Report Calls for Boost in IT Research, Policies

U.S. President Barack Obama's transition team has received a copy of a National Academy of Sciences report that recommends several strategies the United States could follow as information technology research and development becomes increasingly globalized. Randy Katz, a computer scientist at the University of California, Berkeley, who co-chaired the panel that wrote the report, says the top spot that the United States holds in information technology research and development is being challenged. The report suggests the United States would be a better place for innovation if small startups that want to go public did not have to contend with the costly financial reporting requirements for disclosure.[They're talking about Sarbanes-Oxley (SOX)] The patent system should be reformed to discourage litigation, the report adds. Also, more funding is needed for research into the biggest challenges. Katz says the High-Performance Computing and Communications initiative of the 1980s and 1990s is the type of "programmatic research [that can] build communities of researchers that collaborate and also compete while pursuing a particular goal."
Science (01/23/09) Charles, Dan

Related article: Law Profs Decry SOX as 'Debacle,' Express Hope for 'Re-Examination'

HP Grants Aim to Redesign College Engineering & Computer Science

Hewlett-Packard has made more than $2.4 million in cash and equipment available for its "HP Innovations in Education" effort, which aims to reinvent undergraduate computer science and engineering programs through the use of technology. HP is asking for proposals from two- and four-year colleges and universities that offer degrees in engineering, computer science, or information technology. Grant projects must explore the potential for innovation through the intersection of teaching, learning, and technology, with the ultimate goal of "re-imagining undergraduate engineering education." HP plans to award about 10 grants to public or private colleges or universities in the United States. Each grant will include more than $240,000 in HP technology, cash, and professional development. Proposals should describe how the technology will enable innovation in four areas: Leadership Capacity, or creating a global network of administrators and faculty to implement innovative approaches; Digital Learning Environments, or using technology to fundamentally redesign the learning experience in ways that increase student engagement in academic success; the Undergraduate Design and Research Experience, to make engineering real and relevant by involving engineering undergraduate students in design and research challenges that address the needs of society; and Pre-College Outreach, to encourage administrators, faculty, and undergraduate students to work with secondary-school teachers and students to increase student awareness and interest in high-tech programs.
eSchool News (01/20/09)

Operation Comfort Warriors

The American Legion accepts donations for Operation Comfort Warriors, which raises funds to Provide comfort items for wounded U.S. troops recovering in military hospitals.
www.legion.org/ocw
No donations are used for administrative costs.

What's a Serving?

I've been trying to cut back on how much I eat. But in a day and age where it is more common to eat out, I'm not very good at judging a serving. I'm finding it handy to compare food to other objects to judge the serving size. For example, the recommended amount of meat for a healthy meal is 3 o 4 ounces - about the size of a deck of playing cards. Some more examples:

1 oz. meat: size of a matchbox
1 oz. cheese: size of four die
2 tablespoons of peanut butter: size of a ping-pong ball
1/2 cup of pasta: size of a tennis ball
1 medium apple/orange: size of a tennis ball
1 cup of vegetables/fruit: size of a baseball
1 medium potato: size of a computer mouse
1 cup of lettuce: four leaves

You can get a wallet-sized portion guide from the American Cancer Society at:
http://www.cancer.org/docroot/subsite/greatamericans/content/Portion_Control_Guide.asp

Auto Gear-Change Bicycle: Computer Controlled Bicycle Gear Changes Optimize Power, Comfort

Researchers in the Department of Mechanical Engineering at National Defense University in Tashi, Taiwan, are developing a computer system for bicyclists that tells them when to change gears to optimize power while maintaining comfort. The researchers cite ergonomic studies that show cyclists can be in an optimum state while cycling with a fixed output power and peddling speed. The researchers developed an algorithm that provides cyclists with a gear shift strategy to maintain the optimal gear without sacrificing comfort. The algorithm, which has been tested in a simulation of a 12-speed bicycle, provides a gear-shifting sequence with minimal power losses and gear shifts. The algorithm will enable riders to operate the derailleur gearing system more easily, making riders more comfortable because they will be in the correct gear and shifting gears will be smoother. The researchers say the technology could eventually be extended to an entirely automatic mechanical gear-shifting system.
ScienceDaily (01/14/09)

Captivated by R's Power

The R programming language is being used by a growing number of data analysts as data mining is increasingly being used by organizations to set ad prices, find new drugs faster, or fine-tune financial models. The open source language also has become popular among statisticians, engineers, and scientists without a background in programming because of the language's ease of use. "R is really important to the point that it's hard to overvalue it," says Google research scientist Daryl Pregibon. "It allows statisticians to do very intricate and complicated analyses without knowing the blood and guts of computing systems." Statisticians find R particularly useful because it contains several built-in mechanisms for organizing data, running calculations on the information, and creating graphical representations of data sets. Some familiar with R describe it as a stronger version of Microsoft's Excel spreadsheet that can help present data trends more clearly than is possible using information in rows and columns. R also is popular because users can alter the software's code to write variations for specific tasks. "The great beauty of R is that you can modify it to do all sorts of things," says Google's Hal Varian. "And you have a lot of prepackaged stuff that's already available, so you're standing on the shoulders of giants."
New York Times (01/07/09) Vance, Ashlee

NSF Looking for Wicked Cool Visual and Data Analysis Algorithms

The National Science Foundation (NSF) wants to develop highly interpretive mathematical and computational algorithms and techniques to help the U.S. government and private researchers evaluate the data generated by health care, computational biology, security, and other areas. NSF wants to make it easier for law enforcement and the intelligence community to present its data in a visual format, which will require the development of new algorithms capable of representing and transforming digital data into mathematical formulations and computational models that allow for efficient and effective visualization. NSF's research effort is part of a five-year, $3 million project known as the Foundation on Data Analysis and Visual Analytics (FODAVA), which is led by the Georgia Institute of Technology, the NSF, and the Department of Homeland Security. One FODAVA program is a Georgia Tech system known as Jigsaw, which provides multiple coordinated views of large document collections to show connections between entities found within the collection. Meanwhile, the Defense Advanced Research Projects Agency says it wants to develop software capable of capturing knowledge from naturally occurring text and transforming it into the formal representations used by artificial-intelligence reasoning systems.
Network World (01/07/09)

Reply to SciAm Perspective: After the Crash

In the December 2008 issue of Scientific American, the editors wrote a SciAm Perspectives piece entitled “After the Crash,” that focused on making sure that some of the blame for the Wall Street crash is placed on the developers of the software models. The last sentence of the opinion piece really drives home how little the SciAm editors understand about the software models, “Like an airplane, financial models can never be allowed to fly solo.”

What?! The problem wasn’t that the financial models were allowed to “fly solo”. The problem was that the models kept getting changed to appease the developers’ bosses. The SciAm editors also try to make it sound like the developers should have just not made the changes to the models each time “overoptimistic assumptions and faulty data” was given. What should have they done? Just said, “I’m sorry, Dave. I’m afraid I can’t do that.” They would have been fired and then replaced. In some cases maybe they could make a stand and say that making the change is risky and shouldn’t be done without being fired. However, it would amount to the same efforts that Roger Boisjoly made on January 27, 1986, when he convinced his manager, McDonald, to warn the NASA managers not to launch the space shuttle, Challenger, until warmer weather. The push back from the bosses was such that another person was found to sign off on the launch.

I am getting a little off topic. Let us return to this financial airplane that we don’t want to allow to fly solo. Many disasters have been caused by people doing their best to “fix” the problem instead of allowing the established protocols (usually designed using mathematical models) to deal with the problem. I’ll give two examples: Three Mile Island and the incident aboard the drill rig Ocean Ranger in February 1982.

In short, it is wrong to blame the people who work on the product because they are not the people with the power to decide what work is done on the product regardless if the product is a car or a financial model.

Bucket List

Get SCUBA certified: https://www.diventures.com/scuba/dive-certification-costs

Go on an Insight cruise: http://www.insightcruises.com/

---

Winchester Palace, London, UK

November 2, 2017 - Riverside remains of a 13th-century bishops' complex with ruins of great hall, prison and brewhouse. https://www.english-heritage.org.uk/visit/places/winchester-palace

Donald Knuth

Dr. Dobb's Journal published an Interview with Donald Knuth in April of 1996. It was a very good read.

Knuth discusses what distinguishes a "computer scientist" from a "computer programmer" but neatly side steps actually separating the two terms. Choosing to instead explain how computer science is about thinking about problems in a different way. (You'll have to ready the article to see what I mean.)

There is some discussion about Knuth's The Art of Computer Programming. Which Charles Moss tells me is a wonderful read. I'm not sure I believe that but the MSU library has volumes 1-3 in case I want to check them out.

I must quote this because one of my coworkers is just fanitical about C++. To him, everything that is not C++ is crap with the exception of javascript and ruby.

DDJ: You've mentioned Edsgar Dijkstra. What do you think of his work? DK: His great strength is that he is uncompromising. It would make him physically ill to think of programming in C++.

Kunth followed that quote up with some very reassuring comments about trying to find the middle ground between very specific control of code and allowing a library to be used without knowing how it was implemented.

[W]hen you write a program, think of it primarily as a work of literature. You're trying to write something that human beings are going to read. Don't think of it primarily as something a computer is going to follow. The more effective you are at making your program readable, the more effective it's going to be: You'll understand it today, you'll understand it next week, and your successors who are going to maintain and modify it will understand it.