4.0438 Handwriting Technology (5/229)
Elaine Brennan & Allen Renear (EDITORS@BROWNVM.BITNET)
Thu, 30 Aug 90 22:51:00 EDT
Humanist Discussion Group, Vol. 4, No. 0438. Thursday, 30 Aug 1990.
(1) Date: Wed, 29 Aug 90 07:41:35 EDT (18 lines)
From: Germaine Warkentin <WARKENT@vm.epas.utoronto.ca>
Subject: handwriting
(2) Date: 29 August 1990 18:03:43 CDT (16 lines)
From: "Michael Sperberg-McQueen 312 996-2477 -2981" <U35395@UICVM>
Subject: why handwriting recognition, Chinese
(3) Date: 28 Aug 90 22:45:02 EST (12 lines)
From: James O'Donnell <JODONNEL@PENNSAS>
Subject: handwriting or ...?
(4) Date: Wed, 29 Aug 90 10:12:37 EDT (118 lines)
From: "Adam C. Engst" <PV9Y@CORNELLA>
Subject: Re: 4.0430 Handwriting Technology. Why?
(5) Date: Wed, 29 Aug 1990 17:48:47 EDT (65 lines)
From: "James H. Coombs" <JAZBO@BROWNVM>
Subject: Re: 4.0430 Handwriting Technology. Why?
(1) --------------------------------------------------------------------
Date: Wed, 29 Aug 90 07:41:35 EDT
From: Germaine Warkentin <WARKENT@vm.epas.utoronto.ca>
Subject: handwriting
My first response to the message about handwriting for computer screens
was "ridiculous!", but then that was my first response to the notion of
an electric can-opener. Then somebody told me how useful they were for
handicapped people. Jim Coombs' message suggests that the second is the
better one, particularly as I feel the occasional arthritic twinge in
that notorious left hand of mine! But another thought occurs to me: if
the handwriting has to be scanned for transmission, then experience with
scanners (which often still balk at early printed books) would suggest
that we are going to have to improve our wretched scrawls. As I depart
for a much-anticipated research trip abroad, I leave HUMANIST with the
apocalyptic vision of future generations working at their penmanship
lessons, loop after patient loop, just the way I had to even in the
backward forties! cheers, and talk to you all again in six or eight
weeks! Germaine.
(2) --------------------------------------------------------------26----
Date: 29 August 1990 18:03:43 CDT
From: "Michael Sperberg-McQueen 312 996-2477 -2981" <U35395@UICVM>
Subject: why handwriting recognition, Chinese
Recognizing Chinese is a nice goal for handwriting recognition, but
wouldn't it be simpler to catalogue the direction, shape, and order of
strokes and use keystroke input? I assume in fact that it is, since at
least some Chinese-input systems do use stroke-by-stroke construction to
do the job. (It is sometimes suggested the canonical stroke ordering
should radically simplify the task of recognizing the character, so that
recognition could proceed almost at the rate of typing -- but sometimes
it is replied that tests of such systems with native speakers reveal
that people don't always use the same stroke ordering. Back to the
drawing board...)
Michael Sperberg-McQueen, UIC
(3) --------------------------------------------------------------17----
Date: 28 Aug 90 22:45:02 EST
From: James O'Donnell <JODONNEL@PENNSAS>
Subject: handwriting or ...?
Has anybody ever devised or seen a full keyboard (all 26 letters plus 10
numerals plus punctuation if not all the symbols on the shift of the
numeral keys) designed to be used by one hand? Thinking about the
problems in reduction in system size, the two main obstacles are a
display big enough for our gross eyes to read and a keypad big enough
for our gross hands to manipulate. Substantial saving in space and
enhancement of convenience (using with one hand while the other hand is
busy or absent) if there could be such a critter.
(4) --------------------------------------------------------------134---
Date: Wed, 29 Aug 90 10:12:37 EDT
From: "Adam C. Engst" <PV9Y@CORNELLA>
Subject: Re: 4.0430 Handwriting Technology. Why?
-----------------
[What follows is Engst's reply to Coombs' posting in Humanist 4.431.
Coombs' points are presented flush left; Engst's commentary, indented
beginning with a '--', follows each point. This reply was originally
sent directly to Coombs who then forwarded it, on Engst's request,
to Humanist. In the following item in this digest Coombs responds
to Engst's comments. --ahr]
-----------------
>From: "Adam C. Engst" <PV9Y@CORNELLA>
[...]
>What I wonder, and pose to Humanists, is what really is the use of
>handwriting technology?
1. Requires only one hand to write. You can use the other hand to hold a
book open, etc. I use a book holder now, but some books insist on
snapping shut. You can also use one hand to hold the machine and the
other to write.
-- But only one hand is allowed as well. Most of the time when you
write the other hand sits limply on the desk.
2. Handwriting technology permits free-hand drawing as well. For parse
trees, flow charts, org. charts, etc. Not good at free hand? Use it
for rough drafts. Eventually, some pattern matcher will probably be
able to convert a free-hand drawing into structured graphics.
--Perhaps, but this brings in the ergonomics again. I can't imagine
drawing on the SE/30 screen directly in front of my face, where it sits
for optimum reading.
3. Keyboards tend to be too noisy for meetings, classes, libraries, etc.
Handwriting is quiet.
-- Good point.
4. Some people don't type. Some people have only one hand.
-- True in both cases, but a stylus is still a very slow text input
method. Perhaps a better distinction would be "Some people don't input
much text." Those that do (and care about speed) should use keyboards.
People with one hand can use chording keyboards which can be even
faster than normal ones.
5. Perhaps, there is less stress on tendons from handwriting than from
striking keys. On the other hand, I do not get cramped fingers from
typing.
-- I've never heard of carpal tunnel syndrome in relation to
handwriting, but that doesn't imply that keyboards are bad, merely
that current keyboard designs are bad. I get writer's cramp
incredibly quickly these days and had basically stopped taking
unnecessary notes senior year because of it.
6. The proportion of "checking-off" activities to other computing may
not be so small that we should postpone handwriting technology. We
should be careful here about what we compare also: people, hours, $? We
should also consider the possibility that handwriting technology will
cause a shift in practices. For an idealistic example, let's say that
electronic technology enables an aircraft manufacturer to rapidly
disseminate information about critical pre-flight inspections. The
check-off list for an aircraft may suddenly have an extra entry---say,
to check for cracks in a certain high stress area. They find the
cracks. The aircraft that would have crashed does not. This scenario
may never occur. I use it simply to illustrate that we need to think
about the applicability of a technology to "critical" problems, where a
small proportion of hours of use may carry a much higher value to
society.
-- Very true, and I do see the utility in those limited situations. I
was more concerned with the view expressed in the trade magazines that
handwriting was the neatest technology around (because IBM was sinking
a lot of money into it, in part) and that it would become the dominant
input device. One small problem is that the pen is not attached to
the computer and I know that I lose pens with an unerring frequency.
I also have problems keeping them close at hand when I want them.
It's hard to lose a mouse, keyboard, or trackball and they seldom
wander from the computer desk.
I'm sure that there are other advantages. I should also point out that
most people see handwriting as an adjunct to typing, not as a
replacement. From this perspective, I feel little need to discuss the
disadvantages of handwriting in the abstract. I think most of us would
love to have a machine to take to the library, etc., given perfect
handwriting recognition, zero weight, zero cost, perfect compatibility
with our base work stations, etc. I think the main questions now have
to do with the success of the various attempts to produce machines. (I
also assume that we will have handwriting machines before we have a
universal electronic library available instantaneously to every
machine.)
-- I definitely have not gotten the impressions from the magazines
that handwriting technology is to be an adjunct to typing. That's
why these machines are all laptops - because then the keyboard
can be eliminated without causing too much fuss. A laptop screen
may be a terrible ergonomic writing surface (I can't stand writing
on clipboards when standing up, for instance) but it will be basically
impossible to write on standard desktop machine screens (which would
be necessary to remove one level of metaphor, as Gates suggests)
unless the screen are embedded in drafting table-style desktops. As
an adjunct, it is of course hard to contest. I am also not contesting
it for the sake of being spiteful or anything, I'm merely curious and
see more cons than pros. Far be it for me to stand in the way of any
new technology, though I must admit a certain suspicion whenever IBM
is involved heavily. :-)
Jim
Dr. James H. Coombs
Chief Architect
Institute for Research in Information and Scholarship (IRIS)
Brown University, Box 1946
Providence, RI 02912
jazbo@brownvm.bitnet
(5) --------------------------------------------------------------69----
Date: Wed, 29 Aug 1990 17:48:47 EDT
From: "James H. Coombs" <JAZBO@BROWNVM>
Subject: Re: 4.0430 Handwriting Technology. Why?
In response to personal mail from Adam C. Engst:
>-- I definitely have not gotten the impressions from the magazines that
>handwriting technology is to be an adjunct to typing. That's why these
>machines are all laptops - because then the keyboard can be eliminated
>without causing too much fuss.
At least one manufacturer has taken plans to enable the "docking" of a
handwriting machine to a workstation network. We also have a
handwriting pad that can be attached to a microcomputer, which is also
equipped with a keyboard.
>A laptop screen may be a terrible ergonomic writing surface (I can't stand
>writing on clipboards when standing up, for instance) but it will be
>basically impossible to write on standard desktop machine screens (which
>would be necessary to remove one level of metaphor, as Gates suggests) unless
>the screen are embedded in drafting table-style desktops.
I think that writing on a standard screen is out of the question. The
idea is to have a writing surface that one can manipulate much like a
notepad. The writing surface may also be the display on a laptop.
I don't know about the metaphorical aspects of handwriting interfaces.
The question seems much too complicated for me to extemporize about. An
example. It seems to me that to use the word "menu" for those lists of
choices that one sees on the Macintosh is a metaphorical use of the term
menu motivated by someone's perception of a similarity with restaurant
menus. I doubt very much, however, that users use their knowledge of
restaurant menus to help them understand Macintosh menus and how to use
them. So what is the role of the menu metaphor? And isn't the whole
thing supposed to use the "desk top" metaphor? Then what are we doing
with menus on it. And what is that trash barrel doing on my "desk top."
Is it a mixed metaphor? I don't know. I think "direct manipulation"
can be useful, but I have never seen any evidence to support the
contention that these metaphors help people understand and use
computers. Yes, we can in some nice, academic sense count the metaphors;
perhaps we can even eliminate a "level" of metaphor. But what evidence
do we have that we have changed anything other than our vocabulary in
discussing the interface? It's all such a mess (What are these windows
doing in the middle of my desktop?!) that I can't believe that the
metaphors have much to do with people's actual computing. If they did,
I think we would hear people saying, "That doesn't make sense! Desktops
don't have windows. And trash cans go on the floor. And menus belong
in restaurants." Since they don't say such things, I feel justified in
believing that the metaphors are just a veneer of terminology that we
use in discussing our interfaces. If I am right, then it is a little
bit ridiculous to talk about adjustments to this veneer as major
developments in computing. The handwriting technology may be a major
development, but the adjustments to our terminology should not be
treated as necessarily or even probably having a major effect in the
cognitive processing that we perform while computing. And we should not
let marketing gas affect our evaluation of the technology.
--Jim
Dr. James H. Coombs
Chief Architect
Institute for Research in Information and Scholarship (IRIS)
Brown University, Box 1946
Providence, RI 02912
jazbo@brownvm.bitnet
>would be necessary to remove one level of metaphor, as Gates suggests) unless
>the screen are embedded in drafting table-style desktops.
I think that writing on a standard screen is out of the question. The
idea is to have a writing surface that one can manipulate much like a
notepad. The writing surface may also be the display on a laptop.
I don't know about the metaphorical aspects of handwriting interfaces.
The question seems much too complicated for me to extemporize about. An
example. It seems to me that to use the word "menu" for those lists of
choices that one sees on the Macintosh is a metaphorical use of the term
menu motivated by someone's perception of a similarity with restaurant
menus. I doubt very much, however, that users use their knowledge of
Received: by BROWNVM (Mailer R2.07) id 2374; Thu, 30 Aug 90 22:54:57 EDT
Date: Thu, 30 Aug 90 22:51:00 EDT
From: Elaine Brennan & Allen Renear <EDITORS@BROWNVM>
Subject: 4.0438 Handwriting Technology (5/229)
To: Humanist Discussion <HUMANIST@BROWNVM>
Humanist Discussion Group, Vol. 4, No. 0438. Thursday, 30 Aug 1990.
(1) Date: Wed, 29 Aug 90 07:41:35 EDT (18 lines)
From: Germaine Warkentin <WARKENT@vm.epas.utoronto.ca>
Subject: handwriting
(2) Date: 29 August 1990 18:03:43 CDT (16 lines)
From: "Michael Sperberg-McQueen 312 996-2477 -2981" <U35395@UICVM>
Subject: why handwriting recognition, Chinese
(3) Date: 28 Aug 90 22:45:02 EST (12 lines)
From: James O'Donnell <JODONNEL@PENNSAS>
Subject: handwriting or ...?
(4) Date: Wed, 29 Aug 90 10:12:37 EDT (118 lines)
From: "Adam C. Engst" <PV9Y@CORNELLA>
Subject: Re: 4.0430 Handwriting Technology. Why?
(5) Date: Wed, 29 Aug 1990 17:48:47 EDT (65 lines)
From: "James H. Coombs" <JAZBO@BROWNVM>
Subject: Re: 4.0430 Handwriting Technology. Why?
(1) --------------------------------------------------------------------
Date: Wed, 29 Aug 90 07:41:35 EDT
From: Germaine Warkentin <WARKENT@vm.epas.utoronto.ca>
Subject: handwriting
My first response to the message about handwriting for computer screens
was "ridiculous!", but then that was my first response to the notion of
an electric can-opener. Then somebody told me how useful they were for
handicapped people. Jim Coombs' message suggests that the second is the
better one, particularly as I feel the occasional arthritic twinge in
that notorious left hand of mine! But another thought occurs to me: if
the handwriting has to be scanned for transmission, then experience with
scanners (which often still balk at early printed books) would suggest
that we are going to have to improve our wretched scrawls. As I depart
for a much-anticipated research trip abroad, I leave HUMANIST with the
apocalyptic vision of future generations working at their penmanship
lessons, loop after patient loop, just the way I had to even in the
backward forties! cheers, and talk to you all again in six or eight
weeks! Germaine.
(2) --------------------------------------------------------------26----
Date: 29 August 1990 18:03:43 CDT
From: "Michael Sperberg-McQueen 312 996-2477 -2981" <U35395@UICVM>
Subject: why handwriting recognition, Chinese
Recognizing Chinese is a nice goal for handwriting recognition, but
wouldn't it be simpler to catalogue the direction, shape, and order of
strokes and use keystroke input? I assume in fact that it is, since at
least some Chinese-input systems do use stroke-by-stroke construction to
do the job. (It is sometimes suggested the canonical stroke ordering
should radically simplify the task of recognizing the character, so that
recognition could proceed almost at the rate of typing -- but sometimes
it is replied that tests of such systems with native speakers reveal
that people don't always use the same stroke ordering. Back to the
drawing board...)
Michael Sperberg-McQueen, UIC
(3) --------------------------------------------------------------17----
Date: 28 Aug 90 22:45:02 EST
From: James O'Donnell <JODONNEL@PENNSAS>
Subject: handwriting or ...?
Has anybody ever devised or seen a full keyboard (all 26 letters plus 10
numerals plus punctuation if not all the symbols on the shift of the
numeral keys) designed to be used by one hand? Thinking about the
problems in reduction in system size, the two main obstacles are a
display big enough for our gross eyes to read and a keypad big enough
for our gross hands to manipulate. Substantial saving in space and
enhancement of convenience (using with one hand while the other hand is
busy or absent) if there could be such a critter.
(4) --------------------------------------------------------------134---
Date: Wed, 29 Aug 90 10:12:37 EDT
From: "Adam C. Engst" <PV9Y@CORNELLA>
Subject: Re: 4.0430 Handwriting Technology. Why?
-----------------
[What follows is Engst's reply to Coombs' posting in Humanist 4.431.
Coombs' points are presented flush left; Engst's commentary, indented
beginning with a '--', follows each point. This reply was originally
sent directly to Coombs who then forwarded it, on Engst's request,
to Humanist. In the following item in this digest Coombs responds
to Engst's comments. --ahr]
-----------------
>From: "Adam C. Engst" <PV9Y@CORNELLA>
[...]
>What I wonder, and pose to Humanists, is what really is the use of
>handwriting technology?
1. Requires only one hand to write. You can use the other hand to hold a
book open, etc. I use a book holder now, but some books insist on
snapping shut. You can also use one hand to hold the machine and the
other to write.
-- But only one hand is allowed as well. Most of the time when you
write the other hand sits limply on the desk.
2. Handwriting technology permits free-hand drawing as well. For parse
trees, flow charts, org. charts, etc. Not good at free hand? Use it
for rough drafts. Eventually, some pattern matcher will probably be
able to convert a free-hand drawing into structured graphics.
--Perhaps, but this brings in the ergonomics again. I can't imagine
drawing on the SE/30 screen directly in front of my face, where it sits
for optimum reading.
3. Keyboards tend to be too noisy for meetings, classes, libraries, etc.
Handwriting is quiet.
-- Good point.
4. Some people don't type. Some people have only one hand.
-- True in both cases, but a stylus is still a very slow text input
method. Perhaps a better distinction would be "Some people don't input
much text." Those that do (and care about speed) should use keyboards.
People with one hand can use chording keyboards which can be even
faster than normal ones.
5. Perhaps, there is less stress on tendons from handwriting than from
striking keys. On the other hand, I do not get cramped fingers from
typing.
-- I've never heard of carpal tunnel syndrome in relation to
handwriting, but that doesn't imply that keyboards are bad, merely
that current keyboard designs are bad. I get writer's cramp
incredibly quickly these days and had basically stopped taking
unnecessary notes senior year because of it.
6. The proportion of "checking-off" activities to other computing may
not be so small that we should postpone handwriting technology. We
should be careful here about what we compare also: people, hours, $? We
should also consider the possibility that handwriting technology will
cause a shift in practices. For an idealistic example, let's say that
electronic technology enables an aircraft manufacturer to rapidly
disseminate information about critical pre-flight inspections. The
check-off list for an aircraft may suddenly have an extra entry---say,
to check for cracks in a certain high stress area. They find the
cracks. The aircraft that would have crashed does not. This scenario
may never occur. I use it simply to illustrate that we need to think
about the applicability of a technology to "critical" problems, where a
small proportion of hours of use may carry a much higher value to
society.
-- Very true, and I do see the utility in those limited situations. I
was more concerned with the view expressed in the trade magazines that
handwriting was the neatest technology around (because IBM was sinking
a lot of money into it, in part) and that it would become the dominant
input device. One small problem is that the pen is not attached to
the computer and I know that I lose pens with an unerring frequency.
I also have problems keeping them close at hand when I want them.
It's hard to lose a mouse, keyboard, or trackball and they seldom
wander from the computer desk.
I'm sure that there are other advantages. I should also point out that
most people see handwriting as an adjunct to typing, not as a
replacement. From this perspective, I feel little need to discuss the
disadvantages of handwriting in the abstract. I think most of us would
love to have a machine to take to the library, etc., given perfect
handwriting recognition, zero weight, zero cost, perfect compatibility
with our base work stations, etc. I think the main questions now have
to do with the success of the various attempts to produce machines. (I
also assume that we will have handwriting machines before we have a
universal electronic library available instantaneously to every
machine.)
-- I definitely have not gotten the impressions from the magazines
that handwriting technology is to be an adjunct to typing. That's
why these machines are all laptops - because then the keyboard
can be eliminated without causing too much fuss. A laptop screen
may be a terrible ergonomic writing surface (I can't stand writing
on clipboards when standing up, for instance) but it will be basically
impossible to write on standard desktop machine screens (which would
be necessary to remove one level of metaphor, as Gates suggests)
unless the screen are embedded in drafting table-style desktops. As
an adjunct, it is of course hard to contest. I am also not contesting
it for the sake of being spiteful or anything, I'm merely curious and
see more cons than pros. Far be it for me to stand in the way of any
new technology, though I must admit a certain suspicion whenever IBM
is involved heavily. :-)
Jim
Dr. James H. Coombs
Chief Architect
Institute for Research in Information and Scholarship (IRIS)
Brown University, Box 1946
Providence, RI 02912
jazbo@brownvm.bitnet
(5) --------------------------------------------------------------69----
Date: Wed, 29 Aug 1990 17:48:47 EDT
From: "James H. Coombs" <JAZBO@BROWNVM>
Subject: Re: 4.0430 Handwriting Technology. Why?
In response to personal mail from Adam C. Engst:
>-- I definitely have not gotten the impressions from the magazines that
>handwriting technology is to be an adjunct to typing. That's why these
>machines are all laptops - because then the keyboard can be eliminated
>without causing too much fuss.
At least one manufacturer has taken plans to enable the "docking" of a
handwriting machine to a workstation network. We also have a
handwriting pad that can be attached to a microcomputer, which is also
equipped with a keyboard.
>A laptop screen may be a terrible ergonomic writing surface (I can't stand
>writing on clipboards when standing up, for instance) but it will be
>basically impossible to write on standard desktop machine screens (which
>would be necessary to remove one level of metaphor, as Gates suggests) unless
>the screen are embedded in drafting table-style desktops.
I think that writing on a standard screen is out of the question. The
idea is to have a writing surface that one can manipulate much like a
notepad. The writing surface may also be the display on a laptop.
I don't know about the metaphorical aspects of handwriting interfaces.
The question seems much too complicated for me to extemporize about. An
example. It seems to me that to use the word "menu" for those lists of
choices that one sees on the Macintosh is a metaphorical use of the term
menu motivated by someone's perception of a similarity with restaurant
menus. I doubt very much, however, that users use their knowledge of
restaurant menus to help them understand Macintosh menus and how to use
them. So what is the role of the menu metaphor? And isn't the whole
thing supposed to use the "desk top" metaphor? Then what are we doing
with menus on it. And what is that trash barrel doing on my "desk top."
Is it a mixed metaphor? I don't know. I think "direct manipulation"
can be useful, but I have never seen any evidence to support the
contention that these metaphors help people understand and use
computers. Yes, we can in some nice, academic sense count the metaphors;
perhaps we can even eliminate a "level" of metaphor. But what evidence
do we have that we have changed anything other than our vocabulary in
discussing the interface? It's all such a mess (What are these windows
doing in the middle of my desktop?!) that I can't believe that the
metaphors have much to do with people's actual computing. If they did,
I think we would hear people saying, "That doesn't make sense! Desktops
don't have windows. And trash cans go on the floor. And menus belong
in restaurants." Since they don't say such things, I feel justified in
believing that the metaphors are just a veneer of terminology that we
use in discussing our interfaces. If I am right, then it is a little
bit ridiculous to talk about adjustments to this veneer as major
developments in computing. The handwriting technology may be a major
development, but the adjustments to our terminology should not be
treated as necessarily or even probably having a major effect in the
cognitive processing that we perform while computing. And we should not
let marketing gas affect our evaluation of the technology.
--Jim
Dr. James H. Coombs
Chief Architect
Institute for Research in Information and Scholarship (IRIS)
Brown University, Box 1946
Providence, RI 02912
jazbo@brownvm.bitnet