It’s a great time to be a handheld PC enthusiast, not because there are so many good devices to choose from, because I don’t think there are, but because public awareness of handheld PCs are at an all-time high. Consumers have to be aware of the genre, understand the utility of it, and actually want one before the handheld will ever take off like I think it could. Just yesterday I was using the HTC Advantage in a coffee shop and the waitress came over to the table and asked me if it was one of those "OQO computers". The fact that it wasn’t doesn’t matter, it’s that a "regular" person knew what an OQO was and felt that it would be useful for her. What would be really cool is if there was a handheld computer that really worked the way it should.
What’s a handheld computer need to do?
I should clarify my earlier statement that I don’t think there are a lot of good handheld PC devices to choose from. There are in fact quite a few devices with more being released all the time. It’s just that I don’t think they are as good as they need to be to do their intended job. Most all of them being sold today are based on Windows, and that’s a dead-end road. Not just Vista, which is the pits for mobile devices, but XP too. Why? A number of reasons. Windows is too big and too resource-intensive. Mobile computers will never do well with a bunch of things running in the background. It hits performance and battery life too much and handheld devices need all day batteries, or at least most of the day. They will never take off en masse until consumers feel comfortable that they can use them all day without charging. Let’s face it, devices used in the hands are not very usable if they have to be plugged in a lot. Another reason why Windows is not a good platform for handheld devices is that security concerns means Windows needs to be updated regularly. That just doesn’t work with handhelds that need to come out of the box ready to go and keep working that way indefinitely. Wouldn’t you be so much happier if your device never needed updating? OK, not never but seldom. You bet you would and you’d get your work done without unnecessary delays like system updates, especially those updates that break something.
Performance needs to be adequate too, no matter what the user is doing. The core usage scenario for handheld mobile PCs is pretty simple. Users want to check email, surf the web, and work with documents. This isn’t all that comprehensive a scenario but it is essential that handheld computers do this and do it flawlessly. That means without errors, without system slow-downs (or stoppages), and without killing the battery too fast. These particular uses are what people are doing maybe 90% of the time they are mobile. Sure they do other things too and these devices need to do those things too but if you don’t get the core usage down properly then you are just spinning your wheels. One thing is certain- you have to do these core things without compromise or you fail to meet customer expectations.
How can handhelds meet customer needs?
So we’ve looked at the core usage requirements, what’s the best way to do them with a handheld device? This may raise a few eyebrows but the best way to do the core functions is with an embedded operating system. Yep, you heard me. Windows isn’t working well on mobile PCs so that’s out. We have Linux gaining ground, especially with Intel pushing their Mobile Internet Device (MID) based on Linux. There’s a false sense of accomplishment when people discuss Linux-based devices because we invariably think that open source= good. It does but it’s not much different than Windows in this scenario. While variants of Linux can be created to run "softer" than Windows, a version of Linux for mainstream handheld PC usage must be airtight simple for the end-user. I haven’t seen a version of Linux yet that didn’t require a good amount of tweaking to get simple things done, and that will not work on these types of devices. If you don’t believe me then you haven’t been watching the handheld space like I have. We’ve had Linux-based PDAs and Internet appliances for years but they don’t catch on because most users can’t figure out how to make them do what they want. For Linux to be a good handheld PC platform like the Intel MID, they must be airtight out of the box.
So maybe Intel will do this, right? They can’t because they do not make the hardware. This has been Microsoft’s problem forever with both Windows and Windows Mobile and it will be Intel’s problem too. They can build the OS kernel as good as they can but since most of the hardware will be built, installed and sold by someone else we’ll have the same mish-mash of hardware support and driver instability and all of the other bad things we’ve become all too familiar with. Handheld computers must be rock-solid stable through the work day and the only way for a company to produce such a device is to control all parts of the design, including hardware, operating system, applications and specifications. That’s a pretty big limiting factor.
Why am I so sure that the OS needs to be embedded? That’s because a real handheld PC needs to be instant-on in the field. Standby and resume schemes that are used currently are just not good enough to provide the usage experience that is needed in a handheld PC. It is imperative that there be no waiting for the device to wake up and get started and an embedded OS is the best way to do so. Put the OS kernel in ROM so that the working system is always there when needed and using ROM has security benefits as malware cannot change it with bad intent. Operating systems in ROM run faster on lower capable hardware and this is exactly what is needed for a good handheld PC. Having the OS in ROM would also allow the handheld to start reaching the critical all-day battery life.
The solid state disk (SSD) can play a role here too for data storage but not for the OS itself. We need the OS to be a part of the device as it would be embedded, and not installed on what appears to be a hard drive. Sure SSDs are memory and not hard drives but systems today see them as drives and they can be erased or compromised by mal intent. No, I am convinced that embedded is the way to go.
So who will make the first "real" handheld computer?
Apple. There are very few companies who can design and produce the handheld PC like I’ve indicated in this article, in fact there are only two. Microsoft is one and I believe that if the Redmond giant ever took on this task they would produce a killer handheld PC. Let’s face it, they are operating system experts and they control the office document space with Office so they could integrate this into a handheld in a dynamite way. I’m talking embedded Office to go with the embedded OS on our super handheld. That opens so many exciting possibilities that it makes my pulse race when I think of all of that potential. Unfortunately Microsoft doesn’t make hardware more complicated than a mouse or keyboard and they’re not going to make any type of "WinHand" device. They have too many anti-trust worries whenever they take on something like this and I don’t think they’d do it.
That’s why it has to be Apple. Cupertino now has the hardware design, communications experience, OS kernel expertise to produce an awesome handheld computer. I am talking about a current-day Newton handheld, but one that is a full computer. The form is not that important as long as it’s small enough to be used in the hands. Apple can leverage its technologies from the MacBook, iPod, iPhone and iTunes to build a fantastic handheld PC. It would work out of the box like Apple’s other products because they control every aspect of the device from design to production. Apple could also bring virtualization to the table, so even if they developed an embedded OS that is based on Mac OS X they could leverage the ability to run other OSes if the consumer wished. No other company can produce a product like this and that’s why Apple gets my vote for the producer of the first "real" handheld computer. Let’s hope Steve Jobs reads jkOnTheRun because I challenge him and Apple to do one.
You have a point here
You just described the device I’m waiting for.
Currently using HTC Advantage, it’s good but not good enough
Thinking of getting either an OQO 02 or the HTC Shift (whenever it will be released) to replace it, but then again….they run Vista. And the battery isn’t good enough so not sure it will be an upgrade from the Advantage
So why don’t you just give Mr Jobs a call and tell him what we want for christmas
Well, yeah, except that Jobs is the one who killed the (brilliant, exemplary) Newton, and explicitly said that no one would be interested in a “scribble pad”. That was his position years ago, and it seems he’s sticking to it. The iPhone is not a computer, it’s a locked-down browsing, messaging, music device; the focus is not on input.
So the “of course” here is wishful thinking, not likelihood.
Good article James. I also think Jeff Hawkins and crew over at Palm have a chance at creating a hanheld like you described. I know the Foleo never made it to market, but they are getting there. As a old Handspring Visor user/fan, I think Mr. Hawkins does think about the user experience and the fact that Palm stopped the Foleo for now shows us they are listening to the users. I think the major players in the next great handheld computer are Apple and Palm.
Some good things said there.
What does ‘work with documents’ mean? Read? WRite? Office? PDF’s? E-books? Google Docs? Maps?
What do you mean by ‘surf the web’ FIE? Plugins? Inline media? 5 seconds to open gmail or 10 seconds (which kind of negates instant-on.)
Consumers need myspace compatibility? YouTube (Needs plugins) etc etc.
These are the sort of things that no embedded software/RISC based device has been able to do yet. Do you think effort needs to go into RISC or X86? Would Moorestown be good enough. That will give you your battery life.
I Enjoyed the post. THanks
Steve.
yep, it would be very shiny and sleek..
.. and there wouldn’t be any tweaking or hard to do settings.. as they would criple all hw and sw as “they know better what user needs”
no thanks,
I rather do it slower that not at all.
JK,
I definitely agree with you on what a handheld needs to do and all that. I disagree on Apple’s ability to produce that.
As one commenter above pointed out, Jobs has been consistent in not producing truly complete mobile devices. Both the iPod and iPhone are locked down to ONE interface, one experience. That’s not saying Apple’s experience is a bad one (arguably), but that limitation is precisely what has plagued Apple since the founding of the company. When it comes down to it, most people don’t want to be locked into one company for one experience. (Why else do so many complain about Microsoft antitrust activities?)
The nice thing about Microsoft is that despite antitrust claims, there’s a lot of flexibility and options on the platform. Even simple tasks within the OS offer multiple ways to get things done, but it goes beyond that to the compatibility with other products. With Apple it’s their hardware, their OS, their interface, their software, and until they move beyond that mindset, I don’t see their luck changing.
–Aaron
Steve, I mean work with all the documents you mentioned and more if needed. The mobile worker needs to create, edit, and reference all typical documents received and used in a work setting. By web browsing I mean complete experience, Flash, YouTube, etc.
Apple has the pieces to do all of this already.
I should state that I think Microsoft could make a better handheld than Apple for many of the reasons people have stated here. I just believe that is not possible for them to do which is a shame.
Hello everyone,
Very good article.
Three comments:
The “handheld” feature is just one part of the equation. You need to add “pocketable” as I think there is NO MARKET between the 4” and less pocketable and the 10” and more “baggable”.
Nokia is a better candidate than Apple because it does not have a computer business to protect.
The very one feature lacking in Nokias is to be able to link wirelessly to better IO when available.
I would definitely consider an Apple handheld, but that is if it didnt have an Achilles heel.
Seems like most of Apple’s latest products have been beautiful, but have Achilles heels. iphone with no 3g. Touch with no bluetooth. Just to name a very few, there are MANY more Achilles heels.
But if they could make something with the touchscreen of the iphone, with a OPEN OSx, long battery life, and an intuitive interface I would certainly pick it up. They are extremely late to come at least to the tablet party, so I have serious doubts if they will consider the UMPC party though.
Very well said, I agree with almost every point here. I wish Microsoft would sit down and design a real touch OS. No I don’t mean Tablet features taced onto their current OSs. I mean something designed for a computer that can be used 100% without a mouse if needed.