(keitai-l) Re: handwriting as Japanese input method for the keitai

From: Michael Turner <leap_at_gol.com>
Date: 01/02/02
Message-ID: <000f01c1934a$3d147b60$294fd8cb@phobos>
Wolfgang SLANY ( mailto:wsi@dbai.tuwien.ac.at) wonders....

> I wonder why Sharp, Sony, Casio etc do not put their excellent handwriting
> recognition software from their pda's into their keitais. Writing should
> be the most natural and fastest way to enter Japanese characters. I love
> it on my Zaurus since several years, and it's far superior to all other
> input methods I have used or still must use on other hardware. It should
> also be ideally suited for the small screen.

Some possible reasons:

1) Electronics - recognition can take a while, high-power CPUs in keitais
only add to their cost.  Stroke-based lookup requires more ROM as well.

2) Screen size - keitai screens are considerably smaller than palmtop
screens, and pen input requires an input area that could chew up 30-50% of
keitai screens.

3) Input speed -  I'd like to see a side-by-side comparison, but I suspect
that input time savings wouldn't be dramatic.

4) Convenience - you can thumbtype with one hand.  If you're hanging on a
subway strap, carrying a bag, or holding a coffee cup, you don't have the
two free hands that stylus input seems to require.  And then there's that
stylus -- yet another dangling thing to keep track of.

Hardware limitations, if any, should melt away in time.  After all, we have
handsets that play movies now.

The screen size problem might be mostly gotten around by having two screens
instead of one -- a lower screen for input in various modes (touchscreen
keypad emulation, and stylus input for both a full keyboard display and
stroke-based character input) plus an upper screen for display.  The whole
apparatus then still works nicely in clamshell format.  For monolithic
handset designs, there's still a cost justification for having a separate
input screen: it can be lower-resolution and greyscale.  Besides, it's
generally it's easier to get higher LCD yields on smaller areas, all other
things being equal; two LCD screens might be cheaper (even after interfacing
costs are considered) than one LCD screen of the same total area.

Actual speed of stylus input - even if it were slower, this might not matter
so much.  Japanese people are, after all, still learning how to write
Japanese characters in school, even if they are also rapidly forgetting how
to write many of them as kanji henkan becomes the norm for generating
Japanese text.

Finally, though, there's convenience.  Yes, Zaurus was seeing sales growth
here even when the Apple Newton was being laughed off the U.S. market, but
it was never really big, despite Japan's being arguably a far more
compelling cultural market for pen input.  For short messages, thumbtyping
is clearly not a big deal, and can be done in postures and circumstances
where stylus input is difficult.  For longer messages-- well, if you have to
write by hand, pen and paper still do just fine, if you're mainly
communicating with yourself anyway.  And it's in the nature of mobile
communications that long and discursive messages might as well wait.

In short, I think if pen input were THE way to do keitai text, Zaurus (or
someone like them) would have taken the lead in mobile handsets, long ago.
After all, look how long they've been around.

-michael turner
leap@gol.com
Received on Wed Jan 2 07:23:50 2002