Microsoft unveils advanced distributed processing.
Microsoft Corp. is working on a wave of new technologies for the consumer and the enterprise customer that it hopes will change the face of technology as we know it.
Many of the algorithms, agents, security methods and other projects being developed by Microsoft Research focus on distributing processing and intelligence all the way out to mobile and disconnected computing devices.
But industry observers and Microsoft officials alike said that many of the technologies in development at Microsoft Research may never find their way into a shipping product.
Microsoft Chief Software Architect and Chairman Bill Gates and other members of his executive team gave a rare glimpse of some of the technologies at an event held this month on the companys campus here to celebrate the 10th anniversary of Microsoft Research.
Gates said he believes future innovation will revolve around using computers to extend what people can do; improving digital music, videos and images; and building out Microsofts .Net vision for delivering software as a service over a network.
"The tectonic shift over the next 10 years will concentrate on a new approach: the idea of distributed computing where different parts run on different applications," Gates said.
The combination of innovative software with the right hardware would also be used effectively going forward. As computing power continues to double every 18 months, innovative software will allow devices such as monitors, cameras, Pocket PCs and cell phones to do new and exciting things, Gates said. To demonstrate this, he showcased the Sensing Pocket PC, a device that alters handheld applications based on sensors that detect touch, tilt and motion.
Many of the research projects taking place at Microsoft are directed at the initiatives Gates outlined. But much of the research is long-term and will not be incorporated into formal products for a long time, if at all, said corporate IT buyers and company officials.
Take the Loader .Net project, which allows large applications to be transmitted in bits over narrow bandwidth as needed. This saves the user from having to wait for the entire application to download before he or she can start using it. While the technology is fairly advanced, it has yet to be incorporated into any Microsoft product, a researcher on the projects team told eWeek.
Other technologies being developed that have yet to be formally incorporated into shipping products include watermarking for audio content, which helps enable secure multimedia e-commerce; time-warping, which allows video enhancement and correction; and intelligent TV browsing, involving content extraction and advanced browsing tools.
Microsoft Researchs center in Beijing is working on potential groundbreaking projects, including the MiXP project. This is a research prototype that works on top of .Net and records every aspect of a users life experiences and interactions with others. This data is then stored in a single database, categorized and made available on devices from handheld to desktop PCs and from mobile phones to personal digital assistants, according to Ya-Qin Zhang, managing director of the Beijing research operation.
"The current end-user interface of folders, menus and controls will undergo a significant change in the future, and it is imperative that we are able to search and actively find the parts of this data most relevant to us. This is really a step toward the .Net vision, where devices are more intelligent and personalized," Zhang said.
One of the greatest attractions of these components is that they could be used for both the broad consumer market and the enterprise customer. Another early-stage project in Beijing is Upen, a penlike device that traces everything a user writes on paper, an LCD screen or any medium, then automatically processes this information and sends it to a computer over a wireless link. It can also function as an interactive device. The user points at something and then that information is collected by a camera, processed and sent to a computer.
A lot of the research teams .Net focus had been on basic work on algorithms for distributed computing, security methods and algorithms, cryptographic techniques, and ways of sharing and distributing information so it can be protected against any kind of failure of the main system or network.
"We are working on a fail-safe storage project called Farsite, where we are building a symbiotic, serverless, distributed file system that ensures information cannot ever be destroyed," said Rick Rashid, senior vice president of Microsoft Research. "It is distributed in that it runs on multiple machines; its serverless, meaning that it does not make use of a central server or a cluster of servers; it runs entirely on client machines. And its symbiotic, meaning that it works among cooperating but not completely trusting clients."
Robert Rice, an independent computer consultant in Salt Lake City, tracks research trends to see how they could be of use. "New software and gadgets like those being developed at Microsoft increasingly face a hard sell as their usefulness vs. the learning curve required to use them may not justify their use or purchase," Rice said.
Rashid was quick to point out that many research projects quickly find their way into products (see interview, Page 34). However, Microsofts Zhang said users may never see the direct fruits of Microsofts efforts as it was "very difficult to predict when and if some of these fledgling technologies are going to become actual products."