The death of the UMPC? The MID won't do it.

There is no doubt an article published by TG Daily is going to attract a lot of attention in the next few days, especially since Slashdot has picked it up.  I could have predicted that would happen since it's a chance for Slashdotters to stick a sharp stick into the UMPC.  The article is interesting because at its core it is about the direction change that Intel is making with its MID (mobile internet device) line of components and devices the chipmaker recently began touting.  The TG Daily author gives a pretty good account why he thinks that UMPCs haven't taken off in the mainstream, touching on all the usual reasons like size, price, and usability among others.

The Intel MID devices are no doubt interesting, consisting of low-end hardware components to both keep the device cost down and provide better battery life.  It is definitely interesting that Intel is touting some flavor of Linux to run the MID devices.  To get a feel for whether this new class of device will hit the mainstream it's necessary to take a look at why the mainstream is not snapping up current UMPCs, since mainstream consumers has been the target market since the Origami was announced over a year ago.  The announced target price for UMPCs was pegged at $500, aiming the devices directly at the mainstream consumer and Intel is touting the same price point for the MID. 

I have been playing in the ultra-portable device arena as long as anyone and I don't think the Intel MID will be the mainstream device that every consumer is going to long for and buy.  We've had internet appliances before and the fact is that playing on the web is not the only thing that consumers will want to do with a device they carry with them all the time.  Heck, the Nokia N800 is a great internet appliance and retails for less than $400.  I don't see any of my neighbors running around with them and the MID will be no different.  You have to understand that I'm not talking about geeks like me when I talk about mainstream consumers.  I buy a lot of gadgets that most people wouldn't, but then I'm not the target market for these entry-level devices.  The OEMs desperately want to crack the mainstream consumer market who want to walk into Target and buy their device.  The truth is there are a couple of factors that will prevent this from happening.

Microsoft correctly realized back in the Origami/ UMPC early days that the portable computer would need to be able to run all the applications at a consumer's disposal.  This meant some flavor of Windows so users could interact with all their "stuff".  This will be a big negative factor for the MID because once consumers have one they will realize that unless they are a Linux guru they are not going to be able to interact with their data or media.  If nothing else, mainstream consumers will want to use the device with their existing media, and that means with iTunes and their iPods.  Look again at the Nokia N800, a Linux-based web appliance.  It's not limited in what it can do by the device itself, rather by what the average user can do with it and that's not much beyond the intended web usage which it does very well.  But that's not enough to make the N800 fly off the shelves, even at an attractive price point.  People will always want to use a computer with their favorite programs and using the data they have already created.  That's not going to happen with the Intel MID, I'm afraid.  So that dictates a "real" computer with hardware capable of running existing programs and operating systems, and that drives up the price dramatically while increasing the power consumption proportionally.  There's a reason the OQO costs $2,000.

The other factor that will hamper the mainstream adoption of a low-cost portable computer like the MID is connectivity.  Every analysis you see about the ultra-portable computer segment makes it clear that consumers expect to be always connected to the web.  This brings a big impediment to mainstream adoption because it not only increases the cost of the device to put the radios inside to provide this connectivity but it also brings a heavy ongoing price for the connectivity itself.  Even if you could buy a $500 MID device you'd have to shell out $20 - $60 per month for the 3G connectivity needed to make it truly useful as intended.  Mainstream consumers are going to balk at that right from the get-go.  I am convinced that is the reason that wireless device maker Nokia decided to go with WiFi only for the N800.  They know that selling the consumer on an expensive data plan will be hard to do while the public perception is that WiFi connectivity is free.  Of course, it's a far cry from "always connected", too.  This is a negative impact on adoption of mobile devices that is beyond the control of the OEMs like Intel and friends, and no particular class of device is going to change that.  Imagine Joe Public walking out of Target with his MID to find out that he's not really connected anywhere unless he ponies up the big bucks for a data plan.

UMPCs are not dead as the TG Daily article implies, even the author admits that.  But he is correct in pointing out that current 2nd generation UMPCs are not going to appeal to the mainstream consumer.  I don't think that will happen until there is a $500 fully functional PC with a cheap data plan.  It's all about anywhere access and price.  Two things that for the foreseeable future are mutually exclusive.  It might get interesting if one of the big wireless carriers produces a branded ultra-portable device with data service.  Maybe HTC and T-Mobile teaming up?  Now that would be interesting.

Comments

[Ctitanic]

I was reading the comments at the end of that article and found one saying "MID are good but I still need POWER" ;-)

[Ctitanic]

I think that that author is missing something. MID wont have ever the power to run good games. I believe that our UMPC has the potential to be perfect gaming machines in about a year. And that´s what will safe UMPCs. That's what will make UMPC popular machines. ;)

[James Kendrick]

I don't think gaming is going to be a factor at all because they will never run well on this type of device. Games require too much graphic power, are big battery drainers and there is no way to get the popular games onto the device. Dedicated portable gaming devices answer all of the above since they only need to do one thing and do it well. All-in-one computing devices will never do them as well.

[Julien Boyreau]

Hello,

Several Comments :
1) To succeed, "MID" need to REPLACE "phones" as the only device you have always with you.
2) "low end hardware", "power"...are TIME- Relative => At one day in the future, MID will quite power enough to do whatever "mainstream" users wants to do in the same logical path that makes Mainframe then Mini then Desktop niche markets.
3) MID are a step before a needed reinvention of IO systems to computers, beyond mouses, far-eye displays or keyboards to Augmented reality systems.

For me, two unknown :
Who will make it right ? Either the ones coming from PCs (Intel/MS/Apple/Dell...), the ones coming from "Phones" (TI/Nokia...), the ones coming from "Games" (Nintendo, Sony...) or the ones coming from "Web" (Google, Yahoo...)
When will they make it right ? 2009, 2015 or 2020 ?

[bob]

I think intel is doing exactly what it should do, breaking away from the microsoft vision of what these small computers are about. you have to realise that most of the people dont need the power , all they want is basic office , music/video, and internet where they can find a hotspot. There is a minority who want to use there mobile device for much more but they are the minority . if these devices are going to break through it isnt the power that is goign to matter its the price. give me soemthign in my price range that can do those things and i will buy it . once i have it i'll dual boot it with windows xp to run anything i want.

[Mike Cane]

Thanks, James, for writing that. Now I won't have to! As soon as I saw the MIDs, I knew they were dead in the water.

[James Kendrick]

I don't think most people will want a computer that replaces their phone. For one thing, any device big enough to be a serious computer is already too big. Smartphones are pretty powerful devices now but the screen's too small and input methods too dodgy for the average consumer.

[Chris EA]

I used to read "PDA's are dead" articles wirelessly on a Windows CE 1.1 powered clamshell Cassiopeia over GSM from my Nokia 6210 fed by an infra-red link.

Now I'm reading "UMPC's are dead" articles wirelessly on my Sony UX1XN over 3G from my Nokia E61 fed by a Bluetooth link.

My needs haven't changed: namely functionality as close to a desktop as possible in a convenient mobile and wireless package.

PDA's and mobile phones fit the bill the best back then, UMPC's and Smartphones currently fit the bill the best today. Tomorrow I'll be looking for the next solution that meets my needs better than my current set up.

[Kip]

The success of the UMPC or MID is really predicated on factors outside any marketing man's remit - such things as better battery life, the availabilty of widespread and cheap high-speed internet connectivity over wifi, and killer apps that make the these devices at least desirable if not must-have.

[Greg]

I believe the single biggest problem facing the mainstreaming of UMPCs or MIDs, for that matter, is the lack of availability via brick and mortar retailers. Currently there are numerous iterations of ultra mobile products available, online (and some of those are hard to actually locate and purchase). However, only very few are in stores available to examine up close and personal.

Most mainstreamers simply will not take the risk to purchase relatively expensive, new technology via the internet or over the phone.

After all, who really wants to buy new technology, whether it be cell phones, TVs, GPS, computers, etc., without a little fondling first?

Maybe some entrepreneur will open a "UMPCs Are Us".

[Sumocat]

"Microsoft correctly realized back in the Origami/ UMPC early days that the portable computer would need to be able to run all the applications at a consumer's disposal." --That is correct to an extent, but I still think they screwed up by not tailoring the OS for the form factor.

They should have optimized a version of Windows for the slower processor and limited battery life. Microsoft's own specs for Origami called for off-the-shelf components, which carried the expectation of lower prices, but also lower performance. It makes no sense for them to issue these specs then expect the device to run a version of Windows designed for a desktop PC. (performance and battery life)

Even if a UMPC is packing a Core 2 Duo and 4GB of memory, it is still constrained by limited screen space. Instead of just slapping a secondary interface over the full version of Windows, they should have supported the form factor with a customized version of Windows optimized for the small screen. (size and usability)

By not supplying an optimized OS, Microsoft failed to anticipate what have become the common criticisms of the UMPC. Thus, it was no surprise (at least to me) that Intel went looking elsewhere for a more optimal OS for the form factor. The UMPC is not dead, but Microsoft must do more to help it grow.

[Jack Furlong]

I looked at UMPCs, etc for a long time before I plunked down the cash for the Palm TX I carry now. It essentially does what UMPCs were intended to do, less expensively, and with longer battery life. Besides which, it fits in my pocket, which most UMPCs _won't_ I agree with you about the cost of connectivity; a monthly charge is what the average consumer _would_ balk at. But the high price point of the device doesn't help either. And yes, $500 _is_ too high for the "average person".

[Hanno Zulla]

Mobile phones used to be really expensive, business-only devices. Now they are everywhere.

It will be similar with having always connectivity to the web. It will come some time in the near future. There are several locations in the world where a full-time data plan is affordable already.

The N800 / 770 are not mainstream successes because they are meant to be experimental devices.

They did not target mainstream users, they were meant to build a geek community to draw from: Nokia released a relatively open, but knowingly incomplete product to attract open source developers. This has worked. There are already several apps that are better than those shipped with the N800 - and I guess that this was their plan.

Maemo is an experiment. Nokia invests in several such experiments at the moment and waits to see which of them succeeds. (e.g. they invest in three different web browser engines for mobile devices.)

Apparently, they are unsure wether the Smartphone strategy actually works. The N800 is a departure from Smartphones that are always too big for a phone but too limited for actual work. The N800 is smarter-than-your-average-smartphone web-browsing device that you can use with your wifi connection or through bluetooth on your dumb cellphone.

I agree with TG's article's main issue, though. Yes, the marketing promised way too much. I am still waiting for the sub-500$ x86-device capable of running XP while lasting more than a day on one battery. That was the Origami promise and they haven't delivered since then.

[Tax Man]

Gosh, who comes up with these acronyms anyway... don't they know that acronyms scare mainstream consumers? What is a MID? Is that like a WMD? Isn't that what people are mad at Bush about? IDK

LOL

[Nikoooo]

Hi James,

while I do agree with your first arguments about price and having all the software we like in our hand, I disagree with the "always connected" vision of yours.
Indeed I don't see this as a limiting factor as this is not yet available to the public. Indeed I only rarely see people surfing the net using their cellphone, although I am commuting every day to and from NYC! I guess rates remain to high. In consequence the common man has no idea he could be online all the time...
But as many in the marketing area will tell you, One has to create it market to be successful. Thus the day Google or other will create a broad-free-opened wifi, then your argument will be valuable...

Else I personally see only a limited futur in MIDs. I do need to input text, data efficiently... and what I saw so far with MIDs is... "access the internet, read your mail, look at your pictures and videos." Great but... incomplete.

[Nikoooo]

Regarding Handwriting recognition and the lack of it in MIDs...
The newest version of the Kohjinsha was bundled with Windows XP home edition. Guess what is one of the first complain of the testing people? No handwriting recognition!
Here: http://mykohjinsha.blogspot.com/

[Hitchhiker]

Long live the UMPC! I always find these negative reviews lacking in credibility. If the device weighs 2 pounds, the screen isn’t big enough; if the screen is big enough, it doesn’t have a keyboard or a built in DVD or 8 ½” floppy drive to accommodate legacy data from my PDP11… The CPU is too slow, the graphics card isn’t powerful enough. I am tired of reviewers that can’t seem to understand what a business computer is for. It is a tool, it gets valuable work done. The ability to use the same WinTel platform that has made personal computing a success combined with the form factor and ergonomics is why I my choice is and always will be an UMPC or tablet over any PDA or glorified PDA ever offered. Maybe a MID is just what the doctor ordered for some, and more power to you if it is! Diversity is a great thing! I am pleased that we have so many choices among computing devices. But to narrow minded reviewers I say, I’ll make my own choices thank you!

[Stephen Brandon]

I'm going to say something I thought I'd never say to jkontherun. You're falling into the same trap which most in the main stream have fallen into with the UMPC. You're looking for the integrated keyboard rather than seeing the machine (and potential) in front of you.

Online apps, such as the various services offered by google, are the killer apps of the internet tablet. They are already here. To run them you need a platform with the right robust browser. That's it. It's no longer about a robust OS, powerful hardware, and a large hard drive; it's about a hackable OS which is fairly transparent to the end user and a robust browser. Run your apps online. Store your info in the cloud. If there's enough machine to have these, if it can stay connected, if the price stays around the N800 $400 price point, then for most customers, it would be a done deal.

The Nokia N800 is almost there. It's problems include a lack of ability to print, the fact that it's installed browser has trouble with online apps, the fact it currently runs online video SLOWLY, and the fact it's limited to wi-fi. Skype and VOIP on the N800 is right around the corner, literally in the next months. You can install another browser on the N800 which allows you to connect to most online apps. Connect a BT keyboard like the Stowaway you use for your Q1 and a BT stereo headpiece, and connect to online apps and VOIP, and you've got a robust mobile office platform which is limited only by the online apps you can access.

The N800 isn't quit there yet, but that's what the Intel project seems to be about; and, the N800, as another commenter noted, is an experiment meant to build a programmer backing.

But I'm not writing this comment on a N800, I'm writing it on a UMPC. Why? Because it's got the robust browser now. Will the UMPC/Tablet die a quiet death? No. There are programs which will be slow to come to the net. But for the average, non-gaming, non-specialized consumer who wants a phone, mobile office, and digital content player, an internet tablet at $400 will soon be enough. If the choice is between a $700 UMPC and a $400 internet tablet, I know which way I'd go.

I'm keeping an eye on internet tablets because I did something this past semester I never did before. I went a semester without opening a MS product. I ran a hybrid section of English 102 completely online using blogs, docs, spreadsheets, gmail, and online project and personal management. In the process, I also moved my household planning onto google docs and spreadsheets, and I'm in the process of moving my GTD system to google notebooks.

Somewhere along the way I realized something, I'm no longer using the power of the UMPC, even the Q1P. More important, I don't need this much hardware, weight, etc. I'm using the UMPC's browser, and between hotspots, wi-fi on campus, and at home, it's been equal to everything I've asked of it--from planning a budget and watching TV to reading texts online and productively interacting with family and students.

So, the new Intel project doesn't support a full blown MS OS. To do most work and play, it doesn't need to.

[Stephen Brandon]

Oh. I should follow up. It's not that I haven't opened an MS product. It's I haven't opened an MS product outside of the Table PC OS. For word processing and working on spreadsheets off line, I've used OpenOffice. One of my points though is I've used wi-fi, and I've rarely been offline when I need access to my "stuff." With WiMax on the horizion and 3G connectivity likely to fall in price, I'm betting the incidence of being off line would be even less; and, if you're online and you don't need the bells and whistles of the newest Office, etc. you no longer need MS.

My opinion? I'd sell MS stock. It looks to me as if it's the 21st century equivalent of the 1980's IBM. It'll still be around. It'll still be a player, but it will no longer be *the* player for mobile offices.

[Chris K]

The question, really, is simple:

"How much power do you need?"

This is what kills me about the MID arguments. You have to remember that they're also making the platform a WinXP target, which is huge. A low-power Linux handheld doesn't have a lot of appeal to many people, since they can't easily take their MS Office docs with them. They might be a Flash version behind, too, and oftentimes, the Linux handhelds are *harder* to use, since the software choices are a little more scattered, and don't necessarily work from one Linux handheld to the next.

A MID running WinXP, however, would be enough to run the core apps. Office, IE, etc. Sure, you can't game on it, but that's hardly the point. You just need something with the OQO model 02's keyboard/mouse, plus a passive touchscreen, and something better than the OQO model 01's performance. It doesn't need to be a pocketable workstation, or even a shrunken-down laptop. It just needs to work according to typical user expectations.

Users can figure out a Blackberry, so the keyboard isn't an issue. They also figured out Palms and the like, so touchscreens are fine these days. They just need the software options, and they're all set.

XP isn't that much of a high-end OS. In this day and age, it shouldn't be that hard to get it running on low-power hardware, so long as you can throw some RAM at it.

[Aaron]

"A low-power Linux handheld doesn't have a lot of appeal to many people, since they can't easily take their MS Office docs with them."

You're a bit misinformed on that front. Open Office works on all platforms and handles all office docs quite well.

Ditto on the "If nothing else, mainstream consumers will want to use the device with their existing media, and that means with iTunes and their iPods." comment. iPods have been supported for quite some time on all linux platforms. I'm hard pressed to think of any linux media management app that doesn't support the ipod right out the box.

[ThoughtFix]

Remember this article? http://tabletblog.com/2007/04/linux-vs-windows-vs-mac-umpcs-who-wins.html

Having UMPCs on all major platforms will only serve to enhance market reception and adoption.

[Chris K]

Again, a *low-power* Linux handheld isn't going to do the trick. The Nokia 770 and N800 aren't exactly capable of running OpenOffice acceptably. AbiWord and Gnumeric, maybe, but it's not like these ship with the unit, or were made readily obvious to the common user. (Such things rarely are.)

*I* can install a dozen useful, usable applications on a Linux MID/Maemo system. That doesn't mean that the average consumer can, or even wants to. Many times, people want something that just *works* out of the box, without investing time and research into random package repositories and forums and installing hacks just to support a Bluetooth keyboard, and then playing with xmodmap just to make it work right.

When you're looking for a productivity tool, the last thing you need to do is spend hours tweaking it to make you more productive. That's a case of the machine using you, not you using the machine. This doesn't mean that Windows is better, or easier to admin properly, but it is easier for a Windows user (which represents the majority of computer users) to take advantage of that system, than to re-learn everything all over again.

[Woadan]

OK, first, James: Thank you for a well-thought-out and unusually long article/post.

To all of you Linux/OpenOffice fanboys: OpenOffice does not do all that Office does, and if you have to do things with the government, they want it in their templated format. Can you guarantee OO won't mess with that? No? Pass on OO then. Also, let's not discount the emotional issues involved. When you have years of experience in a certain interface, you don't like the change. Everybody here has probably heard about the outcry against the MS Ofc 2007 "ribbon", yes? OO has a different interface, and if users don't want to learn MS' new one for Office, they sure as heck don't want to learn OOs. So, "does most of" isn't a good answer.

Many of the repleis/comments seem to overlook James' comments on the cellular/mobile coverage. Wireless (WiFi or WiMax or whatever flavor) isn't always giong to be available, nor will it always be wise to use it. Government operators, whether the gov agency itself, or a contractor, aren't going to be too keen on it because it is a security risk. That can't be discounted, or at least shouldn't be. Adoption by government agencies/contractors could be key to widespread adoption I would counsel. Those same agencies have no qualms about using Cingular, T-Mobile, Verizon Wireless and/or BlackBerry/RIM, so even if no more secure, they're acceptible. Cellular/mobile service included would be key, too.

Unfortunately the wireless carriers are always fee-hingry, just like their copper/landline-based brethren have always been. (Verizon and AT&T make no money off the landline service itself- they make huge profits on the voicemail, DSL, and other services included with the dial-tone.) And whether in their landline or cellular skins, none of the carriers want to become "dumb pipes". Cingular proxies their PDA-and smart-phones to their network for when selling MP3s/ringtones and the like. This sacrifices any other web surfing unless you disable that proxy. (I know this from experience.) I expect the other companies do, too.

Tech savvy people like those of us who frequent tech blogs like jkontherun probably are smart enough to be able to figure out we can copy our songs to a memory card, and play them from there. And to catch snippits for ring tones as well. But the average user just blithely uses their carriers services because it's easier, even if more expensive. And the RIAA is also happy when you buy the song, and then paya again for part of it when you buy a ringtone.

All of the cellular carriers control the devices, for the most part, that make it to their networks, so they help prevent being a dumb pipe for cellular data. They don't control the devices that connect to their landline services (OK, Sprint and TMobile don't have them.) And they are much more liely to be dumb pipes there.

Do we really want to see a UMPC running Windows XP or Vista with tablet capabilities being hampered in ways similar to Windows Mobile devices often have been by the wrieless carriers? I know _I_ don't. But I also know that would be one way to bring down the price if coupled with a 2-year contract.

Frankly, we know that the cellular carriers, especially Verizon, seem to hate when you use their data service "exessively". So even though their service is nearly essential (as I see it any way), we also know it is problematic. Additionally, at the prices they charge ($50-80/month for an unlimited plan), when you get slower than DSL speeds, well, it boglles the mind that we pay for it. (But we do.)

That Origami/UMPCs have done as well as they have is something of a suprise when you consider all the wrong turns that MS and the makers have made along the way.

Woadan

[Stephen Brandon]

Woadan,

I'm not sure I agree with your reasoning concerning MS office alternatives like google docs or OpenOffice. While you are right, the alternatives don't (currently) have all the bells and whistles of MS Office, most folks use only a portion of MS Office features, and they'll rarely need or miss those which aren't there. The price is certainly better for the alternatives. Moreover, most consumers don't work for the government, and I'm pretty sure institutions as conservative as government are not the right measure of a consumer platform's success. Isn't it government IT who is proposing putting biometric security on DAPs?

Price, ease of use, connectivity, and utility are better measures for platform success. Your point concerning security of wifi, etc. falls in the same category. Most end users don't (and, I'd argue, shouldn't) have the same level of security concerns as government or enterprise.

But for right now, you're probably right in your main point. Until the price and ubiquity of connectivity hit some unknown sweet spot, folks will continue to think and work via the desktop paradigm. They'll hug to bloated platforms and applications. My guess is such traditional government and field enterprise platform needs will be with us for some time, and they'll dictate to many what to look for in a platform. Corporate and government IT will be much more comfortable with a known, desktop paradigm as the world shifts under them.

End users can afford and invite more of a gamble, and that sweet spot is close, very close. For those who move between hotspots and home and work networks, that sweet spot is within a year or two. Some users, more comfortable with a higher price and more willing to pay the price to be tech savvy, are already enjoying what everyone will soon enjoy. The N800 and MID could well be the early adopter version of where the new paradigm begins to make sense to many end users. I'm betting something like the MID project will succeed.

What's cool is James has obviously hit a nerve and uncovered a growing divide among his readers in terms of which paradigm mobile computing will follow. Just look at the number of comments this post has brought compared to most.

One thing studying history has taught me: paradigms interpenatrate. There isn't a point in history where one can say, this idea starts and this one ends. The old keeps going alongside the new; indeed, the new is usually an attempt to replicate the old in some better form. That's why cars originally looked like wagons, and wagons appear alongside of cars in old photos. Drivers of both probably didn't realize the Model T represented a changed world.

I'm just not sure James is giving enough credit to the potential of something like the MID project to create a wide acceptance of a sea change which is already happening. We don't, for instance, hear much about the impact of Web 2.0 (lord, I hate that label)on mobile computing on jkontherun.

That is not, by the way, a complaint. I do appreciate the work and thought both Keven and James bring to this blog.