of the keyboard (e.g. keys that would ordinarily be pressed with the right hand). Only when five keys are held down, does it stop responsing to new keypresses. Thus the MODEL M has quite a bit of rollover. This means that one can type new letters before finishing the typing of previous letters. This ability to have overlap between typing different letters allows a person to type faster because a new letter can be pressed before letting go of the previous letter.
Commercially available chording keyboards such as the Eandykey Twiddler and the BAT don't allow for rollover. Thus typing on the Twiddler or BAT is a slower process.
A goal of the cybernetic keyer is to be able to type much more quickly.
Therefore, an important feature, is the tradeoff between loopbacks at the Delay and Yaled stages (Fig 2), and rollover. If we decide, by design, that there will be no loopback at the Delay or the Yaled stages, we can assume that a chord has been committed to at the Release stage. Thus once we reach the Release stage, we can begin to accept another chord, so long as the other chord does not require the use of any switches that are still depressed at the Release stage. However, because of the sixty nine-fold chordic redundancy, it is arranged that, for most of the commonly following letters, there exists at least one new chord that can be built on keys not currently held down, at the Release stage.
As an example of rollover on a cybernetic keyer, consider the following: With reference to the two switch keyer, suppose we press SWITCH l, then press SWITCH 0, and then release SWITCH 1. W'e are now at the Release stage, and can enter a new command with SWITCH 1, since we know that there will be no Yaled loopback (e.g.
since we know that the chord will not involve pressing SWITCH 1 again). Thus pressing SWITCH 1 again can be safely used as a new symbol, prior to releasing SWITCH 0. In this way, symbols can rollover (overlap).
The chordic redundancy factor can be further increased with a more expressive keyer. The number of possible chords can be increased from 17685 to 7 + 84 +
1260 +
20160 + 302400 + +3628800 + 25401600 = 29354311 by simply adding three switches a.t the thumb position. This provides more than twenty nine million symbols, which is enough that each word in the English language could be represented in approximately a thousand different ways. This degree of chordic redundancy could provide for some very fast typing, if this many chords could be remembered. However, rather than increasing the number of switches, it is preferable to increase, instead, the expressivity of each one.
The pentakeyer is a crude instrument that lacks an ability to provide for tremen-dous "expression'' and sensitivity to the user. It fails to provide a rich form of Human-istic Intelligence in the sense that the feedback is cumbersome, and not continuous.
A guitar, violin, or real piano (e.g. an old fashioned mechanical piano, not a comput-erized synthetic piano data input keyboard), provides a much richer user experience because it provides instant feedback, and the user can experience the unique response of the instrument.
Even using both rails of each switch (e.g. double throw and even velocity sensing) still fails to provide a truly expressive input device. Accordingly, a better keyer was built from continuous transducers in the form of phonographic cartridges salvaged from old record players. These devices provide continuous flow of pressure information along two axes (each phono cartridge is responsive to pressure along two independent axes at right angles to one another, originally for playing stereo sound recordings from the grooves of a record). The dual sensor is depicted in Fig 9 and may comprise either a continuous sensor, or two on-off switches operable on separate axes.
In Fig. 9, Sensors SO and Sl may be switches or transducers or other forms of sensory apparatus. Sensors SO and S1 are operable individually, or together, by way of rocker Block BO1. Pressing straight down on Block BO1 will induce a response from both sensors. A response can also be induced in only one of sensors SO or S1 by pressing along its axis.
In this embodiment, the keyer described here is similar to the ternary data entry keyboard described by Langley; October 4, 1988, in U.S. Pat. 1\0. 4,775,255, in the sense that there are also to axes of each key; so that a given key can move towards or away from the operator, and has a central ''off" position, and a spring detent to make the key return to the central position in the absence of pressure from the finger. An important difference, though, is the fact that the keyboard of U.S.
Pat.
No. 4,775,255 has no ability to sense how fast, how hard, or how much each switch is pressed, other than just the ternary value of 0, +1, or -1. Also, in the keyer of U.S.
Pat. No. 4,775,255, the axes are not independent (e.g. one can't press +1 and -1 at the same time).
The sensor pair of Fig 9, on the other hand, provides two independent continuous dimensions. Thus a keyer rrrade from five such sensor pairs provides a much more expressive input. Even if the transducers are quantized, to binary outputs, there are four possibilities: 0, -l, +1, and ~1. W'ith five continuous transducers, one for each finger or thumb position, the user interface involves squeezing out characters, rather than clicking out characters. The input space is a very richly structured ten dimen-sional timespace, producing ten traces of time-varying signals. This ten dimensional timespace is reduced to discrete symbols according to a mapping based on compar-ison of the waveforms with reference waveforms stored in a computer system.
Since comparison with the reference waveforms is approximate, it is done by clustering, based on various kinds of similarity metrics.
Including just one time constant can greatly expand the functionality of a cybernetic keyer.
If we relax the ordinality constraint, and permit just one time constant, pertaining to simultaneity, we can obtain eleven symbols from just two switches. Such a scheme can be used for entering numbers, including the decimal point.
00 (Open chord not used) LFFL
WW
L': sing this simple coding scheme, the number zero is entered by pressing the LSK.
The number one is entered by pressing the V1SK. The number four, for example, is entered by pressing the MSK first, then pressing the LSK, and then releasing both at the same time, Within a certain time tolerance for which time is considered the same. (The letter ''W" denotes Within tolerance, as illustrated in With reference to Fig 10, Keyer Chords are shown within timing tolerances Y~ o and Wl. Time differences within the tolerance band are considered to be zero, so that events falling within the tolerance band defined by Wo and T~Yi are considered to be effectively simultaneous.
With reference to Fig 11, The decimal point is entered by pressing both switches at approximately the same tune, and releasing both at approximately the same time.
The ''approximately the same time" is defined according to a band around the line t« = tl of Fig 10.
Again, with reference to Fig. 11, a two switch chordic Keyer incorporating timing tolerances can produce eleven symbols. In addition to the unused Open chord, there are eleven other chords that can be used for the numbers 0 through 9, and the decimal point.
This number system can be implemented either by two pushbutton switches, or by a single vector keyswitch of two components, as illustrated in Fig 9. In the latter case, the entire set of number symbols can be entered with just one switch, e.g. by just one finger. :Vote that each number involves just a single keypress, unlike what would be the case if one entered numbers using a small wearable Vlorse code keyer, or the like. Thus the cybernetic chordic keyer provides a much more efficient entry of symbols.
Wearable keyers are known in the art of ham radio. For example, in L.S. Pat.
Vo.
4,194,085, March 18, 1980, Scelzi describes a "Finger keyer for code transmission''.
The telegraphic keyer fits over a finger, preferably the index finger, of an operator;
for tapping against the operator's thumb or any convenient object. It is used for transmission of code with portable equipment. The keyer is wearably operable when walking, or during other activities.
Keyers, such as previously known keyers, as well as the keyers of the present.
invention, such as the pentakeyer, and the continuous ten dimensional keying system, are much easier to use if they are custom made for each user. The most important aspect is getting the hand grip to fit well.
In a preferred embodiment, therefore, the keyer is moulded to fit the individual hand of the wearer. Presently, this is done by dipping the hand in icewater, and draping it with heated plastic material that is formed to the shape of the hand. Once the handpiece is formed, sensors are selected and installed so the keyer will match the specific attributes of the user's hand geometry.
Learning to use the pentakeyer is not easy, just as learning how to play a musical instrument is not easy. The pentakeyer evolved out of a difFerent philosophy, more than twenty years ago. This alternative philosophy knew nothing of so-called "user-friendly" user interface design, and therefore evolved along a completely different path.
Just as playing the violin is much harder to master than playing the TV remote control, it can also be much more rewarding and expressive. Thus if we were only to consider ease of use, vTe might be tempted to teach children how to operate a television because it is easier to learn than how to play a violin, or how to read and write. But if we did this, we would have an illiterate society in which all we could do were things that were easy to learn. It is the author's belief that a far richer experience can be attained with a lifelong computer interface that is worn on the body, and used constantly for ten to twenty years. On this kind of time scale, an apparatus that functions as a true extension of the mind and body, can result.
Just as it takes a long time to learn how to see, or to read and write, or to operate one's own body (e.g. it takes a number of years for the brain to figure out how to operate the body so that it can walk), it is expected that the most satisfying and powerful user interfaces will be learned over many years.
From the foregoing description, it will thus be evident. that the present invention provides a design for a personal cybernetic chordic keyer. .4s various changes can be made in the above embodiments and operating methods without departing from the spirit or scope of the invention, it is intended that all matter contained in the above description or show n in the accompanying drawings should be interpreted as illustrative and not in a limiting sense.
Variations or modifications to the design and construction of this invention, within the scope of the invention, may occur to those skilled in the art upon reviewing the disclosure herein. Such variations or modifications, if within the spirit of this invention, are intended to be encompassed within the scope of any claims to patent protection issuing upon this invention.
SYMBOLS OF A DISCRETE ALPHABET INTO A DEVICE SUCH AS
A WEARABLE COMPUTER OR PORTABLE INFORMATION
PROCESSOR
of which the following is a specification:
FIELD OF THE INVENTION
The present invention pertains generally to an apparatus for data entry, or for trans-mission of code, or the like. The field of the invention may be related to the fields of keyboards, keyers, input devices, human factors, mobile communication, cybernetic sciences, humanistic intelligence, and consumer electronics.
BACKGROUND OF THE INVENTION
Humanistic Intelligence is intelligence that arises in a natural cybernetic way, through having a constancy of user-interface, by way of an "always-ready"
computer system. A handheld keyer that can be easily custom built for each user, and can be used while doing other things such as jogging', running up and down stairs, or the like, is of great use in this context. Such a keyer c:an also be used to secretly type messages while standing and conversing with other people in a natural manner.
What is described; is a simple conformable keyboard, for use with computer sys-tems that vlork in extremely close synergy with the human user. This close synergy is achieved through a user-interface to signal processing hardware that is both in close physical proximity to the user, and is constant.
CA 02309868 2000-OS- .30 f ~~r~ ; - .~,.~~
PNO'r"~4_ ~ ~ ;,, ~ cLi~~ ('UELLE
The constancy of user-interface (interactional constancy) is what separates this signal processing architecture from other related devices such as pocket calculators and Personal Digital Assistants (PDAs).
Not only is the apparatus operationally constant, in the sense that although it.
nay have power saving (sleep) modes, it need not be completely shut down (dead as is typically a calculator worn in a shirt pocket but turned off roost of the time).
More important is the fact that it is also interactionally constant. By interaction-ally constant, what is meant is that the inputs and outputs of the device are always potentially active. Interactionally constant implies operationally constant, but op-erationally constant does not necessarily imply interactionally constant.
Thus, for example, a pocket calculator, worn in a shirt pocket, and left on all the time is still not interactionally constant, because it cannot be used in this state (e.g.
one still has to pull it out of the pocket to see the display or enter numbers). A wrist watch is a borderline case; although it operates constantly in order to continue to keep proper time, and it is conveniently worn on the body, one must make a conscious effort to orient it within one's field of vision in order to interact with it.
The invention can be applied to both traditional devices like calculators and v~rist watches, or to a new class of devices such as those that are interactionally constant.
The WearComp apparatus of the 1970s and early 1980s was an example of an interactionally constant wearable multimedia computer system for collaboration in computer mediated reality spaces.
Physical proximity and constancy were simultaneously realized by the 'WearComp' project. (For a detailed historical account of the WearCornp project, and other related projects, see http:~wvearcam.org and http:~~wearcomp.org~historical.) The W'earComp project of the 1970s and early 1980s comrpised Early embod-iments of the applicant's original "photographer's assistant" system and Personal Imaging systems. WearComp2, an early 1980s backpack-based signal processing and personal imaging system with right eye display comprised two antennas operating CA 02309868 2000-OS-30 ~ I[vj!'~~.; ~.v;w~~~: ~ c~~°,~~~~y N~A'~ 0 2000 z :s ~'HUt'r":. . ~ ., , ; ~s.a.i.~:l'UELLE
at different frequencies to facilitate wireless communications over a full-duplex radio link. WearComp4, a late 1980s clothing-based sigwal processing and personal imag-ing system had a left eye display and beam sputter. Separate antennas facilitated simultaneous voice, video, and data communication. W'earComp was an attempt at building an intelligent "photographer's assistant" around the body, and comprised a computer system attached to the body, a display means constantly visible to one or both eyes, and means of signal input including a series of pushbutton switches and a pointing device SUMMARY OF THE INVENTION
A preferred embodiment of the invention typically comprises an input device with pushbutton switches mounted to a wooden pushbroom hand-grip, or an input device comprising five microswitches mounted to the handle of an electronic flash. A
joystick (c:ontrolling two potentiometers), designed as a pointing device for use in conjunction with the ~~'earComp project, is also often present. The invention may also be used with various hand held devices, or may be mounted covertly to a belt, or in shoes, so that the user can key in data with the toes.
In the flashlamp embodiment, the user can hold the device in one hand to function as a keyboard and mouse do, but still be able to operate the device while walking around. In this way, the apparatus re-situated the functionality of a desktop multi-media computer with mouse, keyboard, and video screen, as a physical extension of the user's body.
An important aspect of the W'earComp is the keyer, which serves to enter com-wands into the apparatus. The keyer, in the preferred embodiment, is attached to some other apparatus, such as a flashlamp, or lightcornb, so that the effect is a hands free data entry device (hands free in the sense that one would need to hold onto the flashlamp, or the like, anyway, so no additional hand is needed to hold the keyer).
However, the keyer invention is useful by itself, or incorporated into various appliances m i t~t_tL ~ nAi.. wv~! ?rtn I r a MAY ~ ~ 2000 S
PROFrsa~ ~ :_ ._ ~~~al.E
such as toasters, kettles, pocket pagers, and the like, to allow a used to control the functionality of these devices quickly with a small number of switches, and without having to pay a lot of attention to a display system.
BRIEF DESCRIPTION OF THE DRAWINGS
The invention will now be described in more detail, by way of examples which in no way are meant to limit the scope of the invention, but, rather, these examples will serve to illustrate the invention with reference to the accompanying drawings, in which:
FIG. 1 illustrates hands free aspects of the keyer.
FIG. 2 Illustrates seven stages of a cybernetic keyer.
FIG. 3 Illustrates cybernetic keyer timing.
FIG. 4 Illustrates a two switch cybernetic keyer FIG. 5 Illustrates examples of some symbols output from a two switch cybernetic kever.
FIG. 6 Illustrates a timing graph for examples of some symbols and corresponding s«~itch timings.
FIG. 7 Illustrates keyer redundancy.
FIG. 8 Illustrates an ordinally conditional modifier example.
FIG. 9 Illustrates a. dual sensor.
FIG. 10 Illustrates a timing tolerance example.
FIG. 11 Illustrates traces of timing information for symbols of a two switch keyer incorporating timing tolerances.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
While the invention shall now be described with reference to the preferred em-bodiments shown in the drawings, it should be understood that the description is not CA 02309868 2000-OS-30 ...~.~,r !t~.I~FLI_.EC:T~3!~ ~=(~aFRTY
MAY ~ 0 2000 r,~
PROPHic i ~. ~, ~ ; ~:.~.tv i llF~ ~ F
to limit the invention only to the particular embodiments shown but"~"'~~'3~'f'' all alterations, modifications and equivalent arrangements possible within the scope of the appended claims.
In all aspects of the present invention, references to "sensor" mean any device comprising a device that can sense force, pressure, displacement, or the like, in either continuous or discrete steps.
References to "processor" , or "computer" shall include sequential instruction, par-allel instruction. and special purpose architectures such as digital signal processing hardware, Field Programmable Gate Arrays (FPGAs), programmable logic devices, Programmable Interface Controllers (PICs), as well as analog signal processing de-vices.
When it is said that object "A" is "borne" by object "B", this shall include the possibilities that A is attached to B, that A is bonded onto the surface of B, that A
is embedded inside B, that A is part of B, that A is built into B, or that A
is B.
V~ith reference to Fig 1; the keyer is preferably hands free. The hands free at-tribute comes from having the keyer borne by the handle of an object that 'would need to be carried by the user anyway. In this example, the instrument is a hand-held fiashlamp that is being used by the user for the painting with lightvectors ( "dusting" ) genre of photographic or experiential imaging. Since the flashlamp handle 120 must be held to direct the reflector 100 of lamp housing 110 at subject matter of interest, no additional hand is needed for the keyer. Alternatively the keyer may be built into the grip of a camera, if traditional photography were the goal. It could also be built into the grip of a ski pole, if, for example, a ski instructor was using the keyer to enter data, or to recall instructions into cybernetic Laser EyeTap (TM) eyeglasses as described in http://eyetap.org.
The original keyer had five keys, one for each of the four fingers, and a fifth one for the thumb; so that characters were formed by pressing the keys in different combinations. A computer or processor can read when each key is pressed, and when INTr~~ ..~ ,.., MAY .° (~ fQ00 PHU1'ri: e=~ 1..~ g each key is released, as well as how fast the key is depressed. A
satisfactor'~"~5'Pa~~'~8t' ' "'"' "
is the 6502 processor of Rockwell corporation. There are four large switches:
~ a thumb switch, SWt;
~ an index finger switch, SWi;
~ a middle finger switch, SWm;
~ a ring finger switch, S~~'r.
.-~ conditional modifier switch 190 has a long lever 199 that makes it. easy for the smallest finger to press it. In this embodiment, it is used less often than the other four switches.
A five switch keyer such as disclosed here, is called a pentakeyer. All five switches of this keyer are affixed to the handle 120 of the flashlarnp, which may be detached from the lamp housing 110 by way of a long screw through the entire handle from the bottom to the top, into a 1/4 20 screw thread in the housing 110. Housing 110 is made of metal to separate the 900 volts main supply in housing 110 from the handle and its associated low voltage switches.
The handle unscrews and is attached to any of a large number of different instru-menu, such as may be tapped with a 1/4 20 thread to accept the pentakeyer. The pentakeyer will screw onto the bottom of almost any camera, whether it be a 35mm still camera, a video camera, or the like. In this way the camera can be used while keying. Therefore, the result is an effective hands-free keyer. The fact that the keyer doubles as a handle for something makes it operationally hands free.
Lamp housing 110 has a cursor pointing device comrpised of housing 130 with four potentiometers 131, two of which are connected to a resistor capacitor tirrring circuit into a wearable computer for cursor control of the wearable computer. The control arm 140 is operable by the thumb of the user, so that the user can type, control the cursor, and aim the flashlamp with one hand.
Velocity sensing capability arises from using both the naturally closed (NC) and naturally open (NO) contacts, and measuring the time between when the common contact (C) leaves the NC contact and meets the NO contact. The velocity sensing timing circuit is similar to the potentiometer timing circuit, so that seven timing circuits in total are needed (five for the five switches, and two for the cursor, one for each of its x and y axes).
The operation of a keyer takes place over seven stages, as shown in Fig 2.
There are seven stages associated with pressing a combination of keys: A
Attack is the exact instant when the first switch is measurably pressed. (e.g. when its common moves away from its first throw if it is a double throw switch, or when the first switch is closed if it is a single throw switch). D Delay is the time betwTeen .Attack and when the last switch of a given chord has finished being pressed. Thus Delay corresponds to an arpeggiation interval (ARPA, from Old High German harpha, meaning harp, upon which strings were plucked in sequence but continued to sound together).
This Delay may be deliberate and expressive, or accidental. C Close is the exact instant at which the last. key of a desired chord is fully pressed. This Closure of the chord exists only in the mind (in the first brain) of the user, because the second brain (e.g. the computational apparatus, worn by; attached to, or implanted in the user) has no way of knowing whether there is a plan to, or plans to, continue the chord with more switch closures, unless all of the switches have been pressed. S
Sustain is the continued holding of the chord. Much as a piano has a sustain pedal, a chord on the keyboard can be sustained. Y Yaled is the opposite of delay (yaled is delay spelled backwards). Yaled is the time over which the user releases the components of a chord. Just as a piano is responsive to w hen keys are released, as well as when they are pressed, the keyboard can also be so responsive. The Yaled process is referred to as an APRA (OIGGEPRA), e.g. ARPA (or arpeggio) spelled backwards. O Open is the time at which the last key is fully (measurably) released. At this point the chord is completely open and no switches are measurably pressed.
It should be noted that the Close, Sustain, Release progression exists only in the firstbrain of the user, so any knowledge of the progression from within these three stages must be inferred, for example, by the time delays.
Arbitrary time constants can be used to make the keyboard very expressive, e.g.
characters can be formed by pressing keys for different lengths of time.
Indeed, a single key alone could be used to tap out l~Morse code, so that only one key would really be needed, if we were willing to use arbitrary timing information. Two keys would give the iambic paddle effect, similar to that described in a Jan. 12, '1972 publication, lay William F. Brown, U.S. Pat. No. 3757045, which was further developed in U.S.
Pat. No. 5773769, so that there would be no need for a heavy base (it could thus be further adapted to be used while worn).
Another example of time dependent keyboards include those with a Sustain fea tune, such as those with the well-known "Typematic'' (auto repeat) function of most modern keyboards. A key held down for a long time behaves differently than one pressed for a short time. The key held down for a short time produces a single char-acter, whereas the key held down for a. long time produces a plurality of the same character.
Some problems arise with time dependent keying, however. For example, a novice user typing very slowly may accidentally activates a timing feature. Although many input devices (e.g. ordinary keyboards and mice, as well as ordinary iambic paddles) have user adjustable timing constants, the need to adjust or adapt to these constants is an undesirable feature of these keyboards.
Moreover, there are problems and issues of velocity sensing, which is itself a timing matter. Sorne problems associated with velocity sensing include the necessity of selecting a switch with less deadband zone ( "snap" ) than desired, for the desirable amount of tactile feedback. There were some other undesirable attributes of the velocity sensing systems, so in this paper, the non-velocity sensing version will be described for simplicity.
Without using any timing constants whatsoever and without using any velocity sensing, nor Sustain, nor measurement of the timing in the Delay and Yaled stages), ( Fig 2) a very large number of possible keypresses can still be attained.
Consider first, for simplicity, a two key keyboard. There are four possible states:
00 when no keys are pressed, Ol when the least significant key (LSK) is pressed, 10 «-hen the most significant key (~MSK) is pressed, and 11, when both are pressed. It is desired to be able to have a rest position when no characters are sent, so 00 is preferably reserved for this rest state, otherwise the keyboard would be streaming out characters at all times, even when not in use.
In a dynamic situation, keys will be pressed and released. Both keys will be pressed at exactly the same time, only on a set of measure zero. In Fig 3, the time the LSK is pressed is plotted on the abscissa, and the time that the MSK is pressed on the ordinate, of a graph. Each keypress will be a point, and we will obtain a sc;atterplot of keypresses in the plane. Simultaneity exists along the line to = tl, a.nd the line has zero measure within the plane. Therefore, any symbol that requires simultaneous pressing of both keys (or simultaneous release of both) will be inherently unreliable, unless we build in some timing tolerance. Timing tolerances require tinning information, such as a timing constant or adaptation, so for no«-, for simplicity. let us assume that such timing tolerances are absent. Therefore, let us only concern ourselves with whether or not the key presses overlap, and if they do, let us only concern ourselves with 'which key was pressed first. and which was released first.
Two keys would be pressed or released at exactly the same time, only on a set, denoted by the line to = tl, which has measure zero in the (to, tl) plane, where to is the time of pressing or releasing of SWITCH 0, and tl is the tune of pressing or releasing of SWITCH 1. To overcome this uncertainty, the particular meaning of the chord is assigned based ordinally, rather than on using a timing threshold.
Here, for example, SWITCH 0 is pressed first and released after pressing SWITCH 1 but before releasing SWITCH 1. This situation is for one of the possible symbols that can be produced from this combination of two switches. This particular symbol will be numbered (4) and will be assigned the meaning of REW (Rewind).
This limitation greatly simplifies programming (e.g. for programming on a simple 602 microprocessor or the like), and greatly simplifies learning, as the pace from novice to expert does not involve continually changing timing parameters and various other subjectively determined timing constants.
Accordingly; without any timing constants or timing adaptation, v~e can, with only two switches, obtain six possible unique symbols, not including the Open chord (nothing pressed), as illustrated in Fig. 4.
In Fig. 4, timing information is depicted as dual traces: SWITCH 0 is depicted by the bottom trace and SWITCH 1 by the top trace. The zeroith symbol 00 depicts the open chord (no switches pressed). The first symbol O1 depicts the situation in which only SVG'ITCH 0 is pressed. The second symbol 10 depicts the situation in which only S~~- ITCH 1 is pressed. The third through sixth symbols 11 arise from situations in which both switches are pressed and then released, with overlap. The third symbol FLFL depicts the situation in which SWITCH 1 is pressed First, switch 0 is pressed Last, SWITCH 1 is released First, and switch 0 is released Last. Similarly LFLF
denotes Last First Last First (fourth symbol). FLLF denotes the situation in which SWITCH 1 is held down while SWITCH 0 is pressed and released (fifth symbol).
I:FFL denotes the situation in which SWITCH 0 is held down while SWITCH 1 is pressed and released (sixth symbol). The zeroith through sixth symbols are denoted by reference numerals 0 through 6, respectively. Each of the active ones (e.g.
other than the Open chord, 0) are given a meaning in operating a recording machine, with the functions PLAY, STOP, FastForward (FF), REWind, RECord, and PAUSE.
The operation of the cybernetic keyer is better understood by way of a simple example, illustrated in Fig 5. In this example, the top trace denotes SWITCH
1, and the bottom trace SWITCH 0. Initially, SWITCH 0 is pressed and then SWITCH 1 is pressed. However, because there is no overlap between these switch pressings, they are interpreted as separate symbols (e.g. this is not a chord). The separate symbols are 1 (PLAY) and 2 (STOP) respectively. This results in the playing of a short segment of video which is then stopped. Then, a little while later, SWITCH 0 is pressed and then SWITCH 1 is pressed. However; because there is now overlap, this action is considered to be a chord. Specifically it is an LFLF (Last First Last First) chord, which is interpreted as symbol number 4 (REW'IND). A REWIND operation on a stopped system is interpreted as high speed rewind. A short time later, SWITCH
is held down while SWITCH 1 is pressed briefly. This action is interpreted as symbol number 6 (PAUSE). Since PAUSE would normally be used only during PLAY or ILECORD, the meaning during RE~YIND is overloaded with a new meaning, namely slow down from high speed rewind to normal speed rewind. Thus we have full control of a recording system with only two switches, and without using any time constants a.s might arise from other interfaces such as the iambic ~~Iorse code keyers used by ham radio operators.
The timespace graph of Fig 3 is really just a four dimensional time space collapsed onto two dimensions of the page. Accordingly, we can view any combination of key presses that involves pressing both switches within a finite time, as a pair of ordered points on the graph. There are six possibilities. Examples of each are depicted in Fig 6. The symbol "X" denotes pressing of the two keys, and exists in the first pair of time dimensions, to and t1. The releasing of the two keys exists in the first second pair of dimensions, which, for simplicity (since it is difficult to draw the four dimensional space on the printed page), are also denoted to and tl, but with the symbol "O"
for Open. Examples of symbols 3 through 6 are realized. Two other examples, for when the switch closures do not overlap, are also depicted. These are depicted as 1.2 (symbol 1 followed by symbol 2) and 2,1 (symbol 2 followed by symbol 1).
With three switches instead of two, there are many more combinations possible.
Even if the three switches are not velocity sensing (e.g. if they are only single throw switches), there are still 51 combinations, which can be enumerated as follows:
~ Choose any one of the three switches (one symbol each) ~ Choose any pair of switches (e.g. omit any one of the three switches from the chord). For each of these three choices, there are four possible symbols (corresponding to the symbols 3 through 6 of Fig 4).
~ Using all three switches, at the ARPA (arpeggio, Fig 2) stage:
- there are three choices for First switch;
- once the first switch is chosen, there remains the question as to which of the remaining two will be pressed Next;
- then there is only one switch left, to press Last.
Thus at the ARPA stage, there are 3 * 2 * 1 = 6 different ways of pressing all three switches. At the APRA (oiggepra, Fig 2) stage, there are an equal number of ways of releasing these three switches that have all been pressed. Thus there are six ways of pressing, and six ways of releasing, which gives 6 * 6 = 36 symbols that involve all three switches.
Therefore, the total number of symbols on the three switch keyer is 3 + 12 +
36 = 51.
That's a sufficient number to generate the 26 letters of the alphabet, the numbers 0 through 9, the space character, and four additional symbols.
Uppercase and control characters are generated by using the four additional sym-bols for SHIFT, CONTROL. etc., of the letter or symbol that follows. Thus the multiplication sign is SHIFT followed by the number 8, and the at sign is SHIFT
followed by the number 2, and so on.
It is preferable to have all the characters be single chords, so that the user gets one character for every chord. Having a separate SHIFT chord would require the user to remember state (e.g. remember whether the SHIFT key was active), and would also slow down data entry.
Accordingly, if a fourth switch is added, we can:
~ Choose any one of the four switches (one symbol each);
~ Choose any pair of switches. For each of these z~44~z~, = 6 choices, there are four possible symbols (corresponding to the symbols 3 through 6 of Fig 4);
~ Choose any three switches (e.g. omit any one of the four switches from the chord). For each of these 4 choices, form the chord in any of the 3 * 2 * 1 =
possible ways, and unform the chord in any of six possible ways, giving 62 =
ways to create and uncreate the chord of the three chosen switches, as described in the three switch example above;
~ Using all four switches, at the ARPA (arpeggio) stage:
- there are four choices for First switch;
- once the first switch is chosen, there remains the question as to which of the remaining three switches will be pressed Second;
- once the second switch is chosen, there remains the question as to which of the remaining two switches will be pressed Third;
- then there is only one switch left, to press Last.
Thus at the ,~RPA stage, there are 4 * 3 * 2 * 1 = 4! = 24 different ways of pressing all four switches. At the APR.A (oiggepra) stage, there are an equal number of ways of releasing these four switches that: have all been pressed.
Thus there are twenty four ways of pressing, and twenty four ways of releasing, which gives 24 * 24 = 576 symbols that involve all four switches.
Therefore, the total number of symbols on the four switch keyer is 4~ (li)z + 4~ (2i)z + 4! (3~)z + 4~ ~4~)z 1!(4 _ 1)! 2!(4 _ 2)! 3!(4 _ 3)! 4!(4 _ 4)!
=4*lz+6*2z+4*6z+1*24z=748. (1) That's a sufficient number to generate the 256 ASCII symbols, along with 492 ad-ditional symbols which may be each assigned to entire words, or to commonly used phrases, such as a sig (signing off) message, a callsign, or commonly needed sequences of symbols. Thus a callsign like "N1NLF" is a. single chord. A commonly used se-<tuence of cozTZmands like ALT 192, ALT 255, ALT 192, is also a single chord.
Common words like ''the" , ''and" , etc., are also single chords.
The four switches can be, one each associated with the thumb, and three largest fingers, leaving out the srrzallest finger. Claude Shannon's information theory, how-ever, suggests that if we have a good strong clear channel, and a weaker channel, that we can get, additional error free communication by using both the strong and weak channels than we can by using only the strong channel. Therefore, we can and should use the weak (smallest) finger, for at least a small portion of the bandwidth, even though the other four will carry the majority of the load. Thus, referring to Fig l, we can see that. there are four strong double throw switches for the thumb and three largest fingers, and a fifth smaller switch having a. very long lever for the smallest finger. The long lever makes it easy to press this switch with the weak finger but at the expense of speed and response time. In fact, each of the five switches has been selected specifically knowing the strength and other attributes of what will press it.
This design gives rise to the pentakeyer.
The result in (1) can be generalized. The number of possible chords for a keyer with N switches, having only Single Throw (ST) switches, and not using any looping back at either the Delay or Yaled (Fig 2) stages of chord development, is:
_, A.
n! 2 2 n~(lV n)i ( ) ( ) TE= 1 Equation 2 simplifies to:
N!n!
), (3) (N-n .
Thus the pentakeyer gives us 5 + 40 + 360 + 2880 + 14400 = 17685 possible chords, without the use of any loopback, velocity sensing, or timing constants.
0.1 Redundancy The pentakeyer provides enough chords to use one to represent each of the most commonly used words in the English language. There are, for example, enough chords to represent more than half the words recognized by the UNIX ''spell" command with a typical /usr/share/lib/dict/words having 25143 words.
However, if all we want to represent is ASCII characters, the pentakeyer gives us 17685/256 > 69, e.g. more than 69 different ways to represent each letter.
This suggests that, for example. we can have 69 different ways of typing the letter ''a" , anc:l more than 69 different ways of typing the letter "b", and so on. In this way, we can choose whichever of these ways follows most conveniently in a given chord progression.
In using most musical instruments, there are multiple ways of generating each chord. For example, in playing the guitar, there are at least two commonly used "G"
chords, both of which sound quite similar. The choice of which ''G" to use depends on which one is easiest. to reach, based on what chord came before it, and what chord will come after it, etc.. Thus the freedom in having two different realizations of essentially the same chord makes playing the instrument easier.
Similarly, because there are so many different ways of typing the letter ''a"
, the user is free to select the particular realization of the letter "a" that's easiest to type when considering whatever came before it and whatever will come after it.
Having multiple realizations of the same chord is called chordic redundancy. Rather than distributing the chordic redundancy evenly across all letters, more redundancy is applied where it is needed more, so that there are more different ways of typing the letter ''a" than there are of typing the letter "q" or "u" . Part of this reasoning is based on the fact that there are a wide range of letters that can come before or after the letter ''a", whereas, for example, there is a smaller range of, and tighter distribution on, the letters that can follow "q", with the letter "u" being in the center of that rc>latively narrow distribution.
Redundancy need not be imposed on the novice, e.g. the first-time user can learn one way of forming each symbol, and then gradually learn a second way of forming some of the more commonly used symbols. Eventually; an experienced user will learn several ways of forming some of the more commonly used symbols.
Additionally, some chords are applied (in some cases even redundantly) to certain entire words, phrases, expressions, and the like. An example with timing diagrams fc~r a chordic redundancy based keyer is illustrated in Fig 7. Example of Keyer with a functional chordic redundancy generator or keyer having functional chordic redundancy. This keyer is used to type in or enter the numbers from 0 to 9 using three single throw switches. Each symbol (each number from 0 to 9) may be typed in various ways. Thus if we wish to type "001" , we can do this as follows:
~ first press and release switch SW'0, to obtain symbol Oo (the zeroith embodiment of symbol 0);
~ then to speed up the process (rather than press the same switch again) we press switch SWl while holding SW2, to obtain symbol O1 which is another realization of the symbol 0;
~ we then choose a realization of the symbol 1, namely 12, that does not.
begin with switch SV'2. Thus before the chord for symbol O1 is completely released (e.g. at the Yaled stage), we begin entering the chord for symbol 12, starting with the available switch S~VO.
This approach, of having multiple chords to choose from, in order to produce a given symbol, is the opposite of an approach taken with telephone touchpad style keyboards in which each number could mean different letters. In U.S. Pat. No.
6,011.554, issued January 4, 2000, assigned to Tegic Communications, Inc.
(Seattle.
W'A), King; Martin T. (Vashon, WA); Grower; Dale L. (Lansing, MI); Kushler;
Clifford A. (Vashon, WA); Grunbock; Cheryl A. (Vashon, WA) describe a disambiguating system in which an inference is made as to what the person might likely be trying to type. .4 drawback of this Tegic system is that the user must remain aware of what the machine thinks he or she is typing. There is an extra cognitive load imposed on the user, including the need to be constantly vigilant that errors are not being made. Using the Tegic system is a bit like using command line completion in Emacs.
~~'hile it allegedly purports to speed up the process, it can, in practice, slow down the process by imposing an additional burden on the user. In some sense, the Tegic system is a form of anti-redundancy, giving the user less flexibility. For example, forming new words (not in the dictionary) is quite difficult with the Tegic system, and when it does make mistakes, they are harder to detect because the mistakes get mapped onto the space of valid words.
Indeed, chordic redundancy (choice) is much more powerful than anti-redundancy (anti-choice) in how characters can be formed.
Ordinally conditional modifiers are a useful feature of some embodi-ments.
A modifier key is a key that alters the function of another key. On a standard key board, the SHIFT key modifies other letters by causing them to appear capitalized.
The SHIFT key modifies other keys between two states, namely a lowercase state and an uppercase state.
Another modifier key of the standard keyboard is the control key. A letter key pressed while the control key is held down is modified so that it becomes a control character. Thus the letter "a" gets changed to ''A" if it is pressed v~hile the control key is held down.
Fig. 8 illustrates an approach of having a modifier that is ordinally conditional, so that. its effect is responsive to where it is pressed in the chord. Fig 8 shows an example of how a four key ordinally conditional modifier is reduced to practice.
Letters are arranged in order of letter frequency starting with the letter "e"
which is the most commonly used letter of the alphabet.. Each of the 26 letters, the ten munbers, and some symbols are encoded with the 51 possible chords that can be formed from 3 switches, a middle finger switch, SW'm, an index finger switch, SWi, and a. thumb switch, SWt. (The more common letters such as e, t, a, etc., are also encoded redundantly so that there is more than one way to enter, for example, the letter "e".) A ring finger switch, SWr, is the ordinally conditional modifier.
If SWr is not pressed, an ordinary lowercase character is assumed. If a chord leads with SWr, the character is assumed to be an uppercase character. If the chord is entered while holding SWr the character is assumed to be a control character. If a chord trails with SV'r, it is assumed to represent a meta character. The ordinally conditional modifier is also applied to numbers to generate some of the symbols. For example, an exclamation mark is entered by leading with SWr, into the chord for the number 1.
One reason chording keyboards can be slow is that they often don't provide rollover. A regular ~W'ERTY... keyboard allows for rollover. A typical 1984 IBM
MODEL 1VI keyboard, for example, will be responsive to any key while the letter "q"
is held down. When the letters "q" and ''w" are held down, it is responsive to most keys (e.g. all those except keys in the q and w columns). When the letters "q", ''w" , and "e" are held down, it is responsive to other keys except from those three columns. When "q" , "w" , "e" , and "r" are held down, it is still responsive to keys in the right hand half of the keyboard (e.g. keys that would ordinarily be pressed with the right hand). Only when five keys are held down, does it stop responsing to new keypresses. Thus the MODEL M has quite a bit of rollover. This means that one can type new letters before finishing the typing of previous letters. This ability to have overlap between typing different letters allows a person to type faster because a new letter can be pressed before letting go of the previous letter.
Commercially available chording keyboards such as the Eandykey Twiddler and the BAT don't allow for rollover. Thus typing on the Twiddler or BAT is a slower process.
A goal of the cybernetic keyer is to be able to type much more quickly.
Therefore, an important feature, is the tradeoff between loopbacks at the Delay and Yaled stages (Fig 2), and rollover. If we decide, by design, that there will be no loopback at the Delay or the Yaled stages, we can assume that a chord has been committed to at the Release stage. Thus once we reach the Release stage, we can begin to accept another chord, so long as the other chord does not require the use of any switches that are still depressed at the Release stage. However, because of the sixty nine-fold chordic redundancy, it is arranged that, for most of the commonly following letters, there exists at least one new chord that can be built on keys not currently held down, at the Release stage.
As an example of rollover on a cybernetic keyer, consider the following: With reference to the two switch keyer, suppose we press SWITCH l, then press SWITCH 0, and then release SWITCH 1. W'e are now at the Release stage, and can enter a new command with SWITCH 1, since we know that there will be no Yaled loopback (e.g.
since we know that the chord will not involve pressing SWITCH 1 again). Thus pressing SWITCH 1 again can be safely used as a new symbol, prior to releasing SWITCH 0. In this way, symbols can rollover (overlap).
The chordic redundancy factor can be further increased with a more expressive keyer. The number of possible chords can be increased from 17685 to 7 + 84 +
1260 +
20160 + 302400 + +3628800 + 25401600 = 29354311 by simply adding three switches a.t the thumb position. This provides more than twenty nine million symbols, which is enough that each word in the English language could be represented in approximately a thousand different ways. This degree of chordic redundancy could provide for some very fast typing, if this many chords could be remembered. However, rather than increasing the number of switches, it is preferable to increase, instead, the expressivity of each one.
The pentakeyer is a crude instrument that lacks an ability to provide for tremen-dous "expression'' and sensitivity to the user. It fails to provide a rich form of Human-istic Intelligence in the sense that the feedback is cumbersome, and not continuous.
A guitar, violin, or real piano (e.g. an old fashioned mechanical piano, not a comput-erized synthetic piano data input keyboard), provides a much richer user experience because it provides instant feedback, and the user can experience the unique response of the instrument.
Even using both rails of each switch (e.g. double throw and even velocity sensing) still fails to provide a truly expressive input device. Accordingly, a better keyer was built from continuous transducers in the form of phonographic cartridges salvaged from old record players. These devices provide continuous flow of pressure information along two axes (each phono cartridge is responsive to pressure along two independent axes at right angles to one another, originally for playing stereo sound recordings from the grooves of a record). The dual sensor is depicted in Fig 9 and may comprise either a continuous sensor, or two on-off switches operable on separate axes.
In Fig. 9, Sensors SO and Sl may be switches or transducers or other forms of sensory apparatus. Sensors SO and S1 are operable individually, or together, by way of rocker Block BO1. Pressing straight down on Block BO1 will induce a response from both sensors. A response can also be induced in only one of sensors SO or S1 by pressing along its axis.
In this embodiment, the keyer described here is similar to the ternary data entry keyboard described by Langley; October 4, 1988, in U.S. Pat. 1\0. 4,775,255, in the sense that there are also to axes of each key; so that a given key can move towards or away from the operator, and has a central ''off" position, and a spring detent to make the key return to the central position in the absence of pressure from the finger. An important difference, though, is the fact that the keyboard of U.S.
Pat.
No. 4,775,255 has no ability to sense how fast, how hard, or how much each switch is pressed, other than just the ternary value of 0, +1, or -1. Also, in the keyer of U.S.
Pat. No. 4,775,255, the axes are not independent (e.g. one can't press +1 and -1 at the same time).
The sensor pair of Fig 9, on the other hand, provides two independent continuous dimensions. Thus a keyer rrrade from five such sensor pairs provides a much more expressive input. Even if the transducers are quantized, to binary outputs, there are four possibilities: 0, -l, +1, and ~1. W'ith five continuous transducers, one for each finger or thumb position, the user interface involves squeezing out characters, rather than clicking out characters. The input space is a very richly structured ten dimen-sional timespace, producing ten traces of time-varying signals. This ten dimensional timespace is reduced to discrete symbols according to a mapping based on compar-ison of the waveforms with reference waveforms stored in a computer system.
Since comparison with the reference waveforms is approximate, it is done by clustering, based on various kinds of similarity metrics.
Including just one time constant can greatly expand the functionality of a cybernetic keyer.
If we relax the ordinality constraint, and permit just one time constant, pertaining to simultaneity, we can obtain eleven symbols from just two switches. Such a scheme can be used for entering numbers, including the decimal point.
00 (Open chord not used) LFFL
WW
L': sing this simple coding scheme, the number zero is entered by pressing the LSK.
The number one is entered by pressing the V1SK. The number four, for example, is entered by pressing the MSK first, then pressing the LSK, and then releasing both at the same time, Within a certain time tolerance for which time is considered the same. (The letter ''W" denotes Within tolerance, as illustrated in With reference to Fig 10, Keyer Chords are shown within timing tolerances Y~ o and Wl. Time differences within the tolerance band are considered to be zero, so that events falling within the tolerance band defined by Wo and T~Yi are considered to be effectively simultaneous.
With reference to Fig 11, The decimal point is entered by pressing both switches at approximately the same tune, and releasing both at approximately the same time.
The ''approximately the same time" is defined according to a band around the line t« = tl of Fig 10.
Again, with reference to Fig. 11, a two switch chordic Keyer incorporating timing tolerances can produce eleven symbols. In addition to the unused Open chord, there are eleven other chords that can be used for the numbers 0 through 9, and the decimal point.
This number system can be implemented either by two pushbutton switches, or by a single vector keyswitch of two components, as illustrated in Fig 9. In the latter case, the entire set of number symbols can be entered with just one switch, e.g. by just one finger. :Vote that each number involves just a single keypress, unlike what would be the case if one entered numbers using a small wearable Vlorse code keyer, or the like. Thus the cybernetic chordic keyer provides a much more efficient entry of symbols.
Wearable keyers are known in the art of ham radio. For example, in L.S. Pat.
Vo.
4,194,085, March 18, 1980, Scelzi describes a "Finger keyer for code transmission''.
The telegraphic keyer fits over a finger, preferably the index finger, of an operator;
for tapping against the operator's thumb or any convenient object. It is used for transmission of code with portable equipment. The keyer is wearably operable when walking, or during other activities.
Keyers, such as previously known keyers, as well as the keyers of the present.
invention, such as the pentakeyer, and the continuous ten dimensional keying system, are much easier to use if they are custom made for each user. The most important aspect is getting the hand grip to fit well.
In a preferred embodiment, therefore, the keyer is moulded to fit the individual hand of the wearer. Presently, this is done by dipping the hand in icewater, and draping it with heated plastic material that is formed to the shape of the hand. Once the handpiece is formed, sensors are selected and installed so the keyer will match the specific attributes of the user's hand geometry.
Learning to use the pentakeyer is not easy, just as learning how to play a musical instrument is not easy. The pentakeyer evolved out of a difFerent philosophy, more than twenty years ago. This alternative philosophy knew nothing of so-called "user-friendly" user interface design, and therefore evolved along a completely different path.
Just as playing the violin is much harder to master than playing the TV remote control, it can also be much more rewarding and expressive. Thus if we were only to consider ease of use, vTe might be tempted to teach children how to operate a television because it is easier to learn than how to play a violin, or how to read and write. But if we did this, we would have an illiterate society in which all we could do were things that were easy to learn. It is the author's belief that a far richer experience can be attained with a lifelong computer interface that is worn on the body, and used constantly for ten to twenty years. On this kind of time scale, an apparatus that functions as a true extension of the mind and body, can result.
Just as it takes a long time to learn how to see, or to read and write, or to operate one's own body (e.g. it takes a number of years for the brain to figure out how to operate the body so that it can walk), it is expected that the most satisfying and powerful user interfaces will be learned over many years.
From the foregoing description, it will thus be evident. that the present invention provides a design for a personal cybernetic chordic keyer. .4s various changes can be made in the above embodiments and operating methods without departing from the spirit or scope of the invention, it is intended that all matter contained in the above description or show n in the accompanying drawings should be interpreted as illustrative and not in a limiting sense.
Variations or modifications to the design and construction of this invention, within the scope of the invention, may occur to those skilled in the art upon reviewing the disclosure herein. Such variations or modifications, if within the spirit of this invention, are intended to be encompassed within the scope of any claims to patent protection issuing upon this invention.
Claims (40)
1. A cybernetic keyer for communication, data entry, or the like, comprising:
~ a plurality of sensors;
~ a processor responsive to an output from each of said plurality of sensors, said processor for ranking at least one measurable aspect of each of said sensors, when at least some of said sensors are activated, said processor producing an output responsive to:
~ which of said sensors are activated;
~ the relative ranking of said some of said sensors.
~ a plurality of sensors;
~ a processor responsive to an output from each of said plurality of sensors, said processor for ranking at least one measurable aspect of each of said sensors, when at least some of said sensors are activated, said processor producing an output responsive to:
~ which of said sensors are activated;
~ the relative ranking of said some of said sensors.
2. A cybernetic keyer for communication, data entry, or the like, comprising:
~ a plurality of sensors;
~ a processor responsive to an output from each of said plurality of sensors, said processor ranking at least one measurable aspect of each of said sensors, when at least some of said sensors are activated, said cybernetic keyer producing an output character from a discrete alphabet of possible output characters, said output character responsive to:
~ which of said sensors are activated;
~ the relative ranking of said some of said sensors.
~ a plurality of sensors;
~ a processor responsive to an output from each of said plurality of sensors, said processor ranking at least one measurable aspect of each of said sensors, when at least some of said sensors are activated, said cybernetic keyer producing an output character from a discrete alphabet of possible output characters, said output character responsive to:
~ which of said sensors are activated;
~ the relative ranking of said some of said sensors.
3. The cybernetic keyer of one of claims 1 or 2, said at least one measurable aspect of each of said sensors being the order in which said sensors are pressed.
4. The cybernetic keyer of one of claims 1 or 2, said at least one measurable aspect of each of said sensors being the order in which said sensors are released.
5. The cybernetic keyer of one of claims 1 or 2, said at least one measurable aspect of each of said sensors being the velocity with which said sensors are pressed.
6. The cybernetic keyer of claim 5, said sensor being a double throw switch, said velocity measured as a time interval from when a common contact leaves a contact point at one throw of said switch, and meets another contact point at another throw of said switch.
7. The cybernetic keyer of one of claims 1 or 2, said at least one measurable aspect of each of said sensors being the force applied to each of said sensors.
8. The cybernetic keyer of one of claims 1 or 2, said at least one measurable aspect of each of said sensors being the earliness in which said sensors are activated.
9. The cybernetic keyer of claim 1, said processor ranking at least two measurable aspects of each of said sensors.
10. The cybernetic keyer of claim 9, one of said at least two measurable aspects of each of said sensors being the order in which said sensors are pressed, and another of said at least two measurable aspects of each of said sensors being the order in which said sensors are released.
11. A cybernetic keyer for communication; data entry, or the like, comprising:
~ at least three sensors;
~ a processor responsive to an output from each of said at least three sensors, said processor for ranking at least one measurable aspect of at least two of said sensors when said at least two of said sensors are activated, said cybernetic keyer producing an output character responsive to:
~ which two or more said sensors are activated;
~ the relative ranking of said two or more sensors.
~ at least three sensors;
~ a processor responsive to an output from each of said at least three sensors, said processor for ranking at least one measurable aspect of at least two of said sensors when said at least two of said sensors are activated, said cybernetic keyer producing an output character responsive to:
~ which two or more said sensors are activated;
~ the relative ranking of said two or more sensors.
12. A cybernetic keyer for communication, data entry, or the like, comprising:
~ at least two sensors;
~ a processor responsive to an output from each of said sensors, said processor for ranking at least one measurable aspect of said sensors when said sensors are both activated during the same time interval, said cybernetic keyer producing an output character responsive to relative ranking of said measurable aspect.
~ at least two sensors;
~ a processor responsive to an output from each of said sensors, said processor for ranking at least one measurable aspect of said sensors when said sensors are both activated during the same time interval, said cybernetic keyer producing an output character responsive to relative ranking of said measurable aspect.
13. A cybernetic keyer for communication, data entry, or the like, comprising:
~ at least two sensors, numbered S0 and S1;
~ a processor responsive to an output from each of said sensors, said processor for ranking at least one measurable aspect of said sensors when said sensors are activated during an overlapping time interval, said cybernetic keyer producing an output character responsive to relative ranking of said measurable aspect, said keyer providing a unique output symbol for each of the following manners in which the same two sensors are activated:
~ said measurable aspect of S0 is greater than that of S1 by at least W1;
~ said measurable aspect of S1 is greater than that of S0 by at least W0;
~ said measurable aspect of S0 and S1 are within a tolerance of W0 or W1 of one another.
~ at least two sensors, numbered S0 and S1;
~ a processor responsive to an output from each of said sensors, said processor for ranking at least one measurable aspect of said sensors when said sensors are activated during an overlapping time interval, said cybernetic keyer producing an output character responsive to relative ranking of said measurable aspect, said keyer providing a unique output symbol for each of the following manners in which the same two sensors are activated:
~ said measurable aspect of S0 is greater than that of S1 by at least W1;
~ said measurable aspect of S1 is greater than that of S0 by at least W0;
~ said measurable aspect of S0 and S1 are within a tolerance of W0 or W1 of one another.
14. A cybernetic keyer for communication, data entry, or the like,comprising:
~ at least two switches, numbered S0 and S1;
~ a processor responsive to an output from each of said switches, said keyer providing at least three different output symbols when S0 and S1 are both pressed only once during an overlapping time interval, said at least three different output symbols each corresponding to the following situations:
~ switch S0 is pressed before switch S1 by at least a positive constant W;
~ switch S1 is pressed before switch S0 by at least said positive constant W;
~ switch S0 and S1 are pressed within a time tolerance of W of one another.
~ at least two switches, numbered S0 and S1;
~ a processor responsive to an output from each of said switches, said keyer providing at least three different output symbols when S0 and S1 are both pressed only once during an overlapping time interval, said at least three different output symbols each corresponding to the following situations:
~ switch S0 is pressed before switch S1 by at least a positive constant W;
~ switch S1 is pressed before switch S0 by at least said positive constant W;
~ switch S0 and S1 are pressed within a time tolerance of W of one another.
15. A cybernetic keyer for communication, data entry, or the like, comprising:
at least two switches;
~ a processor responsive to an output from each of said switches, said keyer providing at least two different output symbols when the switches are pressed only once during an overlapping time interval, said at least two different output symbols each corresponding at least two of the following situations:
~ switch S0is released before switch S1 by at least a positive constant W;
~ switch S1 is released before switch S0 by at least said positive constant W;
~ switch S0 and S1 are released within a time tolerance of W of one another.
at least two switches;
~ a processor responsive to an output from each of said switches, said keyer providing at least two different output symbols when the switches are pressed only once during an overlapping time interval, said at least two different output symbols each corresponding at least two of the following situations:
~ switch S0is released before switch S1 by at least a positive constant W;
~ switch S1 is released before switch S0 by at least said positive constant W;
~ switch S0 and S1 are released within a time tolerance of W of one another.
16. A cybernetic keyer for communication, data entry, or the like, comprising:
~ at least two switches;
~ a processor responsive to an output from each of said switches, said keyer providing at least two different output symbols when the switches are pressed only once during an overlapping time interval, said at least two different output symbols each corresponding to the following situations:
~ switch S0 is released before switch S1;
~ switch S1 is released before switch S0.
~ at least two switches;
~ a processor responsive to an output from each of said switches, said keyer providing at least two different output symbols when the switches are pressed only once during an overlapping time interval, said at least two different output symbols each corresponding to the following situations:
~ switch S0 is released before switch S1;
~ switch S1 is released before switch S0.
17. A cybernetic keyer for communication, data entry, or the like, comprising:
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said processor for ranking at least one measurable aspect of some of said sensors activated with overlapping time intervals, said cybernetic keyer producing an output character responsive to relative ranking of said measurable aspect.
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said processor for ranking at least one measurable aspect of some of said sensors activated with overlapping time intervals, said cybernetic keyer producing an output character responsive to relative ranking of said measurable aspect.
18. A cybernetic keyer for communication, data entry, or the like, comprising:
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said keyer producing an output symbol in response to activated sensors, said processor responsive to an Attack of a first activated sensor, a. Close of a last activated sensor, a Release of a first unactivated sensor, and an Open of a last unactivated sensor, said symbol depending on at least three of the following:
~ which of said sensors is said first activated sensor;
~ which of said sensors is said last activated sensor;
~ which of said sensors is said first unactivated sensor;
~ which of said sensors is said last activated sensor.
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said keyer producing an output symbol in response to activated sensors, said processor responsive to an Attack of a first activated sensor, a. Close of a last activated sensor, a Release of a first unactivated sensor, and an Open of a last unactivated sensor, said symbol depending on at least three of the following:
~ which of said sensors is said first activated sensor;
~ which of said sensors is said last activated sensor;
~ which of said sensors is said first unactivated sensor;
~ which of said sensors is said last activated sensor.
19. A cybernetic keyer for communication, data entry, or the like, comprising:
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said keyer producing an output symbol in response to activated sensors, said processor, for at least four of the possible output symbols that can be generated by said keyer, responding to an Attack of a first activated sensor, a Close of a last activated sensor, a Release of a first unactivated sensor, and an Open of a last unactivated sensor, said output symbol depending on at least three of the following:
~ which of said sensors is said first activated sensor;
~ which of said sensors is said last activated sensor:
~ which of said sensors is said first unactivated sensor;
~ which of said sensors is said last activated sensor.
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said keyer producing an output symbol in response to activated sensors, said processor, for at least four of the possible output symbols that can be generated by said keyer, responding to an Attack of a first activated sensor, a Close of a last activated sensor, a Release of a first unactivated sensor, and an Open of a last unactivated sensor, said output symbol depending on at least three of the following:
~ which of said sensors is said first activated sensor;
~ which of said sensors is said last activated sensor:
~ which of said sensors is said first unactivated sensor;
~ which of said sensors is said last activated sensor.
20. A cybernetic keyer for communication, data entry, or the like, comprising:
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said keyer producing an output symbol in response to activated sensors, said processor, for at least four of the possible output symbols that can be generated by said keyer, responding to at least one of:
~ an ARPA stage when at least two of said sensors are activated;
~ an APRA stage when at least two of said sensors are unactivated.
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said keyer producing an output symbol in response to activated sensors, said processor, for at least four of the possible output symbols that can be generated by said keyer, responding to at least one of:
~ an ARPA stage when at least two of said sensors are activated;
~ an APRA stage when at least two of said sensors are unactivated.
21. The cybernetic keyer of one of claims 19, or 20, where said keyer produces the symbols PLAY, STOP, Fast Forward (FF), REWIND (REW), RECORD
(REC), and PAUSE.
(REC), and PAUSE.
22. The cybernetic keyer of claim 21 where said four symbols are Fast Forward (FF); REWIND (REW), RECORD (REC), and PAUSE.
23. A cybernetic keyer for communication, data entry, or the like, comprising:
~ a plurality of sensors, numbering N;
~ a processor responsive to an output from each of said sensors, said keyer producing distinct chords, said distinct.
chords being redundantly mapped to a lesser number of symbols output by said keyer, at least some of said symbols being responsive to an arpeggio or an oiggepra.
~ a plurality of sensors, numbering N;
~ a processor responsive to an output from each of said sensors, said keyer producing distinct chords, said distinct.
chords being redundantly mapped to a lesser number of symbols output by said keyer, at least some of said symbols being responsive to an arpeggio or an oiggepra.
24. A cybernetic keyer for communication, data entry, or the like, comprising:
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said keyer producing a symbol in response to a first chord entered on said keyer, said first chord not having Delay loopback, said first chord not having Yaled loopback, said keyer eventually producing a second symbol in response to a second chord, when entry of said second chord begins before sensors forming said first chord are all fully released.
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said keyer producing a symbol in response to a first chord entered on said keyer, said first chord not having Delay loopback, said first chord not having Yaled loopback, said keyer eventually producing a second symbol in response to a second chord, when entry of said second chord begins before sensors forming said first chord are all fully released.
25. A cybernetic keyer for communication, data entry, or the like, comprising at least three sensors, said keyer providing output symbols in response to the activation of various combinations of said at least three sensors, said output symbols comprising at least the numbers 0 through 9, where at least the number 0 is output in more than one way, said more than one way comprising at least two different combinations of pressing said at least three sensors to produce the number 0.
26. A cybernetic keyer for communication, data entry, or the like, comprising at least three sensors, said keyer providing output symbols in response to the activation of various combinations of said at least three sensors, said output symbols comprising at least the numbers 0 through 9, where at least some of the numbers are output in more than one way, where said keyer outputs each of said at least some of the numbers in response to at least two different chords.
27. A cybernetic keyer for communication, data entry, or the like, comprising:
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said plurality of sensors including at least one ordinally conditional modifier.
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said plurality of sensors including at least one ordinally conditional modifier.
28. The cybernetic keyer of claim 27 where a function of said ordinally conditional modifier includes being a shift key or being a control key, said function as to whether it operates as a shift key or as a control key being responsive to an order in which it is pressed within a chord.
29. The cybernetic keyer of any of claims 1 to 28, where said sensors are switches.
30. The cybernetic keyer of any of claims 1 to 28, where at least some of said sensors are dual axis switches.
31. The cybernetic keyer of any of claims 1 to 28, where said sensors are transducers.
32. The cybernetic keyer of any of claims 1 to 28, where at least some of said sensors comprise dual sensors, said dual sensors responsive to independent axes and having a common rocker Block.
33. The cybernetic keyer of any of claims 1 to 28, where said sensors are pressure transducers.
34. The cybernetic keyer of any of claims 1 to 28, where said sensors are force transducers.
35. The cybernetic keyer of any of claims 1 to 34, where said sensors are mounted into a detachable handle.
36. The cybernetic keyer of any of claims 1 to 34, where said sensors are mounted into the detachable handle of an electronic flashlamp.
37. The cybernetic keyer of any of claims 1 to 34, where said sensors are mounted into a transferable grip.
38. A method of manufacturing a cybernetic keyer, said method comprising the steps of:
~ heating a material to a temperature at which said material becomes soft;
~ forming said material to the hand of the intended user;
~ allowing said material to cool to a temperature at which said material becomes hard.
~ heating a material to a temperature at which said material becomes soft;
~ forming said material to the hand of the intended user;
~ allowing said material to cool to a temperature at which said material becomes hard.
39. A method of manufacturing a cybernetic keyer, said method comprising the steps of:
~ heating a material to a temperature at which said material becomes soft;
~ forming said material to the hand of the intended user;
~ allowing said material to cool to a temperature at which said material becomes hard;
~ installing a plurality of sensors in said material.
~ heating a material to a temperature at which said material becomes soft;
~ forming said material to the hand of the intended user;
~ allowing said material to cool to a temperature at which said material becomes hard;
~ installing a plurality of sensors in said material.
40. A cybernetic keyer for communication, data entry, or the like, comprising:
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said plurality of sensors borne by a transferable hand grip, said transferable hand grip providing an operationally hands-free keyer.
~ a plurality of sensors;
~ a processor responsive to an output from each of said sensors, said plurality of sensors borne by a transferable hand grip, said transferable hand grip providing an operationally hands-free keyer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA 2309868 CA2309868A1 (en) | 1999-06-29 | 2000-05-30 | Cybernetic keyer for transmitting or entering symbols of a discrete alphabet into a device such as a wearable computer or portable information processor |
Applications Claiming Priority (20)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| USCA2275798 | 1999-06-29 | ||
| CA002275798A CA2275798C (en) | 1998-06-29 | 1999-06-29 | Wristworn or handheld video production facility or videoconferencing system |
| USCA2275784 | 1999-06-29 | ||
| CA2,275,784 | 1999-06-29 | ||
| CA2,275,798 | 1999-06-29 | ||
| CA002275784A CA2275784C (en) | 1998-06-29 | 1999-06-29 | Wristwatch-based videoconferencing system |
| USCA2280022 | 1999-07-28 | ||
| CA002280022A CA2280022A1 (en) | 1999-07-28 | 1999-07-28 | Contact lens for the display of information such as text, graphics, or pictures |
| CA2,280,022 | 1999-07-28 | ||
| USCA2280420 | 1999-08-12 | ||
| CA002280420A CA2280420A1 (en) | 1998-10-13 | 1999-08-12 | Selective vitrionic viewing concealed by material such as polymer diffuser |
| CA2,280,420 | 1999-08-12 | ||
| CA2,280,425 | 1999-08-16 | ||
| CA002280425A CA2280425C (en) | 1998-10-13 | 1999-08-16 | Aremac incorporating a focus liberator so that displayed information is in focus regardless of where the lens of an eye of a user is focused |
| USCA2280425 | 1999-08-16 | ||
| US42193799A | 1999-10-21 | 1999-10-21 | |
| US42255999A | 1999-10-21 | 1999-10-21 | |
| US09/422559 | 1999-10-21 | ||
| US09/421937 | 1999-10-21 | ||
| CA 2309868 CA2309868A1 (en) | 1999-06-29 | 2000-05-30 | Cybernetic keyer for transmitting or entering symbols of a discrete alphabet into a device such as a wearable computer or portable information processor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA2309868A1 true CA2309868A1 (en) | 2000-12-29 |
Family
ID=27570293
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA 2309868 Abandoned CA2309868A1 (en) | 1999-06-29 | 2000-05-30 | Cybernetic keyer for transmitting or entering symbols of a discrete alphabet into a device such as a wearable computer or portable information processor |
Country Status (1)
| Country | Link |
|---|---|
| CA (1) | CA2309868A1 (en) |
-
2000
- 2000-05-30 CA CA 2309868 patent/CA2309868A1/en not_active Abandoned
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8125440B2 (en) | Method and device for controlling and inputting data | |
| US6262355B1 (en) | Method and apparatus for glove-based chording | |
| TWI274271B (en) | Data input device for individuals with limited hand function | |
| AU667688B2 (en) | Hand held computer input apparatus and method | |
| US20110209087A1 (en) | Method and device for controlling an inputting data | |
| JP6737996B2 (en) | Handheld controller for computer, control system for computer and computer system | |
| CN102549531B (en) | Processor interface | |
| EP0213022A1 (en) | Electronic single-handedly operated keyboard | |
| TWI278769B (en) | Portable input method for wearable glove keyboard | |
| US6429854B1 (en) | Stealthy keyboard | |
| JP2002169653A (en) | Touch-type input unit | |
| CA2309868A1 (en) | Cybernetic keyer for transmitting or entering symbols of a discrete alphabet into a device such as a wearable computer or portable information processor | |
| Williamson et al. | Efficient human-machine control with asymmetric marginal reliability input devices | |
| US5151873A (en) | Calculator with music generating device | |
| US6848083B2 (en) | Data input method and device for a computer system | |
| KR20020053988A (en) | Apparatus and method for pointing a braille | |
| US11249558B1 (en) | Two-handed keyset, system, and methods of making and using the keyset and system | |
| ATE67446T1 (en) | ONE-HAND BUTTON HOUSING. | |
| KR20060022984A (en) | Keypad glove apparatus | |
| US20030090456A1 (en) | Method of mapping linguistic symbols to a sequence of keystrokes and assist memorization with musical notes | |
| TW201931071A (en) | Input sensing system input sensing device and method of operating the same | |
| JP2001242986A (en) | Information input device | |
| Felzer et al. | Experiences of someone with a neuromuscular disease in operating a PC (and ways to successfully overcome challenges) | |
| CN102405456A (en) | Data entry system | |
| Lee et al. | Chording as a text entry method in mobile phones |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request | ||
| FZDE | Dead |