The Future PC as a Set of Pens?
Strudelkugel writes "The Wave Report covers a concept PC that NEC is working on, called P-ISM. (Maybe the name doesn't work, but it looks cool.) The design concept uses five different pens to make a computer. One pen is a CPU, another a camera, one creates a virtual keyboard, another projects the visual output and thus the display and another a communicator (a phone). All five pens can rest in a holding block which recharges the batteries and holds the mass storage. Each pen communicates wirelessly with the others."
As often as I lose pens...
ESCAPE POD - The Science Fiction Podcast Magazine
Think how easily supplies disappear from the workplace..... this just makes it easer to fit a "box" of computers in your briefcase/purse. Clemster.
True. Unfortunately, people have been trying to develop good speech recognition for years, and it is still a long way from being viable as the main input method.
If I seem short sighted, it is because I stand on the shoulders of midgets
> what does a virtual keyboard have to do with a real piano vs a midi keyboard
Both a virtual keyboard and a midikeyboard are shitty versions of what they're based on - they both have feedback mechanisms - which let you know which key you've pressed and how hard you've pressed it - which are simply not up to the job of serious typing or piano playing.
Of course, there could be some sort of storage in the CPU pen, but not that much, surely.
Mod parent up!
Without tactile feedback it is impossible to use a keyboard effectively without looking at it. I can detect typos by the feel of the keys, I know when I hit the wrong one. without tactile feedback, I'll be looking at what I'm doing rather than what I'm typing from. That slows me down a lot.
No-feedback keyboards have a place, but not for general use.
Speech recognition will never replace the keyboard completly.
Take any action you do on a computer.
Now figure out a way to SPEAK that action, without any ambiguity. Now compare how long it takes to SAY that, with how long it takes to do via a keyboard or mouse.
Virtually every action take longer to explain than to simply DO. And the only way to have voice recognition come anywhere close is to have the computer try and guess what you mean. I don't know about anyone else, but I prefer it doing what I tell it to do, even if I occasionally make mistakes. The last thing I want is the computer misinterpreting something I said through no fault of my own.
As for simple text input, the biggest draw for voice recognition is for people who have bad or nonexistant typing skills. It is a valid issue, however it is one that as time passes becomes less and less important. Keyboard use has become ubiquitous among the young today, who will eventually push out the older people who never developed thier typing ability.
Also, can you imagine programming, or writing out abbreviations, or non-words with voice recognition?
int main left parenthesis int argc comma char asterisk asterisk argv right parenthesis left curly bracket printf left parenthesis doublequotes Hello World backslash n right parenthesis semi-colon right curly bracket
Frankly, speech recognition has some good uses for AUGMENTING input, but it's useless as a keyboard replacement for nearly every usage area.
I'm surprised at this. I figured more of the /. crowd was used to typing while looking at the screen. Isn't that the *best* way to know you typed what you meant? Don't you need to know that the *computer* knows that you meant to call the function certainFunction() instead of reference the array cwrtinFunctin[]?
/. posts...
Just typing this, I must have hit the 'i' key 4 or 5 times incorrectly, but you'll notice no typos after I hit "Submit" because I can see what comes up on the screen in *realtime* rather than stare at the keys until I'm done and then search for the typo in the paragraph-stack (I'm a good speller always, but a bad punner on Monday mornings). Then again, this does explain the massive amount of errors in
Much love all the same!
Emacs: for people who just never know when to