Incremental Blogger

Let’s face it, the OEM community has failed to deliver a knock out mobile Internet device. Microsoft didn’t deliver with the UMPC and its sleek Haiku concept. Even Apple has been unable to deliver a larger version of the iPhone–the iTablet that many have dreamed of. And come to think of it, Intel is poised to seed this market with its MID initiative and from what I’ve seen so far I’m skeptical they’ll hit the sweet spot too–despite its fortunate timing.

But no one, no one, has delivered what quite a few people are looking for: An inexpensive, portable, Internet-tuned device that’s large enough to use on the web and light enough to want to keep using.

Of all the devices I’ve owned over the years, the iPhone is the closest, but frankly its display is too small.

So what’s a person to do? How about try designing your own? That’s exactly what Michael Arrington and the folks over at TechCrunchIt have in mind.

Michael writes:

I’m tired of waiting - I want a dead simple and dirt cheap touch screen web tablet to surf the web. Nothing fancy like the Dell latitude XT, which costs $2,500. Just a Macbook Air-thin touch screen machine that runs Firefox and possibly Skype on top of a Linux kernel. It doesn’t exist today, and as far as we can tell no one is creating one. So let’s design it, build a few and then open source the specs so anyone can create them.

Here’s the basic idea: The machine is as thin as possible, runs low end hardware and has a single button for powering it on and off, headphone jacks, a built in camera for video, low end speakers, and a microphone. It will have Wifi, maybe one USB port, a built in battery, half a Gigabyte of RAM, a 4-Gigabyte solid state hard drive. Data input is primarily through an iPhone-like touch screen keyboard. It runs on linux and Firefox. It would be great to have it be built entirely on open source hardware, but including Skype for VOIP and video calls may be a nice touch, too.

As I’ve blogged for awhile, I’m all for an idea like this and frankly since no one has been able to put it together I agree it’s time to look elsewhere. Is the community the right place? We’ll see. Even if a good reference design is all that comes out of this effort, I’m all for it. We need something.

It sure looks like I’m not alone in this. Since the first post earlier this afternoon there are upwards of 700 comments sprinkled about on serveral posts about this idea. Yep. People are interested.

I think there are lots of existing players here to blame as to why we haven’t seen something like this. Some of this is due to the sluggish momentum of the large players. For instance, Microsoft has been unable to deliver an OS that makes sense for devices like this. Intel hasn’t delivered on the performance/cost side for low-cost mobile devices. And the OEMS. Don’t get me started. Even on the sales side we’ve seen the Best Buyification of notebooks to the point that they’re all pretty much alike. Even Apple can’t keep up–particularly with all of the calls for a larger iPhone-like Tablet.

Will an open project like this succeed? Who knows. Cost is going to be the biggest challenge I imagine. Even if TechCrunch were to get commitments from say 5,000 people online to buy a first gen device, I doubt any of the manufacturers will give the price breaks they need to reach the $200 price most of us want. Why should they? In no time flat it would make all the over priced UMPCs and similar devices look, well, over priced.

I’m game though. Sounds like a fun project to try. Sorry Intel. Sorry Apple. We’re tired of waiting.

Wouldn’t it be funny though if Apple announced something like this in the days to come? Yep. When something is achievable for one, it’s achievable for all.

This InfoWorld article makes it sound like Intel is going to take a wait and see attitude when it comes to marketing MIDs. InfoWorld suggests that at Computex this past week there were several hit, low-cost PCs using Intel’s forthcoming Atom processors, but there wasn’t the same thing for Atom-natural MIDs.

Executive Vice President and General Manager of Intel Sales and Marketing Group Sean Maloney takes a wait and see attitude with respect to MIDs, “”By the end of this year, you will have seen a whole bunch of new MIDs coming out and we’ll see which ones are hits.”

Excuse me. There already is a hit MID. It’s called the iPhone. And in terms of early adopter power users the Nokia N95 isn’t far behind. If Intel or anyone else is looking for permission to do a good job with a MID, they have it. The market has spoken. Now get to work and make it happen.

Maloney’s cautious words, which I can appreciate, are making me cringe. I smell another UMPC-scale disaster. And I’ll be totally bumbed if this is what happens, because I really want a larger device than an iPhone for my Internet browsing, videos, home remote control, and the like.

Why do we need MIDs when we can get a smartphone that’s similar? Yes, the MIDs will have many of the same features (great connectivity, diminutive dimensions, cost), but some key differences: They will have larger displays. They will have better browsing experiences (complete with Flash). They will have more open platforms.

Now here’s where it’s going to be hard for MIDs–outside of the OEMs dragging their feet and saying “Show me first”:

First, MIDs will be secondary devices. Why is this an issue? Connectivity. WiFi is OK for starters, but many MIDs will need 3G radios or similar. And I’m like more and more people that have multiple radio-based devices and the carriers don’t see to have good offerings for users like us. Why should I be paying as much for a second or third or fourth device as my primary phone? I shouldn’t. Intel is going to have to work with the carriers to solve this or if they won’t do it, Intel is going to have to subsidize the market and prove to the carriers that they should be working with customers that have multiple connected devices differently.

I see another looming problem with MIDs, like the UMPCs before them? Price. Remember when the UMPC was going to be only $500? Didn’t happen. Instead prices reached three times that. There’s no doubt about it, price has been a big problem when it comes to the small UMPC. Huge problem. And if Intel doesn’t work with manufacturers, in terms of incentives, hardware and software engineering support, and marketing, MIDs are going to be a failure. There’s no doubt in my mind that the manufacturers will not commit themselves fully. Look at the UMPC.

So here are a couple more suggestions I have for Intel if they wanting win:

1) Watch the convergence of cameras and connectivity.

There has to be good a camera built into these devices. Most UMPCs failed miserably here. MIDs are large enough to have cameras and lenses that knock the socks off most cell phones.

Further, Work with QIK (etc) to get video out of the box in these devices. One advantage the MIDs will have over the Nokia N95 or other cellphones running Qik is the larger screen size which will come in handy for interacting with chat. Yes, this means that there has to be a 3G/WiMAX or comparable radio on board. For those that don’t have radios that go beyond WiFi there’s still hope here, but the message should still be the same: You can broadcast live, record video, upload video, and we’ll help make it easy for you.

Think a live broadcasting video feature is too esoteric? Think again. Look at who’s doing it. 1) Early adopters, like the ones you want to be standing in line to buy your device. 2) Key communicators. You want the likes of Michael Arrington of TechCrunch or Robert Scoble of FastCompany to use your product for their videos. Why? Because they communicate with a large number of early adopters. Just look on Twitter to see how much impact each 140 character sentence they say can have.

2) You better get your content ducks in order. Out of the box there ought to be three month coupons for Skype or a video download service or Rhapsody or all three. MIDs need to have bundles. I’m not talking about icons that clutter the screen. I’m talking about services that people really want. If users don’t want them, they’ll never get in the way. That’s key. But there better be blow away bundles. The bigger, the better.

3) The browser better be real and tuned for these devices.

4) The software experience is as critical as the hardware. In particular, what OS to use? Intel is in a pickle here in large part because Microsoft doesn’t appear to be onboard with this. It’s a crying shame. Maybe all of this is stating the fact that it’s not time for MIDs yet–except for companies like Apple and Nokia. For everyone else they need to wait a couple of years. Again, I hope not. I think someone at Intel needs to spend a bit more time with a few people outside of their normal group. I think they’re going to miss the obvious.

No matter what the OS, work with developers. Provide loaners. Help give exposure for MID apps. If there’s multi-touch, work with developers to come up with creative multi-touch games and apps. If there’s an accelerometer do the same there. Same goes for GPS. Whatever you do, don’t think you can slap together a bunch of hardware and some existing apps. It’s not going to work that way.

Finally, I disagree 100% with Maloney’s view that the issue is going to come down to fashion. He’s reading the iPhone tea leaves incorrectly. Sure the Apple crowd crows over the design sense of the iPhone, but that’s the Apple mistique. You’re not going to win this game. Just do a good job when it comes to design, but don’t focus in that direction.

Anyway, I look forward to see which MIDs reach the market. I hope I’m not going to be as disappointed as Maloney makes it sound.

James Kendrick ponders what the mobile PC marketplace will look like once Intel’s Atom processor hits the streets. I’m guessing it will enable many new platforms, primarily because of its size, lower power consumption, and cost. Where I have my concerns is on the performance side. James wonders the same thing. Will it be faster? Or fast enough? Or will performance take a hit.

I’m hoping that performance will not suffer, but I can fully see how from an engineering perspective we might see incremental improvements here rather than the performance we’d really like to see all at once.

I imagine we won’t know how it’ll all pan out until the Atom starts shipping and MIDs and the like start using them. Can’t wait. My credit card is already warming up to the idea of getting one…or two.

Update: UMPCPortal reprints some graphical details on the expected performance of the Atom processor. From this charts it looks quite promising. We’ll have to see how it all pans out in the real world.

Darren Waters of the BBC News takes Intel’s Classmate PC for a spin. Although Darren thinks the OLPC and the Classmate PC “herald from polar opposite philosophies about how to end the digital divide,” he sees Classmate PC as “a rival to the XO’s ambitions in the developing world generally and in education in particular.”

One feature I really like with the Classmate: Mesh networking. This alone makes its something that I want to play with. How far can the Mesh networks be pushed?

I’ve been going through a series of muddled mess with Verizon Wireless over the last couple months and today I decided it’s getting old. The problem? I can’t pay my bill online.

It all started three months ago when I purchased a second broadband device for my Tablet PC. I’m a heavy Internet user and two devices make sense for the way I work. I knew from the get go that Verizon isn’t set up for users like me, but I took a deep breath and figured it would all work out, though maybe more expensive that in ought to be. But that’s the price you often have to pay when you’re on the technology edge.

Anyway, I purchased the second device and went on my merry way, paying my bill online as always. A little over a month later I got a call from Verizon that my account was overdue. Seemed odd at first because I always paid online. And when they asked for a credit card over the phone I began to wonder if this was all a scam. After all, they called me, I didn’t call them. It took me a few minutes to realize that the real problem was that the bill was for the new device, which Verizon didn’t automatically add to my existing account–even though I bought it all online while signed in under my account.

After several calls and callbacks and failed attempts to merge my two accounts, reps from Verizon finally called me to tell me that everything was set up and the two accounts were merged. I went on my merry way.

Then a month later I tried to pay my bill online. It didn’t work. It said my account information wasn’t available and try again later. I waited a couple days and tried again. Same message. I called Verizon. They scratched their head and told me that the problem was that I didn’t owe anything and to wait a couple weeks until the next billing cycle. So I did.

After a couple weeks I tried again. Still didn’t work. A couple more days of waiting and I realized there was something wrong. Another call. This time I was told that my old account had been closed. Would have been nice if the website had told me that rather than “try again later.” The rep told me to create a new account under the new number because the two accounts had been merged under it and the old account had been closed.

So I then tried to register the new number and the registration page kept silently failing and popping me back to the landing page. Another call. This time the Verizon rep figured out that it was because I was using Safari and it was a known bug. Whew. So I switched computers (no way I was going to try the IE8 beta I had on that machine!) and used IE. The registration worked and a confirmation email notification popped up and from it I was told I could get my temporary password. So I hung up the phone. Ooops. The email just says that the temporary password will be text messaged to me. But, ugh, these are broadband devices. So I called back. They said, “no problem, go into the vzaccess software and…”—-uh, oh, I interrupted. I don’t use that software. It too clumsy on a Tablet PC and I need another boot up process like I need another IE toolbar. I use Vista’s DUN (which is also clumsy, but that’s another story.) No problem, they have a solution for that. They’ll mail me the password via USPS.

I’m beginning to scratch my head at this point. Is it really any more secure to do what they are doing than to send a timed out, encoded link in the original email that takes me back to their site where I can enter a password???? I’m realizing the problem by now. I don’t think the way the designers of the Verizon’s wireless service thinks. And I don’t think that most other commercial sites would either.

So I started to tell the rep I was talking to how I could now see what had been going on–that at each step along the way there were Verizon processes that were messing things up. I just wanted to pay my bill. So I explained would they just make a note of what was going on and review their processes? And then I heard some laughing. And that did it. The rep was not interested in hearing my suggestions. So I said goodbye and hung up. She told me to call another number anyway to pay over the phone. I was beginning to get annoyed. Actually, I went over the top when I got thrown into a maze of “push this” and “push that” instructions on the phone. I gave up and pressed “0.” I needed a human. I paid my bill (which was now late as I had expected) over the phone pondering in my mind if all of this was really more secure than this or that.

Ah, the fun of it. We’ve all been there. No biggy, just annoying.

But just think of the messes that are going to occur as broadband wireless gets more prevalent. The wireless companies are too cell phone centric and not net-savvy enough. My guess is they will be when all is said and done.

Warner Crocker’s declaration that the MID market race is over now that the iPhone/iPod Touch SDK has been announced may be a bit provocative, but I think he’s onto something. What do I mean? For the near term the iPhone/iPod Touch is essentially going to lead the MID market in terms of price, availability, developer community, quality of user experience, and funding of its ecosystem.

For Microsoft, this doesn’t mean much, because Microsoft isn’t playing in this game per se. Yes, there’s XP and Vista for some beefier units and I guess “CE” is a choice too. But let’s be serious here, for the near term none of these reach the platform model of the iPhone/iPod Touch. Sorry.

There are some Linux distros that are pretty slick that I’ve seen on some MID prototypes (at CES), but I don’t see these as major contenders either. Why? Because they are fragmenting. There’s not one single way to do things. It’s a little bit of this and a little bit of that. Nice efforts in some cases, but for the immediate future they are not going to lead the industry. They won’t condense into an industry. Sorry.

Where I think the big issue is, is not particularly in the OS “wars,” but rather in how essentially Intel’s MID initiative is left without a software partner. If Microsoft isn’t going to do it–if Linux isn’t going to provide a cohesive answer–Intel’s view of the MID world is going to have a tough time competing with the iPhone/iPod Touch family.

Actually, for Intel it probably doesn’t matter. I’m sure Intel will be quite happy if Apple switches to its CPU family and takes over the MID market. It doesn’t need the Intel-named MIDs to “win.” However, since it’s unlikely that Apple will ever license their OS platform to others to build like-minded devices, Intel will lose out in the total market it could achieve. So I’m sure Intel is struggling to figure this out.

As I see it there’s one possible software partner left–for the near-term–that could come to Intel’s aid. And that’s one that’s been quite good at filling holes left by Microsoft–Google. I don’t think it’s likely–just possible. Google could rework its Android initiative to make it MID friendly and this together with Intel’s hardware partners could possibly create an ecosystem that might have a chance at broadening the MID market. A chance. A thin one. But one nonetheless. I’m partially in doubt because of Google’s boardroom ties with Apple and Google’s lack of prior experience in this area. Besides they are thinking “phones” rather than MIDs at this point.

Note that in all of this I’m talking near term. Three, five, ten years out anything could happen. However, for the next year and possibly two Apple is in the lead and will most likely stay that way.

For all of you that are thinking I’ve fallen victim to the hype, let me encourage you to do a few things: First, get an iPhone. Use it and then tell me that it’s connectivity and form-factor don’t draw you in. Second, play banker for a minute. Where would you place your bets? In an unproven, fairly fragmented software market geared around an unproven family of MID products from hardware-based Intel and its partners or with Apple and its “iPod” family including the recent iPhone device that’s done remarkably well for its first year? And third, if you’re a developer, check out the iPhone SDK. Tell me it doesn’t have many of the things we’ve all wanted as rich-app, mobile developers. We’ve seen similar things before and beyond as well. But have they been so well packaged before? If you’re still not convinced, watch the demos during Apple’s SDK announcement event. And if none of this convinces you, I bet you’re not shaken by near term events, so no problem. I agree with you that in time the world will reach equilibrium and all will be well. What’s a couple years here and there among friends?

Yes, several years out, things will most likely be much different. And no doubt, there will be plenty of room for niche players across the board in the meantime. And Microsoft’s and Intel’s coattails are big enough to carry several efforts along the way too. But for the Kliener Perkins level break away products, I’ll be looking in the iPhone/iPod Touch direction.

As a Tablet/UMPC enthusiast and developer, I’m not going to dump what I’ve been doing and radically change directions, but you can rest assured that I’m keeping my options open. Like with the OLPC and Eee PCs I see an evolution occurring here and I don’t want to be left behind.

Good technology and good implementations deserve respect. And I’m treating the iPhone with such. Doesn’t mean my competitive instincts aren’t kicking in. Just means I’m very mindful and doing my best to do my best.

Wired magazine takes a look at the forthcoming MID (Mobile Internet Devices) that will be hitting the markets this year.

Will the MIDs find an eager market? That’s the $64,000 question.

The article points out that Intel is making great progress at creating PC-grade, mobile processors that scale down to pocketable devices. Intel is pinning great hopes on the Silverthorne family leading the way.

There is something very important that the article leaves out: The software. It’s not just a matter of building a small device. That’s only part of the problem. These are for the most part “sealed” units (from a hardware standpoint) and as such people are going to expect them to work as is–and work well. That means the software is going to have to be up to par.

You see, the big difference between let’s say a MID and a Pocket PC is not just the hardware. It’s the fact that a MID can run a full OS and full apps. However, it doesn’t take much to realize that most apps are going to have to be tuned for these small devices.

From what I’ve seen (at CES, for instance), some MIDs will run a variant of Linux and some will run Windows. The question is which one will yield an experience on par of iPhone’s OSX-lite?

A MID needs to be fast on, have easy access to apps, and boast apps out of the box that run well–really well (a browser tuned to reading on a small display, live maps, live weather, and so on). 

Where’s the initiative to create any of this? I don’t see it. This means that the manufacturers are on their own which will mean spotty results and platforms that may be extensible, but will take work for third parties to take advantage of.

So I don’t think the MIDs necessarily will be DOA from a hardware standpoint. It’s going to be the software that makes a huge difference here. And once the software is “good enough” it’s going to be a matter of whether the hardware and software blend together to make a device that really signs.

All this being said, if the prices are low enough consumers will be more acceptable of any flaws–at least the early adopters will. My prediction is that this is going to be an evolutionary product. We’ll see.

Intel has decided to make an about face and has resigned from the OLPC board according to the Wall Street Journal.

News.com reports that the OLPC effort asked Intel to stop working on the Classmate PC as well as with others on small computers such as the ASUS Eee PC. Of course, Intel said no.

In the competitive world, I don’t blame the OLPC effort for trying to force Intel’s hand, but on the other side, I also don’t blame Intel for saying no and resigning.

Well, I guess that insures that the OLPC will have at least one more competitor. I’m not sure that’s what the OLPC team wanted in the first place, but oh well.

Things are definitely heating up in the low-cost, laptop space.

My brother Layne stirs up the MID vs UMPC debate.

A couple of points: There are Tablet PC and UMPC evangelists at Microsoft. They just appear to be spread out. I think that happened naturally as more and more people within Microsoft adopted Tablet for their own use. But more so, there are some key people that believe just as strongly today about the Tablet PC as they ever have, such as Bert Keeley. There are many others.

Now in terms of Intel, I agree with commenter Chuck, Intel has been focusing on “downgraded” processors in large part to conserve power. Processor speeds have been going the wrong way for us Vista folks. If I understand the verbiage correctly from Intel’s Developer Forum (IDF) there will be faster processors down the road, but for now I’m not sure if Vista runs on the current generation. I’m a bit confused. Intel is promoting Ubuntu on these platforms and it looks to me like Vista may be “supported” later. I can’t quite tell. My guess is that XP would work fine, so I’m unsure why Intel didn’t show that. Either way, it’s going to take more than a few months to get Ubuntu up to mobile speed. I bet the same thing could be done to XP and would be a better overall product. How hard it is to bring in a new shell and fix a handful of overly large dialogs? I can guess why Microsoft wouldn’t be for this, but in light of the MID (sub-UMPC) market going Linux (courtesy of Intel) I don’t see that Microsoft has a choice. Windows CE isn’t going to cut it and Embedded isn’t either. These need to be full OS’s. That’s the beauty. That’s the innovation. Of course, if Microsoft doesn’t want to play along, I guess Intel had no other choice but to go with Linux.

Finally, Microsoft marketing takes a hit quite a bit of the time. They’ve tried, but the effort has been mediocre at best. I think it had a lot to do with marketing timing–not just marketing. When the Tablets first came out the hardware channel was changing significantly. The reseller channel lagged and was crippled by direct sales from the OEMs. And the OEMs had little insensitive to sell a niche product. Sure they would if the market pulled it through, but that wasn’t going to happen with all of the negative online commentary that occurred in large part not because of the Tablet per se, but the transition to notebooks. Centrino helped and now notebooks are more accepted and the Tablet would do much better today if it were released. I’m convinced if Tablet PCs were launched today you wouldn’t see endless Slashdot or Digg commentary about how heavy they are, how overpriced for its size, how the battery life is too short—because ALL of these issues impact ALL notebooks. People understand the issues better. Now that being said, most Tablets and UMPCs are still overpriced to me, but that’s just me.

Oh, and one more thing, I do shed a tear each time I see an iPod Nano commercial on TV. That could have been a Tablet PC or UMPC commercial.

The Moorsetown MID mockup that Intel showed at IDF (shown below) sure looks like a stretched iPhone.

So much so that I began looking for an Apple logo:

No way that’s really an Apple logo, but funny if it were.

Update: That’s not an Apple logo. It’s a camera lens. Still, it would be funny if it were the former.

I’ve have an industry rule: Watch where Intel goes. So often where Intel goes, so does the industry.

It makes sense. Intel provides the basic circuitry and know how on how to build the many devices we use. We have to follow where they go. Yes, there’s competition, but Intel is the predominant force that sets the path that we can practically follow.

So with this rule in hand, I watched a webcast of Intel’s Senior Vice President and General Manager of the Ultra Mobility Group’s IDF presentation yesterday.

Much of the webcast discusses Intel’s next two generation of mobile-focused technologies: Menlow and MoorseTown. Both are lower power X86 technology platforms. Both will enable even smaller mobile devices.

[Menlow devices about to become available]

That’s good news. Intel hasn’t had a very good selection of very low power devices that live up to the dreams and aspirations of UMPC owners. It’s now getting there.

The news for those that follow the UMPC platform may signal a bumpy road ahead though.

Intel is squarely marketing its platforms as a Mobile Internet Device, not a UMPC–even though they toss in UMPC support every now and then. It looks to me like Intel is moving its marketing might away from the UMPC. That’s very unfortunate for those that develop and support this platform. Now some could argue that Intel hasn’t done all that much to advocate the UMPC anyway. As a software developer, that would be my opinion. Outside of announcing the UMPC before Microsoft and providing some sneak peeks as various reference designs, I’m not sure what else Intel is doing in terms of UMPC evangelism. I’ve been at Intel events where the people don’t even know one lick about UMPCs, other than some group in Intel works on them. Then again, maybe I’m looking in the wrong places.

Anyway, now Intel’s marketing effort for the UMPC is making more sense–it’s all about the MID. Now in some respects, what’s the difference between a MID and a UMPC? Maybe they’re a little smaller, but other than that we’re talking about very similar devices. In fact, this Moorsetown-enabled MID mockup sure looks a lot like the Haiku device Bill Gates was showing around as the future of the UMPC as well as a little iPhone splashed in.

So why aren’t these devices called UMPCs? Yes, many are smaller. Here’s why: Vista isn’t going to be running on the first generation ones. Why not? Vista is too big and expensive for devices like this where the no-brainer goal should be to drive down costs and yield the best experience. Instead Intel is turning to Linux.

Normally, I’d say no biggy. There’s plenty of room in the market for multiple types of devices running different platforms. But here’s the thing: Microsoft has no OS for these devices that can compete at the Linux level. Zippo. And as a developer that’s been developing for Microsoft’s Tablet PCs and UMPCs this is a big deal.

Yes, I expect someone will get a Vista version of one of these devices running and wave it around on stage maybe at CES maybe six months later (Moorestown should be minimally sufficient with its 16GB spec and processing might), but I’m going to be wondering properly so if Vista is not the best choice for these devices?

The real bummer here is that Windows XP probably would scale right for these devices (in terms of footprint and cost)–outside of teaks to the Shell and the various desktop-sized dialog boxes sprinkled throughout the OS. But Microsoft has turned all its attention to Vista and XP’s days are numbered. So XP isn’t going to get a MID-slash-UMPC-sized refresh. It’s a shame for Microsoft developers.

It’s time to dust off your C/C++ code. For MIDs it’s going to be a Linux world–at least in the short term.

Now there is some hope for Microsoft-focused technologies. And that comes from Novell. Maybe, just maybe, Mono will have a small enough footprint. And in terms of Silverlight technologies, maybe, just maybe Moonlight will be up to par. However, who knows.

There’s another possible future here. Intel may only half-heartedly promote this new platform. And if the OEMs are left to their lack-luster design ideas as they have shown a proclivity for doing over the last few years, then MIDs probably will be outshone by iPods and iPhones and similar devices. And so the MIDs will fade leaving room on store shelves and in time to get Vista even better on these lower power devices.

I also noticed that in all the live demos people were using styluses and that indicates to me resistive displays. That’s not up to par with the current market expectations. Think capacitive touch. Think gestures. Don’t assume that everything is going to be touch like in an old-style POS or Kiosk. Can Linux be tuned fast enough to create the kind of rich experiences that people like me now expect? Give anyone an iPhone or an iPod touch and they won’t want to go back. Sorry folks. I didn’t see anything on stage that indicated that Intel or the Ubuntu representative understood this.

So anyway, this is going to be an interesting time. Intel is clearly driving power consumption down and expects many low-end devices with small drives and limited screen real-estate. We’ll see if this is what the consumers want or if they want more power in their hands.

I just watched the live UStream.tv feed of the IDF interview with Intel co-founder Gordon Moore. Npr host Moira Gunn did the interviewing.

Absolutely wonderful!

First, the UStream technology worked extremely well. The video was great. The audio was great. The chat room was (for the most part) well mannered and lots of questions were taken from the chat room and used in the interview. Wonderful job.

A couple tidbits I picked up from the interview that I was able to jot down:

* In terms of Moore’s Law, he sees that eventually it’ll hit a wall. He predicts a decade to a decade and a half out.
* If he were back starting in school today, he go into some form of biological sciences field.
* Early on people costed out systems at 68 cents a transistor. Today the cost is in the range of 10 pico bucks.
* Early on lots of people had concerns about yield problems with ICs. They thought that the yield would exponentially decay with the number of transistors. It didn’t. Another issue at the beginning was that people were used to measuring things at the discrete component level. They didn’t need to with ICs. It took people awhile to warm up to that.
* He credits co-founder Noyce for coming up with the idea of pricing ICs less than the cost of the discrete components.
* He predicts that once language recognition becomes a reality that computer interfaces will become conversational and that we’ll interact with computers the way we do so with colleagues. In the meantime, he describes himself as a Luddite: He uses a keyboard and a mouse.
* Oh, and about the Intel name. They went through about 15 names. The first several they couldn’t get. Eventually they settled on Intel. Amazingly, they later found out that the name was already taken. They had to buy it from a hotel company in the Midwest.
* Gordon Moore enjoys deep water fishing….and his foundation is active in conservation.

I see that Omar Shahine, who’s done some great work in the past with Tablets and OneNote, is sharing his enthusiasm for Mac PCs. Why? Because of the ways Apple simplified running Vista on the Macs via Bootcamp. In fact, Omar is so satisfied that he proclaims: “I think all my future PCees will be Macs.” Yep, the crowd is growing. The PC OEMs–and Microsoft too–need to take notice, if they haven’t already.

Chris Pirillo blogged about the Mac migration the other day too. He’s sees a growing number of “influentials” turning to Macs and was surprised at how many Macs were at his recent Gnomedex event.

I don’t think it’s just because of the state of Vista that’s at play here, although that’s a contributing factor (drivers, sleep/hibernate issues, IE stabillity, and the like). I think Apple has simply done a better job of running Vista than other OEMs. The key is in the drivers, the installation, the small touches that improve upon Windows itself. Apple deserves a star for effort in all of these areas. Microsoft is being “out-Windowed.”

As many people point out, Microsoft has a great challenge ahead of it because of all the partners it works with. It doesn’t control all the hardware. Well, Microsoft and the OEMs can’t keep going on like this. They are going to have to adjust. Here are some ways I think Microsoft and its partners can get back on track and create platforms we want to use for the next 10 years.

1. Bifurcate Windows. Yep, it’s time to split Windows into two versions: One that’s more legacy biased and one that’s focused on the future. Whether Vista fits into this, I’m not so sure. In some ways, I see Vista itself needing to be divided: The Desktop Composition Engine and the graphics moving forward in a new version and at the same time the old GDI world taking a step back. This could mean bringing an updated shell, network manager, and the like to XP for the backwards-compatible version and then moving on to a new version of Windows that’s more graphics intensive and network-minded. This version, for instance, would make the DWM API look like the straight-jacket it is. It would also be responsible for making sense of the whole .NET Framework concept. Either go all the way or take a seat in the luggage compartment, I’d say. This half-and-half world is optimized for too few. I also think a new OS would be a good time to bring C++ development back up to par. It’s a shame how it’s been languishing within the Microsoft language stack–much of it induced by backwards-compatibility concerns. This would be the chance to set things on a better path.

You know, for all the pains Longhorn went through, I think the original intent was right. It was time to make a break from the past. The problem was some of its most important design principles didn’t have the same priorities that most of us had–especially considering the rapid movement to mobility. There’s no doubt the challenge in changing so much on the only OS road that Microsoft had was too difficult. The solution was to not dump Longhorn. It was to split the effort–to leave two roads open. The single road effort was too restrictive.

2. Microsoft, Google, and Intel should join forces to build out an “open” WiMAX network. Rapidly we’re on another networking cusp as wireless broadband becomes available on the hardware and carrier side. However, I have big concerns that the major carriers are going to get it wrong. If history is any gauge: they’ll want to charge and control each connection point. That’s a recipe for market suppression and one that favors fewer devices at the endpoints under centralized, walled management. Old-school. Clearly this does Microsoft and Intel no good–both companies want to sell as many devices (or OSes) as possible. This isn’t going to happen if it’s a pain to maintain too many contracts on too many devices. So why should Google join the wireless broadband trio? First, because they have a philosophy consistent with solving the problems that others set out before them–whether it’s desktop search or ease, simplicity in maintaining tons of email, or minimizing deployment of a service. There’s a problem here and maybe Google will want to join forces to help solve the problem. The other reason is that Google doesn’t want centralized control to devalue its offerings. There’s lots at stake here and to cede control at such a fundamental level–the network is not good strategy.

When I read about how Sprint, for instance, is dedicating billions to WiMAX it makes me hope that the pending problems will be self-correcting, but I’d rather not hold my breath. Yes, a billion here and a billion there adds up, but with Google, Intel, and Microsoft joining financial forces what they could achieve could be phenomenal and set the industry on a strong path for another decade.

At the OS and app level, Microsoft and Google could work together to create standards and infrastructure that mitigates issues with moving apps and data online and offline. Today everyone is going their own way, which is fine. However, at some point–which I hope is sooner than later–we need a concerted effort that brings developers and networks together. The deployment of wireless broadband could be this chance.

No doubt that the industry will have many changes going forward–I just hope that some of them will help out to expand the industry as much or more as it has over the next ten years as it has the last twenty plus. I’m ready.

Update: Engadget asks what can be done to improve Vista. As expected there are quite a few commenters–many of them challenging Windows itself, which doesn’t help all that much and for the most part the commenter suggestions are all over the place.

There is a bit of a pattern that I see here and that I think I’ve been noticing elsewhere:

Doesn’t it seem like desktop owners of Vista are more satisfied than their notebook counterparts? Maybe I’m reading too much into things, but it seems like this is a pattern. It kind of makes sense. My guess is that for the most part Vista was developed on desktops. Likewise, people who have desktop systems can swap parts in and out to get things working as they need. Likewise, in desktop systems there’s more pound-per-pound investment in the hardware, so some Vista features may not be as critical or as noticeable as others–the machine itself is the king.

Am I being too mobility sensitive?

I do wonder if the mob isn’t simply piling on here. Vista is an improvement in several important ways over XP. The question is collectively (in terms of both the hardware and software) is Vista and the Vista experience where the market wants it to be. I’d say it depends. As a developer things are in a bit of a state of flux in terms of Microsoft development, but as a developer that’s OK. I’m looking down the road. In terms of customers, it’s not too bad as long as things are tweaked–but that’s nothing new, give a new computer to anyone and it takes time to get it “just right.” The issue I think is that people want Vista to “just work” the way they do even more. That’s a good thing. It shows how much they depend on Windows as it is.

This video by Robert Scoble of Tom Rolander (of Digitial Research fame) has to be the best first-hand account of the story behind the OS included on the first IBM PCs.

If you’ve heard the stories about how Digital Research’s Gary Kildall lost out on one of the biggest business opportunities in recent times because he wanted to fly a plane instead of meeting with IBM or that he refused to meet with IBM because of an NDA, then you’ve got to watch this interview. The stories wrong and some very important ways. To hear Tom Rolander tell it, in large part the failure of Digital Research’s CP/M came down to price–and I’d also add in their belief that the IBM PC would fail.

What’s so fascinating about Tom’s account is that it highlights how established, successful businesses can have difficulty adapting to expanding business opportunities—opportunities that challenge existing cost structures.

We’ve seen a similar challenge and churning going on over the last decade plus with the Internet. It’s changed the cost structure of distributing information, of managing services, or of what it means to stay in contact with others or to communicate with them. Costs are being driven down.

Amazing video. You have to watch it.

Robert Scoble and others are pointing to Microsoft’s recent notice that this year’s PDC (previously planned for the fall) is being postponed.

To me, it’s no big deal whether PDC is postponed or not, but PDC does have the tradition of standing out among Microsoft conferences, as the one event to attend if you want hard-core information about where Microsoft technologies are going for not just this year, but for years down the road. For this reason, PDC is an important event to keep an eye on.

Some are speculating that Microsoft is postponing PDC because there isn’t much new to talk about. Others are suggesting that the timing simply isn’t right, because several products (Silverlight, Orcas, etc) will be about ready to launch and PDC is traditionally about new technologies rather than products just being released.

Whatever the reason, this has sure helped to establish Mix07 as the Microsoft developer event of 2007 (where several new major technologies, most involving Silverlight, were announced)–especially for developers that want to stay on the edge of forthcoming Microsoft technologies. This makes it even more unfortunate that Mix07 was unable to handle the numbers of developers that wanted to attend. I imagine that if Microsoft had known that it was going to be postponing PDC, it would have opened up Mix more–at least I hope so.

So what would I like to hear about at the next PDC?

* Bring C++ libraries up to the level of the .Net platform in terms of ease of use. I’m not talking about leveraging managed code. I’m talking about higher-level, easy to use libraries.
* And, of course, improve the C++/XAML story too.
* Enhance the programming model of the Office Ribbon so that it not only restores capabilities that were available before, but goes beyond them.
* Additional commitments to mobility in Windows. Even faster boot times. More efficient use of processor(s)–especially while on batteries.
* Even better networking support in Windows–particularly for ad-hoc networks and the like.
* Multi-processor programming strategies and support
* XAML debugging
* Better design tools within Visual Studio, ranging from icon development on up.
* Even better productivity through editing/refactoring tools in Visual Studio
* Silverlgiht on Linux
* Optional handwriting recognition download for Silverlight–calls local Vista recognition engine on Windows machines and Apple’s InkWell techology on OS X, and leaves open hooks for Linux developers to add their local own recognition engine.
* Scaling and rotation of windows (frames and all, although it would be OK if this feature is only available for frameless windows). Fixing the .Net platform and Windows itself to better handle DPI would be OK too.
* Improved startup time for Vista speech recognition engine
* Ability to access or programmatically supply a video stream in Media Center
* Better support for lower resolution displays running Vista. It would be great if all Vista dialogs fit within 800×480 resolution, which is standard on many UMPCs.
* I also wouldn’t mind seeing C# change its tune on destructors and its garbage collector, but that’s just me. I won’t hold my breath for this though. * Added: A vision library. Cameras are now standard equipment in cellphones. They should be an integral part of the computing experience too. Some companies have started filling in the gaps, but there’s nothing like a cohesive, stable core with an easy to use API to make the market sing.

I can think of many more things I’d like to hear at the next PDC, but this is a good start. I’m sure Microsoft has a good number of surprises up its sleeve too.