danbricklin.com/log
Home	Tablet PC: First Impressions
	Back in the early 1990's, I was heavily involved in the pen computing world. I co-founded Slate Corporation which developed application software for GO's Penpoint as well as Microsoft's Windows for Pen Computing and for the Apple Newton. I was exposed to software and hardware development, both at the OS and application level, and had experience using a wide variety of machines. With the release of the new Tablet PCs based upon Microsoft's new software, I felt it was appropriate for me to comment upon that, given my perspective. You will find here my comments based on my general feelings as well as actual experiences using various equipment. How I decided what to buy At various points in time, starting with Bill Gate's Comdex 2000 talk, I have been seeing public prototypes of Microsoft's Tablet PC software and the accompanying hardware. Most recently, I got to play with an Acer unit for a few minutes while waiting to get into Jeff Raike's talk at TechXNY (PC Expo). My Thinkpad was well over 3 years old, and I knew it was time for a new laptop. I decided to wait until the Tablet PCs became available, and probably get one of them. I had used a Grid Convertible (a machine from late 1992) for years after it came out, and knew that a convertible could be a fine laptop, even if you didn't take advantage of the pen much, and I needed a lightweight laptop. Now that the Tablet PCs are shipping, I decided it was time for me to buy. Since I was spending my own money and knew this would have to last me for a while, I looked carefully. Given my needs, I narrowed it down to the Acer, Compaq, and Toshiba. The Acer has a maximum of 256M of RAM, so I decided that wouldn't work for me in case I wanted to do any development work, or run some of the photo manipulation apps I'd probably end up using for my photography work. I was really torn between the Compaq and the Toshiba. From the specs, both looked like they'd meet my needs. The Compaq is supposed to have much better battery life, has a stalk as the keyboard resident pointing device, has more buttons, and is lighter when the keyboard is detached. Its pen, though, is not pressure sensitive. The Toshiba has a bigger and supposedly brighter screen (better for showing pictures and presentations, which I do a lot), faster processor, a touch pad (which I don't like as much as the stalk), and is in a more traditional form factor. Configured as I wanted them, with warranties, etc., they were close enough in price. I've had two Toshiba Portege's over the years, so I was more familiar with them, but I worked with Compaq on a similar, earlier machine, so I had some loyalty (and they ended up with Slate's assets when we closed it down). In the end, I couldn't find a Compaq to look at, but I found a Toshiba 3505 that was available at a CompUSA store in my area. After looking at it, I decided I might as well get the Toshiba. (I've presented the whole story of how I made my decision here not as an endorsement, but as an example of the type of thinking you might want to go through. I expect to spend time with some of the other units over the next few months, and will comment about them when I can.) My philosophical feelings about tablet computers The most important thing to know about the Tablet PC, as far as I'm concerned so far, is that Microsoft did a great job...of naming it. Much as the press wants to call it a "pen" computer, it is a Tablet computer. You must understand that. The basis of the machine is that it is (or can be turned into) a tablet. The pen is secondary, and not always important. I think they did the right thing in concentrating on the tablet aspect. Being a tablet means that it is much more mobile than laptops or desktops. You can do the things you do with a PC (read, web surf, email, etc.) in more situations (sitting without a desk, standing, etc.). The big change since earlier tablet computers like the Grid Convertible is that so many more people read so much more on a computer. PCs used to be mainly for composing, doing "what if?", etc. Now we use it for those applications, but even more we spend time reading (web, email, emailed documents) and quick communicating of simple stuff (IM). Another big change is that the main thing connected to a PC is not a printer, but rather all of computerdom, through LANs and the Internet. With 802.11, that connection can now be ubiquitous in more locations -- we are no longer tied to a particular wall connector. These factors increase the value of a tablet, and define its use more. When you read, you mainly select things on the screen (which email message to read, which links to follow, or which "favorites" to revisit), or scroll. When you compose, you are much more concerned with text. The pen is an obvious choice for an input device on a tablet. Since a tablet is often used in a horizontal position, and you can't be sure of a firm place to rest something like a mouse, a pen is appropriate for a pointing device. In a vertical orientation, like on a laptop, the pen isn't as appropriate for a pointing device as a mouse. In those cases we either connect an external mouse, or put up with even more limited pointing devices like touch pads or stalks. (I find the pen a much better pointing device than either a touch pad or stalk.) In both vertical and horizontal orientations, for any large amount of input of plain text, a keyboard (or in some cases dictation) is a very good solution. A mouse is a lousy text input device, and a touch pad even worse. A pen, though, is better than a mouse for text input, using either a touch keyboard on-screen or handwriting recognition. Handwriting recognition, though, is not the point of a pen, just like it isn't for a mouse. The pen is also much better than the mouse (or keyboard, touch pad, or stalk) for inputting graphical information, and has been well received in the graphic arts world for years. We've put up with all sorts of kludgy UI workarounds to "draw" with a mouse, and consequently rarely use drawings in informal electronic communications as much as we would in a room with a whiteboard. The pen opens up new areas for applications using drawing, of which "digital ink" is one. Of course, every new input/output device added standard to a PC opens up vast new areas, from laser printers to CD-ROMs to sound cards to modems. The pen will be no different. So, they are "pen" computers only in that being a tablet to some extent implies a pen. Like the touch pad and stalk in a traditional laptop, they are a reasonable compromise for doing the type of input you want to do with a PC. If you really wanted a "pen" computer, you'd probably want a desktop with a large dedicated writing surface like you find with the excellent pen tablets used by graphic artists. It's much easier to build a good pen system when it doesn't have to run through an electrically noisy color display, and it's much better to make a screen without protection layers for a pen. Impressions after using it I'm not going to do my normal job of showing pictures, etc., of the computer in operation, since you can find pretty good Flash and video presentations on the web sites of Microsoft and the manufacturers. I'll just comment on what I've encountered. I assume most readers here have read lots of reviews and other material about the machines and software. As I write this, I've had the Tablet PC for about a day or so. After I spend more time, I'll publish updated impressions. It's a geek magnet Like most any new device, but maybe more so, other people are very interested in seeing it. On the way home from buying it, I stopped to go to a Massachusetts Software and Internet Council committee meeting. When I explained that I was a bit late because I had just bought a Tablet PC, I was immediately kicked out of the meeting until I would go back to the car, get it out of the trunk, and return to show it to everybody. The machine was still in the sealed carton, so I had to open it up and show how the screen turned around and how light/heavy it was. (I didn't take the time from the meeting to turn it on and configure it, though.) Later that night, my friend Bob Frankston strongly encouraged me to come over and show it to him and let him play a bit (he's probably going to get one of the Tablets -- he was also at Slate for part of the time). Anybody I tell about it says the same thing: "Can you bring it over for me to see? Please?" Just a warning. You have to spend time learning how to set it up best for you It took some fooling around (and there's still more to be done) to learn how to set up my Tablet PC to work the way that fits me best. For example, the Toshiba had "hibernate" as the default for closing the case, and "power off" for the power switch when it's in battery powered mode. I had to change them both to "standby", so that I can wake up the Tablet in a few (about 5) seconds. I didn't want it to go into portrait mode when I switch to tablet, only on command, so I changed that setting. Inking problems with my chicken scratch The most interesting thing for me was discovering something about inking. Just like I think I remember with the Acer prototypes last June, the inking is a little slow for my writing. It lags behind my pen a bit. (I haven't had a chance to compare the effects on different implementations of the pens.) In addition, the ink seems "smoothed" to remove the jaggies from the digitizer or something. It just doesn't seem to sample fast enough for my writing. I write with a quick, jerky, "chicken scratch" style. I print, and almost never use cursive (ever since a teacher in 8th grade told me my handwriting was so bad that I should either print or type from now on for her). In addition, the amount of pressure at which the pen is set to be "touching" feels relatively arbitrary (though consistent). With my way of writing, the combination of smoothing and strict determination of "touching" makes my quick notes look completely different than with a pen, and very difficult for me (or the recognizer) to understand. With cursive writing, done a bit slowly and large for me, things are much better -- but I'm not at all used to cursive writing. For input, I find that handwriting recognition works OK, so I sometimes use it for input in Tablet mode, and sometimes use the on-screen keyboard (the keyboard is better for passwords...). The combination of cursive recognition and printing recognition is a nice step up from the machines I used in the old days. When I write in a way that my ink is readable (slowly and big), the recognition is surprisingly good, but not wonderful. Still, it sure beats writing with a mouse... For large amounts of text, I'll use the keyboard -- that's why I wanted a convertible. I was buying a new laptop, not replacing a keyboard. I found a partial solution to the weird ink, though, for note taking: I set the Journal application to use the pressure sensitive feature of the pen. Suddenly, the ink of my handwriting (even printing) looked much better. In fact, while I was trying out the new settings, my cell phone rang, and I had to quickly jot down a number to call, and then take notes during the resulting call. It worked great. I just had to write somewhat bigger than I normally did, but with pressure sensitivity making it do different thicknesses as I pressed down it was readable for me, and using a pen was a lot easier than holding a cell phone to my ear with one hand and typing with the other -- and I got to doodle to boot! The Tablet PC may not be better than paper for taking notes (ignoring the storage and searching features I haven't tried yet), but it's sure makes the PC a more useful device. Some note taking is much better than none. Also, I'm happy I have the Toshiba with its pressure sensitive pen, but I'm waiting to find out how to tune other pen parameters. (I tried downloading Wacom's new driver for Tablet PCs that lets you set sensitivity, etc., but I had problems getting it to calibrate correctly for some reason.) Buttons help reading and probably other things An important part of a tablet is being able to read, and an important part of reading on a computer screen is scrolling. The Tablet PCs have buttons you can push. The Compaq has a rocker switch, too. The Toshiba has basically 3 buttons, by default set to Up-Arrow, Down-Arrow, and Enter. This means when you're reading, you barely even need the pen. Very nice. I think buttons are real important, just like on PDAs. Remember, it's a Tablet PC, not a Pen PC... An interesting thing about reading: I've noticed how pervasive the RIM Blackberry has become with financial people like venture capitalists and analysts. They sit there in meetings, and every once in a while hold their RIM's in their laps and check their email. A Tablet PC with 802.11 or connection to cellular wireless with Bluetooth or its own cell phone PCMCIA card gives you an even better way to read real email and share what you've found with others in a meeting. Having a personal communications or data storage device you can read with on your lap with the right form factor is already successful (the RIM, PDAs). Here is a device with wider applicability and real Internet connectivity and lots of storage, with ink as a socially acceptable reply method/medium. The ability to turn the Tablet PC on or off in about 5 seconds with the push of a button is very helpful. (I think it takes a few more seconds to reconnect to 802.11, though.) Portrait mode is a win One of the properties of a tablet is being able to run in portrait as well as landscape mode. For reading on screen this can be very helpful. Much of what you read fits better on a machine the size of a pad of paper when it's in portrait mode, especially when you only have 1024x768 resolution (or is it 768x1024?). I received some email that I read over breakfast that included images of several fax pages. Reading it on my Toshiba in portrait mode was really great. I never needed to scroll, I just tapped the Next Page button every once in a while. Of course, for many regular computer applications and web pages, landscape mode works better. Being able to switch is important. On my big desktop machine, with an 18" 1280x1024 LCD display, I don't mind wasting the screen space when I read a portrait format page. With something I carry around, though, I don't want to have something any bigger than it has to be. General purpose machine If reading on screen is so important, why not just build an electronic book for reading? The answer is simple. You need to have a portable general purpose machine like a laptop anyway for composing, calculating, and running specialized applications. By the time you build a good enough "book" machine that can also connect to the Internet with whatever technique you have available (dial up, 100baseT, 802.11) and connect to the devices you'd like (USB), and be upgradable, etc., you're already spending enough for most of a laptop. It's silly to pay twice, so the more general laptop has always won out. It's only in the case of a completely different form factor, and a price down in the range of a software package or PC peripheral (which is what a Palm cost and was positioned as) that you'd buy both. By making the Tablet PC a full-fledged Windows machine, with access to all the normal peripherals and applications, you don't have that tension of needing to pay twice as much. How far have we come? Using the Toshiba (and remembering the little time on the Acer, and seeing the demos at conferences and on the web), Bob and I were both struck with how little advance there had been since the last try for pen computers in some respects. The pen/tablet software and hardware aspects appear just a bit better, especially given the huge increase in speed and capacity of today's computers vs. the ones of the early to mid-1990's (using the Grid Convertible as an example). Of course, making it work with color displays, and integrating things into full Windows XP, did take work, I assume. The advance in features, though, seems more like a "next release" or two of things rather than 10 years passing. This is not bad, though. Things were pretty good in the old days. The Grid got hammered for its black and white screen just as color became standard (B&W being necessary for some of the digitizers of the day), and there wasn't as much advantage to a tablet back then. As I pointed out above, the big change is the environment in which the PC works. There are finally lots of real reasons for tablet computers. Also, the hardware has improved enough where the cost in weight and price is little different than the amount we've repeatedly spent for other new additions to laptops during that timeframe: CD drives, bigger screens, wireless, etc.  Moore's law eventually brought those costs down to where they became standard. Tablet-ability is the latest in that long line of new capabilities. What's exciting to me, though, is that the way Microsoft is doing this will hopefully encourage tablet-centric innovation to start again throughout the hardware and software industry, so we'll continue these advances, and the rate of improvement will return to what it was in the early 1990's. The wide variations in Tablet PC form factors shows the start of that innovation. There will be trial and error to learn all sorts of important issues, from number and placement of buttons, to inking techniques, to new ways to take advantage of the pen and sound. Bottom line so far So, one day in, my verdict: I can't see ever buying a portable laptop that isn't a convertible -- the benefits are too great for me. It's a Tablet PC, not a Pen PC, and not a Clamshell PC, and that's a win. While these are clearly still basically a version 1 or 2, they are still very useful. If you read a lot on a PC, and move your laptop around a lot, and have benefited from 802.11, and don't mind using early software that works but is basic (like the original VisiCalc was), and are in the market for a new laptop, take the next step and move up to a tablet. Corporate evaluators must start learning about these systems, because as they improve and the price difference disappears, you'll have to figure out how to configure them, what type of software to insist upon, etc. If you always wanted to do your composing with a pen, and expect handwriting to be as reliable as a keyboard, stick with the keyboard, and wait for "handwriting computing" to happen, if it ever does. It's not that important. Tablet computing is. It will make reading on a computer even more pervasive. I think Microsoft and the hardware manufacturers who were willing to take a chance trying to advance the state of mainstream personal computing are to be commended for what they've done. - Dan Bricklin, 14 November 2002 In response to reactions to this essay, I've written another one looking at the reasoning behind some of my assertions. I explore a bit of the history of tablet hardware, as well as the state of application software for them in the early 1990's. I even link to a few patents to get detailed descriptions. See: "About Tablet Computing Old and New". - Dan Bricklin, 22 November 2002
	Home © Copyright 1999-2002 by Daniel Bricklin All Rights Reserved.