Otto Berkes, increasingly being referred to as the ‘father of UMPCs’, was the manager of Microsoft’s Origami team. In March 2006 the team announced the Origami ultra mobile PC device category and presented a handful of devices to the public. Today there are well over 30 different designs that have built on the work that was done and form the ultra mobile PC device sector. Christoph Derndorfer of Epiacenter.com was in Tapei in June and agreed to do some work for UMPCPortal. We tracked down Otto Berkes and managed to set up an interview. Here’s the result. Many thanks to Otto, Christoph and Epiacenter.com for the time they gave.
Interview with Otto Berkes
Taipei. June 2007
What are the most important themes when it comes to the ultra mobile PC market right now?
I think the themes of high integration and miniaturization of the silicon and hardware building blocks are very important. I feel that PC design – especially mobile PC design – has been limited by the difficulty of packaging all the components that make up a PC into a small form factor. So far the form factor and the design has been dictated by the constraints of putting all of those pieces into a package rather than with the product designer creating the ideal form factor and design for a particular use.
It’s also important to emphasize that it’s not only just the size of the components but also other factors such as power efficiency that determine the overall design. So for example if you can reduce the pressure on the battery size by decreasing power consumption, that will have a positive effect on the form-factor. Or if you don’t need a fan to cool the components then the thermal-solution can also be much simpler and more elegant.
Intel are indicating that they think a split (UMPC/MID) is necessary in order to position ultra mobile PC products correctly in the market. Does Microsoft support this or will they continue to focus on the higher-end ‘UMPC’ definition?
Looking at the trends in technology miniaturization I don’t think that this is about size. There’s nothing to prevent full PC functionality from being delivered in small devices. I also don’t think that people are going to be confused about what is and what isn’t PC. From the beginning the goal of ultra mobile PC was to create a new category of a full-fidelity, full-functionality Windows PC and that remains the goal of our ultra mobile PC efforts.
It’s possible of course to have a whole spectrum of devices that deliver a more limited set of functionality – for example, a device that only does a few things such as internet browsing and media playback. However, I think the value proposition of these devices will be fundamentally different from a ultra mobile PC device.
When you started the Origami project, did you expect battery technology to move forward? Do you think that the battery industry is really interested in low-power devices that can negatively affect their revenues? If so, how can we solve this issue?
There’s certainly some exciting longer-lead research going on in battery technology. But it takes a very long time for new battery technology to make it into the market. So going into ultra mobile PC and Origami development my assumptions about battery technology were always very conservative, assuming maybe 8% to 10% improvement per year. I’m not counting on radical battery improvements to increase battery life. We really need to address battery life from a demand side rather than the supply side. Having said that I believe that there’s really been great progress on the demand side not just in the CPU and chipset but also for example in displays thanks largely to LED-backlights.
So with regards to power-consumption, what are the main challenges considering the improvements made in the recent past?
I think chipsets still remain a challenge, so if I had to pick between CPU and chipset I’d choose optimizing the chipset power. Hard-drives are obviously also power-consumption hot-spots. Intelligent automatic backlight control can also be a helpful step to reduce display power consumption.
In terms of the software side do you think that the current solutions that Microsoft and others are providing are good enough or even adequate?
I think this is going to be an evolutionary process; we’re absolutely not there yet. Both the software and hardware need further improvement but there’s good progress being made on both fronts. I don’t think there’s a silver bullet and there’s still much that needs to be done across a number of areas to improve the experience for ultra-mobile hardware. And the ongoing process of user and market feedback will help guide development efforts in the right direction.
For example, if you compare the first generation of software we released based on XP for UMPCs and the Origami Experience software that we released for Vista, it’s clear that we made major improvements in overall functionality and in the user interface. For example, the latest version of the Origami Experience software enables a complete media playback experience with touch-based interaction which doesn’t require the use a keyboard or stylus. That said there’s still a lot more work to do optimize the user interface in many other areas for smaller touch-enabled displays and we will continue improving the user experience around these concepts.
So do you feel that the current Vista platform is a good basis to go into that direction? Or could a more dedicated and specialized operating-system, built from the ground up for these applications be better?
The beauty of Windows and the Windows PC is that it provides a very consistent framework for both developers and consumers and a very high base level of compatibility with software, peripherals, hardware, and the user experience.
I’m a big advocate for lightweight hardware and I’m also a strong believer in the emergence of Flash as an alternative to the hard-drive, especially for smaller PC form factors like UMPCs. Of course taking advantage of Flash for mass-storage of the operating system will require an operating system footprint that’s much smaller. I’m optimistic that the needs of ultra-mobile devices can be addressed by more efficient scaling of the underlying operating system.
I don’t believe there’s anything inherent in the challenge of optimizing for a greater range of hardware that would require a new operating system as such. I think that the Windows operating system can evolve over time to encompass a more dynamic range of hardware capabilities.
One of the important aspects of UMPCs and mobile-computing is about being connected. With people being connected 24/7 and having access to web-based services and applications such as Gmail, Google Calender, Salesforce, etc. do you think that the importance of the operating system on a device that’s constantly connected will diminish compared to now?
I think the importance of the operating system will remain quite high because at a fundamental level the operating system provides the complete substrate for rendering digital information. The user ultimately doesn’t want to see any break between web-based, hybrid, and local applications in their computing experience. Even a small break, a missing plug-in for example, may be very painful to the user if he or she can’t get to the content they need. The beauty of the Windows PC is that in a very real sense it’s the most powerful renderer of digital information regardless of whether that information is local or across an internet connection. There’s no digital content that a Windows PC can’t render in a full-fidelity way and I think that’s incredibly powerful.
What’s your favorite ultra mobile PC design at the moment?
I haven’t really had the luxury of putting the latest designs through their paces. The product from the first generation of devices that I found myself to be using the most was the Samsung Q1B with the extended battery. It was and still is fairly magical to not have to be worried about charging a PC throughout the day. I was even brave enough to leave my charger behind on day-long business trips in the same way that I can leave behind the charger for my mobile phone. And that’s really the right model for this type of ultra-mobile computing experience.
If you had to describe the most important characteristics of a ultra mobile PC for you personally, what would they be (apart from the previously mentioned battery life)?
I think weight is a huge component of easy-of-carry so that for me would be second. The third one is a little less tangible but it’s a combination of industrial design, aesthetics and ergonomics. This also extends to attributes like heat-dissipation, fan noise, and durability.
So if Apple were to come up with a ultra mobile PC – of course running Windows Vista via Parallels – you might be interested?
*laughs* I’d certainly be very interested in seeing what Apple had done given their track record with industrial design…
Can we expect major marketing efforts from MS in the near future or do you think that MS has given the segment enough moment for it to sustain and grow itself?
Different players involved in the market will have a different approach to marketing but I believe that in general a more unified effort would be beneficial for the ultra mobile PC market. It’s important to realize that with a new device category such as this one it will take time and effort to clearly communicate the basic message to people. I think that marketing efforts will definitely influence how quickly this market segment will accelerate and grow.
Where do you see UMPCs 12 months from now?
I think there’s going to be a lot of fine tuning since most of the hardware design basics have reached the “good enough” point. One of the aspects that companies will continue to focus on is the total user experience. I think that the mobile phone and device market will influence the evolution of ultra mobile PC with regard to things like cost, battery life, hardware user interfaces, connectivity, data service subscriptions, etc.
It’s important to understand that the original Origami initiative didn’t dictate a design but was rather a framework to foster device-like PC hardware development. The goal of the Origami project was to serve as a starting point the ultra-mobile computing segment. When introducing a new device category it’s always challenging to do Version 1 but I think what really matters is what happens in the long-run. Version 1 got the ball rolling and now we can all build on that.
I’m also guardedly optimistic that 2007 can become a springboard with a critical mass of customers being reached.
While we’re talking about a critical mass of customers, what role does the price of UMPCs play?
I think the most important aspect of the price point is that it will strongly influence the public’s perception of ultra mobile PC as something that could fit into their lives. In my opinion $500 to $600 would be a very good price range for many people.
Today there’s still quite a different between a “UMPC” and a “sub-notebook”, do you think this is going to change in the future?
Overall we’re going to see a blurring of these differences. Some UMPCs will come in a form more closely resembling sub-notebooks while some sub-notebooks will incorporate some of the features that are usually attributed to UMPCs such as touch displays.
Can you also see UMPCs really replacing regular PCs? SOme people user their UMPCs in extended-desktop-mode connected to other peripherals such as monitor, keyboard and mouse. And do you think that traditional computer manufacturers might not want to support UMPCs because they’re afraid their boxes are going to be replaced by them?
By design it’s absolutely possible to use UMPCs in the way you just described since manufacturers are including VGA outputs and you can use standard USB peripherals. After all a ultra mobile PC is a real PC with full PC functionalities. In terms of the marketing I’d really expect the focus to be on the mobile aspect because the value of the mobility is such a core element of ultra mobile PC products. There are some other aspects that lead me to believe that such a desktop usage won’t be too common: For example the pricing of the low-cost 15″ laptops which is around $500 to $600 is currently still less than the typical UMPC. These laptops are of course significantly larger, use more powerful but more power-hungry processors and graphics and include a DVD-drive because that’s both expected and feasible in this form factor. Even assuming similar pricing, the difference in terms of overall computing capabilities and features is going to be significant for most users. So I don’t really see this type of use being mainstream and expect it to be an augmentation to existing PC usage models rather than a replacement.
Hopefully we’ll be able to catch up with Otto again in the future and re-visit some of these topics. Thanks again to Otto and Christoph for giving their time to UMPCPortal.
