|
|
|
First Look at Tablet PCAcer TravelMate TM100 with Windows XP Tablet PC Edition (July 2002 issue) [Click here for a pdf version of the Pen Computing feature]
Shingle Springs, CA, June 25, 2002 -- On June 4th and 5th, a small group of journalists
participated at a Microsoft Tablet PC Reviewer's Conference in Seattle. Although
Microsoft has been promoting the Tablet PC concept for almost two years, this was
the first time that the press was given working Tablet PC prototypes with beta
versions of the Tablet PC software. The actual hardware/software combination
consisted of an Acer TravelMate TM100 "convertible" notebook loaded with Windows
XP Tablet PC Edition.
Pen Computing has reported in detail on the Tablet PC concept in the July 2001
issue and then again in a special Tablet PC insert in our May 2002 issue. During
that time the Tablet PC has undergone many changes and revisions. What initially
began, at least in my opinion, as another attempt to create a true slate with an
optional external keyboard morphed into something different, a mobile computing
device that may come in different forms and shapes, and whose primary focus is
the use of "ink" as a new data type. There has been significant discussion among
the staff of Pen Computing about the meaning of this shift. I personally believe
that Bill Gates really did want a "pure" slate and that it was the
practical-minded factions within Microsoft, and perhaps resistance among hardware
partners, that eventually morphed the concept towards a more conventional form
factor. One interpretation of this morphing is that, given the past failures of
pen slates, the teams got cold feet and chose to play it safe, resisting efforts
to create a true pen slate. Another interpretation is that Microsoft's Tablet PC
group concluded that the project was really about adding ink and the pen to the
mainstream computing arsenal while leaving the decision about form factors up to
OEMs and the buying public.
Be that as it may, we now have had a chance to take the "Tablet PC" for a first
test drive. Thanks to the move to a form factor-agnostic approach, the term
"tablet" is definitely relative as the little Acer TravelMate notebook is not a
tablet but a "convertible" that can function both as a standard notebook and also
as a pen tablet. It accomplishes this transformation with a display that swivels
180 degrees, then folds down flat on top of the keyboard, LCD side up. While the
Acer is a competent piece of hardware, this article is not a review of the little
transformation artist. (Click to see a sidebar piece on the Acer). Rather, it
describes my first impressions of the look and feel of Microsoft's "Tablet PC,"
or to state it one more time, the Microsoft Windows XP Tablet PC Edition running
on a piece of hardware that represents one of several form factors designed for
this extra-feature version of Windows XP. The inclusion of notebook form factors
has definitely made it more difficult to figure out what to call the combo. After
all, I am not calling my IBM ThinkPad a "Pro PC" just because it is running
Windows XP Professional.
To put it bluntly, those of us who have been around the pen technology scene for
a while may have a hard time not comparing the XP Tablet Edition to Pen Services
I (1991) and Pen Services II (1995). Microsoft claims much better integration
this time around. That may well be so, and the Tablet PC edition deserves an
unbiased evaluation. So here is what you can expect from a computer running the
new version of XP.
So let me tell you what's actually different in the Tablet PC edition. You start
with standard Windows XP Professional. On top of that you get an Input Panel that
toggles between an on-screen keyboard and a handwriting input area, Windows
Journal to work with ink, and the ability to use ink in various Windows
applications. In addition, you also get speech input, which we'll get into in a
future article.
It is important to realize that Journal is meant to be a notepad for electronic
ink and not an ink processor. Predecessor technology to Journal did use the ink
processor approach where words would wrap and where you could place the cursor
between words and insert more ink. Microsoft deliberately departed from that
approach in favor of a more paper-like interpretation of ink--ink that stays put
unless we erase it. That is not to say that you cannot manipulate the ink to some
extent. You can insert space between lines of ink text. You can also scale ink
and change its color. Finally, you can have ink recognized and converted to text
that then either replaces the ink or is sent to the clipboard.
Do use the recognition feature, you write something, then select it and use the
"Convert Handwriting to Text..." You then see what the recognition result is. If
something is wrong, you click on the word. A very small green angle shows up. Tap
on it and a list of alternates shows up. If none applies, you can either delete
the word or rewrite/respeak it. At first sight, the Text Correction panel seems
to lack some simple edit functionality, like erasing or adding words and such.
However, you quickly find that you can actually edit in several ways: via the
Acer's physical keyboard, via the Input Panel both in recognition and keyboard
modes, and via write-anywhere recognition. Correcting in ink is a different
issue. You cannot add horizontal space between two words in ink (as you did in
the predecessor technologies), so you cannot simply place the cursor between two
ink words and add a word. You can, however, select ink and move it, then insert
new ink. There is no doubt in my mind that Journal may be a revelation to some
people, especially when used in conjunction with the utility's ink search and
sorting features.
The Input Panel works either in pen or keyboard mode and is heavily customizable
(see screenshot). You can play with basics such as pen width and color, size of
the input window, number of lines for data input, time between lifting of the pen
and the start of recognition, and so on.
The much maligned handwriting recognition works amazingly well. I was not
initially a fan of the Input Panel method, especially when it looked like this
was going to be the only way to use the recognition engine. The final version
does allow writing into the Input Panel, write anywhere, or both at once. "Write
anywhere" actually means writing inside a box that shows up, but you can
alternate between anywhere and writing in the input panel. That box can be a bit
annoying as it sits on top of whatever else is on the screen. Go outside that box
and there is no recognition. Another thing that is a bit distracting is a
horizontal line that acts as a guide to help you write horizontally, thus making
the recognizer's job easier. I wish you could turn those lines off. My
handwriting is not particularly nice and I do mix printed and cursive. I do,
however, follow a few basic common sense rules for handwriting recognition. I was
easily able to write 20 words or so all over the screen and it recognized every
single one of them.
To be honest, given handwriting recognition's history, I was puzzled that
Microsoft wasn't better prepared in their demo. I did not expect perfection at this still relatively
early stage, but I also did not expect them to walk into landmines they way they did, did not
expect numeric fields to return alpha characters, and remain perplexed why the company chose
not to include the Shape Selector feature that can greatly increase recognition accuracy. That
feature has come with every version of CalliGrapher/Transcriber ever shipped, and why Microsoft chose
to remove it is a mystery.
All of that said, I must say that the recognition is the best I have ever seen, and
that the recognizer worked extremely well for me. What it boils down to is that
in order to make a recognizer work for you you must learn how to use it. A bit of
practice and observing a few simple rules is all it takes to greatly increase
accuracy. Of course, if you simply want to prove that it does not work, then it
won't.
By far the most important question is what the "Tablet PC" will eventually
become. At this point it is almost certain that the majority of Tablet PCs won't
be slates, but notebooks with a degree of pen functionality. None of the
ink-enabled software I've seen so far is compelling enough to qualify as a
"killer app." It is entirely possible that there is no Tablet PC killer app, and
that the success (or failure) of the concept will depend on a combination of
factors and abilities that,when combined, can make for a more compelling user
experience. Some of those factors and abilities are ink, low weight, long battery
life, no heat build-up, quick and reliable standby, public acceptance and
software support, ease of integration, low cost premium, the evolution of
wireless LANs, discovery of untapped markets. You get the idea.
It's simply too early to tell if this is a great new thing or just Pen Extensions
3.0.
|
|
|
[Homepage] All contents ©1995-2001 Pen Computing Magazine, Inc. All rights reserved. |