Oh, my. A whole paragraph on sci-fi interfaces and not a word of=20
cyberpunk. As a fan, I must (light heartedly, I assure you, but perhaps =
with some interesting material if anyone bothers to reads on) correct=20
this omission. Most works in this sub-genre make use the ubiquitous=20
direct neural interface which provides both input and output usually in=20
the form of interactive sensual imagery/cyberscapes. Obviously=20
producing neural input is a long ways off but we have plenty of ways of=20
displaying visuals of the cyberscape to substitute (and a booming 3d=20
game industry pushing such technology). Output wise we already have=20
external transducers that can report brain activity so developing this=20
into extracting some motion/approval/disapproval and other control=20
commands isn't impossible. The more pared down variants of this output=20
method are even more doable. I can easily imagine something you wear on =
your arm/wrist/back of hand that tracks tendon movement or motor nerve=20
firing for input (similar systems used in cyber punk for weapons=20
control, of course ^^). So with potential input and output available a=20
cyberscape interface is possible at some point. The act of moving=20
around a generated world to control operations and communications and=20
approving and disapproving of various options encountered is something=20
people shouldn't have trouble adopting; it's basically how we handle the =
real world.
Next up in cyberpunk is information retrieval via eye commands and=20
tracking. Not a complete interface in and of itself it is still almost=20
second nature to look at something you want more information on. The=20
real world actually has work progressing on augmented reality so I guess =
I shouldn't get into the wide range of augmented senses in cyberpunk. =20
Other interfaces are much like very advanced speech recognition variants =
which you quickly dismissed. Subvocal microphones placed on the throat=20
are used to prevent the need for being alone/quiet (and similarly=20
unobtrusive speakers for return audio info are obviously doable). I=20
haven't seen anything about subvocal mics in the real world. I suspect=20
vocal cord only speech would be much more difficult to piece together=20
than the normal speech and possibly require a specialized language=20
rather than just advanced computer interpretation/training. Still=20
speech is another natural choice for getting data to and from humans and =
often used in cyberpunk far more effectively than star trek. Some=20
novels have used the "thought transducers" from the previous paragraph=20
to piece together language like commands and others have visual=20
prompting to help make the voice commands more specific. One cyberpunk=20
roleplaying game I play keeps track of how many words per action segment =
your character can handle using various technologies. ^^
Hmm, I guess it's odd to write so much in response to one paragraph of a =
letter focused on more important issues. Covering some more relevant=20
details I would like to mention I prefer MS/drill down menus as opposed=20
to esoteric/lengthy/hard to remember commands or macro combinations and=20
buttons. I spend a lot of hours on windows computers and hardly ever=20
even use those meaningless function keys and never, ever have to look up =
a command which I see as big pluses. You seem to place a lot of weight=20
on popular acceptance and widespread usability and MS certainly has=20
that. Clickable menus can present more options and more information=20
about the options than a more complex predefined interface since they=20
have the powerful advantage of interactive visual feedback. Maybe I=20
missed what exactly you all dislike about MS interfaces since I don't=20
have much experience with alternative GUIs.
Just as an aside on the more mechanical aspect of present day interfaces =
I can't imagine ever using a trackball for anything (maybe that was the=20
joke ^^). Maybe we need a message chain about what's out and usable,=20
but never having used one they look both inaccurate and scarily=20
repetitive with that petting motion I see people using. Optical mice on =
the other hand work anywhere, I use one in bed on the sheets all the=20
time and am working on building a wireless one into a wrist band. For=20
laptops microjoysticks/trackpoints require zero motion for mousing, just =
pressure. My dream interface has always been to mount a wireless,=20
thumb-controlled microjoystick on the base of my index finger. Mousing=20
capability aside I think I could out type all those chording interface=20
people using a two stroke matrix (ie. up then up is one key, up then=20
up-right is another, etc) but that is starting to fall out of the realm=20
of an interface that is acceptable and usable by everyone which is what=20
I got for the goal of your letter.
Carol Stein wrote:
>>>Wearable, just like PDAs, are not desktop and they
>>> =20
>>>
>>An excellent point. While people will undoubtedly want=20
>>something along the same lines, much as Palm/OS echoes
>>many of the UI conventions used in MSWindows, Windows'
>>acceptance for desktops does not mean that it will be the
>>predominant wearable interface.
>>
>>Or, at least I hope not...it's klunky enough with a stationary
>>computer and trackball!
>> =20
>>
>
>N.B. This turned into a long piece. Call it a Sunday sermon, and skip it=
if you're not interested.
>
>First, any wearable computer that does not REPLACE a laptop or desktop w=
on't succeed... by which I mean, if we need to keep the desktop/laptop as=
well as the wearable (to take care of tasks the wearable can't handle), =
why bother?
>
>Second, yes, the MS interfaces are awful. What bothers me, however, is t=
hat usually we (humans) only build what we first imagine. And in sci-fi, =
where most of the imagining for this sort of thing takes place, the imagi=
nings of an interface remain very primitive. All the Star Trek series, fo=
r example, either rely on scanners (and you have to look at a small, clun=
ky handheld device even to make out the result) or a voice-driven, somew=
hat structured, query system ("Computer, access the personal log of ...";=
"How many references to ...?"). Or we see Captain Jean-Luc sitting at hi=
s desk, working on a laptop. Okay, we can agree tv-based sci-fi is aimed =
at L.C.D., and thus has to account for stupidity... but most real sci-fi =
is not much more imaginative in this arena.
>
>Voice-based interfaces are a problem -- unless we expect to use wearable=
s only when we're alone, or we're in a relatively quiet environment and w=
e're the only one speaking (or at least the main speaker, as in facilitat=
ing a business meeting, teaching a workshop, etc.). Mouse-gesture interfa=
ces, broadly defined, are still going to be limited to a relatively narro=
w band of options, unless we want to enshine the silly menu drill-down pa=
radigm for another decade. To borrow the progression of humanity's progre=
ss in an earlier medium, what is efficient will be "interchangeable type.=
" We might not necessarily need a keyboard, but we sure do need a relativ=
ely small symbol set (gestures or whatever) that can be meaningfully comb=
ined in infinite ways.
>
>
>For now, I figure the symbol set might as well be the familiar characers=
of the English language, which has already been widely adopted by the co=
mputer-literate part of the world.
>
>Now here's a cautionary tale: Kenneth Iverson, around 1964, sugggested a=
new set of symbols to be used in addition to those characters. In his sc=
heme, the familiar characters represented data, and parameter values; his=
new set represented very powerful operators or "primitive functions." He=
devised his operators and primitive functions in order to describe effic=
iently a powerful, then-new computer. Because at the time he was an IBM f=
ellow, his new symbol set was taken from characters that could fit on an =
IBM selectric typewriter ball -- many Greek characters, for example.
>
>Later on, someone realized since his new notation uniquely represented o=
perations, it could become A Programming Language. So they wrote an inter=
preter for it. (Yes, I'm talking about APL. How many of you even knew tha=
t much? How many have heard of APL? And you are the smart ones.)
>
>Now, as years passed, quite a few intelligent humans became devotees of =
APL. (I have worked with many programmers in my 20+ years documenting sys=
tems & programs, and I can tell you APL programmers are simply not like o=
ther programmers. They tend to be noticeably more intelligent, creative, =
and fun to be with. I personally would far rather work in APL or J than i=
n any other language I've ever learned or known of.) APLers formed groups=
and user groups and developed new languages. (How many COBOL programmers=
get together after work, enthusiastic about enhancing or exploring COBOL=
??) APLers argued that the reason their far superior language (which was =
intrinsically object-oriented, btw, treating data as one big amorphous bl=
ock that could be addressed in any fashion, not just linearly -- i.e., li=
near =3D read in record, compare field values, process, store or write ou=
tput, read next record) was not spreading because one needed a special ke=
yboard or stickers or something.
>
>
>And, lo, one of Iverson's sons, working on Wall Street for an infamous b=
roker (infamous among Wall Streeters for paying extremely high wages but =
expecting ALL of your time in return), developed an offspring of APL whic=
h used three-letter English-language combinations as the primitive functi=
ons and operators (i.e., the character represented by the Greek letter rh=
o, which looks a bit like a curly "p," would henceforth simply be represe=
nted by the string "rho"). The new language also, to some extent, altered=
the operators and primitive functions and their meanings. And the entire=
brokerage house invested all its programming capabilities in the new lan=
guage, and lo, they prospered (at least until greed at upper levels of ma=
nagement pretty much did them in).
>
>And lo, Iverson senior came back with J, a marvellous language which use=
d the same trick to remove the need for special keyboard symbols, also no=
rmalized operators and primitive functions, and generally was a fabulous =
language. Iverson won a Turing award for his earlier work. He called APL =
a "tool of thought," because (once you get your head around it), you can =
actually do very powerful things (that is, manipulate very large blocks o=
f any kind of data, including "nested" data, in extremely powerful ways) =
very, very quickly. There are actually the equivalent of "help screens" i=
n APL which, based on English-language requests you input (like, "separat=
e this mess into sentences and count the number of words in each and tell=
me the mean number of words/sentence), hand you "idioms": one-liners (or=
even a short string of characters) that replace pages of code written in=
other languages. And btw a good APL or J programmer can write a simple p=
rogram to do the above which would in fact be a one-liner... unless s/he =
was concerned to make it simple to follow, in which case it might take, o=
h, maybe 3 lines!
>
>
>Now here's the caution: Almost no-one programs anymore. Of those who do,=
almost no-one uses APL, or J, or has even heard of those languages. The =
higher-generation languages, mostly devoted to doing things with so-calle=
d databases -- themselves very limited -- are very limited in what they c=
an do (nothing near as flexible as J, or even something like FORTRAN).
>
>APL was an elegant, precise, powerful, general-purpose meta-tool. Many o=
ld APLer's still use home-grown WP, stat programs, and other tools they'v=
e developed over the years with this language. Iverson spent his life bet=
ting on the intelligence of human beings; he still (AFAIK) works up there=
in Toronto with kids in schools, teaching them how to think/program/expl=
ore. MS, in contrast, bet on the stupidity and laziness of human beings i=
n the U.S. of A. (Again, they were winning until greed at upper levels et=
c.)
>
>And now, the conclusion I draw from this sad little history: If we want =
wearables to be popular (thereby making them *affordable* for the rest of=
us), they have to run all current popular software, and do what the mass=
es want done now (but maybe look cooler as well as be more convenient). I=
f we design them primarily as tools for those of us who are intelligent, =
far-seeing, creative thinkers, they will never become popular. So the str=
ategy of those of us who ARE intelligent, creative, etc. should be to mak=
e them modular, make them adaptable, make them flexible.
>
>Eventually, as we feel our way into the right interface and the right to=
ols, one of us will come up with a killer-ap that will make the good inte=
rface, the good tool, turn into the popular tool and/or interface. In the=
meantime, to popularize an interface/tool, it would help just to have so=
me great gaming software, full of violence and sex, probably with some as=
pect of pornography thrown in.
>
>I'm not really cynical about humanity; in the long run I'm hopeful. I re=
cognize that intelligence and spiritual maturity are rare now, but I thin=
k they're generally considered attractive (as well as survival traits). (=
After all, _Buffy the Vampire Slayer_ is popular, right?)
>
>I think wisdom implies the maturity to recognize that we have to work wi=
th very imperfect humans, that we're very limited individuals. In the cur=
rent corporate culture of the U.S. especially, many are so pressed for ti=
me they feel they cannot learn a new interface before they get the next t=
ask accomplished, and the rest of us must respect that if we're to change=
it. We should use a stealth campaign (games?) to insinuate the new GUI, =
OS, program, or interface.
>
>Cheers --
>Carol
>
>
>
>--
>Subscription/unsubscription/info requests: send e-mail with subject of
>"subscribe", "unsubscribe", or "info" to 
>Wear-Hard Mailing List Archive (searchable): http://wearables.blu.org
>Please, *PLEASE* don't subscribe through a forward/expander/false domain=
>
> =20
>
--
Subscription/unsubscription/info requests: send e-mail with subject of
"subscribe", "unsubscribe", or "info" to
Wear-Hard Mailing List Archive (searchable): http://wearables.blu.org
Please, *PLEASE* don't subscribe through a forward/expander/false domain
From Wear-Hard Mailing list Archive (WH)
Maintained by R. Paul McCarty
Archive created with babymail