Friday, August 31, 2007

Mobile Phone – The Next Platform for Innovation

One lesson I’ve learned in working with emerging technologies is that you can never stop innovating. In the following article there’s a quote that I like,

“Technology only continues to be successful when it looks to the future. “

This certainly holds true for the mobile phone which in my opinion has now overtaken the PC in terms of “innovation attention” and for the future will be the platform of choice. Innovation can be found on all fronts with the mobile phone.

In the memory department, 8 Gb is now the standard memory space for a mobile phone. The iPhone comes with that amount of memory. (I’m holding out for the 32 Gb version myself.) Samsung just announced an 8 Gb SD memory card for its phones.

ARM just announced a 1 GHz processor for the mobile phone. Going forward, multicore processors from Intel will be targeted at the mobile device with an eye towards cost.

In the camera department, the camera on the mobile phone has grown from 100K resolution to 3 mega pixels today. In fact mobile phones are on the verge of invading the digital still camera space. Check out this whitepaper on the topic.

The number of mobile phones with GPS capability will increase 10x in less than five years according to this article.

The use of Web services will be extended to the mobile phone in a much tighter synchronization taking into account a user’s personal set of preferences.

Inside the mobile handsets, the power amplifier will be integrated into a single silicon chip and will use a silicon-on-insular substrate for signal isolation.

Research at Georgia Tech predicts wireless data transfer rate at the 1 meter distance will reach 15 Gb/sec.

If Clayton Christensen were writing the “Innovators Dilemma" today he could use the shift from PCs to mobile phones as his example case rather than the shift from the mainframe to minicomputers to PCs which he did in the first version.

Best regards,
Hall T.

Friday, August 24, 2007

NI Labs – Generic Containers

One of the new NI Labs technologies now available on NI Labs is Generic Containers. LabVIEW provides a set of standardized interfaces for manipulating data (saving, reading, writing, etc.) to the developer which are called APIs. These APIs are grouped into “containers”. By using generic container methods the user can store, read or handle any type of data. These APIs provide type-independent interfaces and abstract away most of the common operations such as search, sort, insert, delete, etc.

A “container” is a data structure used in a wide range of applications. It provides methods for accessing its elements which are data type independent. A generic container can store elements of any data type. The Generic Container tool on NI Labs is geared toward searching large amounts of data of varying types. Since the container is data type independent, it can handle any kind of data. I’m interested in hearing from you about the use of G programming tools. Please contact me with ideas for tools you would like to see.

Best regards,
Hall T.

Friday, August 17, 2007

NI Labs – Bioinformatics/Biological Data Management

The Lab Automation market has grown from $465M in 1998 to $2.1B in 2003 based on research by Strategic Directions Inc. Given the government and private sector funding going into the biotech industry, these numbers are not surprising.

In talking with lab automation specialists one of the first topics that come up is the volume of data generated by lab automation systems in the drug discovery process. It’s not unusual for scientists to download 500 megabyte files from several of the 750+ databases containing genomic/proteomic/biological data and then merging it with their own research data. Given three or four file downloads and your into the gigabytes of data.

To solve this, NI has come up with a starter kit for managing Bioinformatics/Biological data and offers it through their NI Labs program.

The software kit works with Diadem which is a technical data management tool that provides a user interface for managing and analyzing data and is ideal for combining data from multiple sources and indexing the data so it’s searchable.

It’s common for researchers to download data from say the Stanford MicroArray Database, or the GEO (Genomic Expression Omnibus) database to compare to their own test results.

Finally, scientists studying biological interactions generate many times the volume of data today than 20 years ago. The toolkit can analyze and manage that data as well.

Best regards,
Hall T.

Friday, August 10, 2007

Chris Anderson of the “Long Tail” Fame – Utilizing Mobile Phones for UAV Control

Chris Anderson the editor of Wired Magazine and author of The Long Tail delivered the keynote address at NI Week 2007. Chris originally wrote the Long Tail as a concept article in Wired Magazine back in 2004 when he noticed that the digitization of media would eliminate the economics that drove publishers to focus on megahits. Once the delivery of media became virtually free customers would have their choice and would thus exploit the “long tail” of additional choices.

Chris Anderson’s speech launched with a statistic that showed a bell-shaped curve representing the 20th century product development. When deviating too far from the mean raised the cost of supplying a product. Vendors develop products that appeals to the mainstream. He then showed a Pareto curve and described the inequality that comes into it. In traditional markets shelf space was limited so only high-volume products were displayed and then sold. With the advent of the internet, shelf space went to infinity and now provides a greater variety. The number of products in the niche segment is equal to the number of products in the mainstream section.

If you can lower the cost of distribution then you can offer more stuff. The internet lowers the cost dramatically. For software you lower the cost of production, distribution and search, and that drives the rise of the “long tail”. As software languages and tools become more sophisticated the production cost goes down. As we move to web-based applications, the distribution cost of software goes down.

The film business is an example of an industry that has reached the limits of its distribution channel. Chris showed a graph that with a dramatic falloff in revenue after the 30th film. The main reason is that the distribution capacity of screens to show films is at most 120 per year. Those films below the top 30 receive reduced revenues due to lack of screen exposure and national marketing.

Chris showed a Globalhawk UAV which cost $123M dollars. From this concept he and his family created a sub $1K UAV using Lego Mindstorms. His nine year old programmed an autopilot onto a remote controlled plane. He then organized a competition similar to a scavenger hunt. Each plane must travel to a prescribed point at which it then gathers new coordinates for the next target to reach. He mounted a cell phone in the plane for communication and used text messages to provide latitude and longitude. He mounted a camera on the bottom of the plane to send back pictures of what the plane sees. The 3G cell phone network provides two-way communication at a low cost. He used infrared sensors on the wings to provide stabilization and then the LabVIEW code provides steering and direction.

In his next project he used a cell phone to control the plane. Everything he needed for the project was already in the cell phone – GPS, onboard processing and storage, camera, and communications. In his flight, he was able to get 3 cm resolution on the images taken with the UAV. It’s interesting to note how the mobile phone continues to come up as a robust and enabling platform.

Best regards,
Hall T.

Friday, August 03, 2007

NI Labs – Emerging Technologies for Virtual Instrumentation Available On-Line

For several years now I’ve posted on emerging technologies and how they relate to virtual instrumentation. Well, now there’s a site that allows users to download and test out some of those technologies. The site is called NI Labs and contains software (mostly LabVIEW code) that NI engineers developed and now share with users to join in the exploration. The experience is similar to Google Labs.

On the site you’ll find several technologies. Each one lists the required software/hardware you’ll need to run it. Most are software-only so no hardware is required for those. If you don’t have the required software, you can find a link to an evaluation version of it. After downloading, you’ll find a Word document that describes the installation and use of the code. Most have installers but a few contain zip files of VIs to place into your folder. Also, upon downloading you’ll receive a link that shows you where to post questions and any feedback you may have.

One of my favorite technologies is the Graphical Timing Software which lets users build diagrams using icons and wires that represent hardware units and timing signals rather than dataflow. This is a mind shift for the long-term LabVIEW user but for those with FPGA-based applications that deal with timing and synchronization, the technology cleans up the diagram and makes for a more efficient implementation in software.

BTW – In April I posted on a new technology called Twitter which let’s users post messages to a common board on the web and enables others to see where you are and what you’re doing. I mentioned this would be a great tool for NI Week given the scale and scope of the conference. Well, this year at NI Week 2007, Twitter is available. You can login here.

We’ll see you at NI Week 2007.

Best regards,
Hall T.