The problem with computer science

I've commented before on the issue of decreasing computer science enrollment, and especially the enduring decline in women opting for a computing career. While many say that our profession has an image problem, the implication is usually that we need to improve our public relations. I've argued that we need to do something about what our field is actually like. The article from The Institute, "Why won't Jane go to engineering school? (Hint: Jane is not dumb.)", linked from the title above, says the same thing, extends the argument to how we educate engineers, and compares this to other professionals.

The Institute is published by the Institute of Electrical and Electronic Engineers (IEEE), the major professional society for electrical engineers. The article, while written about engineering in general, applies fine to computer science as well. Here's the major point:

... engineering school continues to be a dreary and stressful affair. Typical curricula still struggle to include 'all that is important,' and as a result they are overstuffed and unattractive. More study subjects are likely to be crammed into the heavy course list; fewer obsolete old favorites are likely to be retired. The post-university workplace is not much better. Compared with the fields of education or health care, the ethos of the engineering workplace -- long hours, high stress, competitiveness, a 'one size fits all' mind-set -- is uninviting. This is especially true for women, who still carry child-rearing duties in our society much more heavily than men.

The lackluster engineering education experience and the often unaccommodating (and increasingly unstable) engineering workplace have affected men as well as women. During the last 20 years, enrollment in U.S. engineering programs has lagged significantly behind the overall growth in college and university enrollments. There were demographic changes as well-engineering students in the United States are increasingly recruited from communities that struggle to lift themselves into the middle class (most notably, first-generation college attendees and first- and second-generation immigrants).

The author, Moshe Kam, has really pegged the field. Computer science disproportionately attracts students who are pre-disposed to hard, even excessively hard, work. First-generation college students and immigrants have a very clear idea of what life will be like without a college degree that translates into a well-paying job. They're not as willing to take the risk of pursuing a course of study that may not "pay off". I know that's the way I thought of things when I was an undergrad: failure was not an option, nor was the prospect of what I perceived (rightly or not) to be the potential low-paying positions associated with other degrees. This fed back to engineering students' psychology: that students in other majors were either lazy or lacked the "mental firepower" to be engineers. We never thought that maybe they had considered the option and decided that life was just too short and that an engineering career wasn't worth the amount of effort demanded. In other words, the problem isn't with potential students, the problem lies with the computing profession.

Kam contrasts engineering with fields such as law or medicine, and how those fields have attracted women by offering a more rewarding, human, and humane educational and career environment. Is part of the problem that these other professional programs are graduate programs, while computer science/engineering is undergraduate?

When we also tell them [women] (as we do) that in order to be an engineer one must be 'a fan of science and math' and 'juggle projects, lab exercises, and reading assignments' they take one last look at us and flee... the popular CollegeBoard Web site describes several college study majors. The lawyer-to-be will 'engage in intense discussion of thorny legal problems.' The computer engineer? She will 'spend lots of time solving tough math problems.' Take your pick.

Young women are not dumb. The problem is not that they need to change. The problem is that we need to change. In the view of many young people, women especially, engineering represents a collection of majors that promise hard work during college, often in a tense and demanding atmosphere, with the prospect of ultimately gaining a stressful job of questionable permanence. What will help us most is not to say that this ain't so, but to make it so that it ain't.

Topics: careers, computer science.

[+/-] show/hide the rest of this post

The KU soap opera continues

It really is starting to sound like a soap opera. This article quotes Prof. Paul Mirecki saying that he was forced to resign as Chair of the University of Kansas' Department of Religious Studies. The article linked from the title above quotes KU administrators saying that he resigned voluntarily. Though these two stories seem in conflict, they really aren't. Voluntarily acceding to a strenuous request by your boss is still a response to coercion. At most schools, department Chairs serve at the pleasure of Deans, so any refusal on Mirecki's part would likely have been futile anyway. As I stated earlier, the KU Chancellor made some remarks that raise questions in my mind about support for academic freedom there. This is consistent with Mirecki's statements.

I'm sure there will be more happening here, as it also appears that Mirecki's computer has been seized by the police, for reasons they won't state.

Is Today's Engineer lame or clueless?

Along with my most recent copy of IEEE Spectrum, I received a print copy of Today's Engineer Digest, which is supposed to be "the US IEEE members' quarterly digest for building careers and shaping public policy". And all I can say is, "holy crap, can't they spend a minute or two reading the articles before they print the thing out?"

For those who don't know, the IEEE is the main professional organization for electrical engineers, and Spectrum is their main publication. Generally speaking, articles in Spectrum are excellent, not as original research publications, but as tutorials and technology overviews for members who may be in any one of a large number of different EE subfields.

Then there's Today's Engineer. Let me choose two example articles from the December 2005 issue of the Digest: one with a topic that I agree with and one with a thesis I disagree with.

The first article is "A New Frontier: The Privatization of Space". The online article is choppy, essentially a sequence of paragraphs that vaguely follow one to the next but without much in the way of transitions, let alone a coherent overall argument or story. Clearly, the current state and future development of the space industry is of professional interest to electrical engineers. That's about all I get from the article, despite the fact that the author is IEEE-USA's Technology Policy Editor. And the print version in the Digest is much, much worse.

Then there's the "Student Voice" article: "On Social Security". Now, I can forgive a student for not being very clear about the basics of Social Security or investments. After all, it will be decades before he has to consider Social Security benefits (though it would behoove him to learn about investing, including retirement investing, as soon as possible). But I would expect some sort of editorial fact checking to occur before the article sees print. In this case, the author parrots the disinformation he's apparently heard: that "Congress has been spending it [the Social Security surplus] and placing IOUs in the 'trust fund'". I guess I can't really blame him, as he's mostly quoting this article from the Congressional Research Service.

I always wonder if people who write things like that think that Social Security contributions would be sitting in a savings account if not for the spendthrift Congress. Of course, the Social Security system invests in securities that are more secure than a bank account: US Treasury securities, generally considered to be the safest investment in the world. These securities serve as the foundation for many mutual funds with conservative policies and are purchased by the hundreds of billions of dollars worth by foreign investors (it's what they do with all of the trade surplus dollars they get). If, as the author fears, the government defaults on these debt obligations, Social Security will be the least of our problems: it will mean a worldwide economic collapse that will make the Great Depression look like a little bump in the economic road.

Back to Today's Engineer Digest. I'd be interested to hear from other US IEEE members regarding this publication. Have you gained useful information from it? Are you aware of useful public policy work that it has been involved with?

Topics: IEEE, electrical engineering, public policy.

[+/-] show/hide the rest of this post

Digital convergence: death knell for civilization?

For a few years now, many of us have been salivating over the prospects of "digital convergence": technology that will unify all media (radio, TV, etc.) on computer platforms. Well, I'm here to tell you that digital convergence is starting to look like a VCR blinking 12:00 because its clock isn't set.

The story begins with our hero and his family tiring of watching a 15-year-old 29" (75cm) TV from a distance of 15 feet (5 meters). The only things making the situation tolerable were the fact that we don't watch much TV and that, because of our nearsightedness and the distance to the screen, it wasn't so noticeable that the right side of the screen was purplish and the left side was greenish. So, we decided to get a new TV, a bigger TV, a digital TV.

Considering the current flux in HDTV pricing and technology, we decided on a 52" (130cm) rear projection micro-display model, to keep the price down. I'll keep the brand secret, but it's a large Japanese company. the TV is big, and the picture is much better than our old TV. That's the good news. The bad news is that, even after almost a week, there are still some issues with the TV setup that I'm not sure of. That's right, someone with a PhD in Computer Science and almost 30 years' experience programming computers, including Unix system administration and device driver development, has trouble setting up his new TV.

Lest you think I'm some special kind of idiot (well, you may still conclude that, but I'll try to dispel it), let me describe what I've gone through. First and foremost are the problems of the TV user interface and manual. The TV's user interface is composed of a series of approximately 100,000 menu screens, with items grouped more-or-less by topic (by the way, the user interface runs under Linux, not that that really matters in terms of user interface design). The only way to move from screen to screen is to scroll up or down the items of the current screen, one by one, until hitting the end, at which point the next screen will pop up. The screens are arranged in a loop, so, as any CS major will tell you, the worst-case distance to your desired menu item is N/2 button presses, just like a circular, doubly-linked list. The manual even has a listing of the menu screens. Unfortunately, the listing in the manual seems to be four-dimensional, as I still cannot figure out whether the shortest distance to a menu item is "up" or "down".

The TV has four sets of inputs, and each has different capabilities, in terms of combinations of component, composite, S-video, HDMI, analog audio, and two types of digital audio connections. It also has two sets of outputs: one with video and audio and one with audio only (one with a choice of analog or digital audio, one with only analog audio). In my case, I have a older, analog (Dolby Pro Logic) surround sound stereo system. One nice feature of this TV is that its built-in speaker can be used as the center speaker for a surround system, and I planned to take advantage of that. I also have a TiVo with integrated DVD player and a VCR.

I planned out how to hook everything up. Unfortunately, the manual's connection instructions only include a set of "quick start" examples; there's no detailed information on any non-obvious distinctions among the different inputs and outputs. Anyway, I filled in the blanks in my own mind, hooked everything up, and then prepared to enjoy some "Stargate SG-1".

Everything works fine when the video source is the TiVo, DVD, or VCR. But, when the video source is the TV itself, there's trouble. First of all, when the TV is set up so that its built-in speaker will act as a center channel for a surround sound system, no audio is output to the stereo when using the TV tuner. Since the center channel must come from the stereo, that means there's no sound at all. This necessitates going into the menu system every time I switch between external devices and the internal tuner to switch the speaker mode.

Then there's the tuning itself. Our basic cable comes with a limited selection of digital programs, including local stations' HD broadcasts. Our cable company lists them as having three-digit channel numbers, but the TV accesses them via channel/subchannel numbers that seem to have no relationship to the cable channels. My intuition is that the TV's channels are physical numbers, while the cable company's channels are logical numbers. The cable company representatives don't seem to understand what I'm talking about; my emails bring back only repeated references to a URL that lists the channel lineup by "logical" channel number. So, I'm left scanning about a hundred channels myself to figure out what each is.

Meanwhile, my parents' VCR still blinks 12:00, assuming that there has been a power outage since the last time I visited them (which, considering the fact that they live in Florida, is quite likely). I can't see the typical TV watcher being able to install these things, or having the patience to go through their setup. The fact that consumer VCRs have been around for something like 25 years and most are still too difficult or annoying to program for most people bodes ill for digital convergence. And I'm not sure Apple can save us...

Topics: television, consumer electronics, user interface.

[+/-] show/hide the rest of this post

Outsourcing game playing and homeworks

Daniel W. Drezner blogged about the NY Times article linked from the title above. The article is about a new kind of Chinese "factory", in which workers play online games, building up high-level characters and collecting various in-game objects for sale to other gamers.

As they grind through the games, they accumulate virtual currency that is valuable to game players around the world. The games allow players to trade currency to other players, who can then use it to buy better armor, amulets, magic spells and other accoutrements to climb to higher levels or create more powerful characters.

I like the adjective "grind" in the quote above. Remember, these are supposed to be games. If you find playing them to be a grind, then maybe you should find something else to do. If a player needs to have a high-level character to get to the fun parts of the games, then maybe the game companies need to work on their software a bit. You'd think that the games would be designed so that exploring and developing one's character would be fun, regardless of level. Then again, I guess there will always be people who confuse "winning" with "having fun". Unless you're getting paid to do it, why do you care what level your character is?

I was going to suggest that if you have enough money to burn that you spend it on paying someone else to play games for you, it would be better to give it to charity. But, then again, maybe that's what's being done: rich kids in developed countries giving their money to kids (and companies) in developing countries. And they're fostering increased interest in technology and technology-related education in those countries, too.

"It's unimaginable how big this is," says Chen Yu, 27, who employs 20 full-time gamers here in Fuzhou. "They say that in some of these popular games, 40 or 50 percent of the players are actually Chinese..."

"What we're seeing here is the emergence of virtual currencies and virtual economies," says Peter Ludlow, a longtime gamer and a professor of philosophy at the University of Michigan, Ann Arbor. "People are making real money here, so these games are becoming like real economies."

The game companies don't like this, I assume not because of any sense of fairness, but because in the long term it threatens their business. Who wants to play a game when half of your opponents are only in it to develop characters for sale? Then again, maybe the game companies can embrace this, and advertise the percentage of players who are "professionals" as a way to sell their games as being more competitive than others'.

How different is this than students outsourcing their homeworks? This was a subject in a recent Communications of the ACM article, but I had my own personal run-in with it a couple years ago. A colleague emailed me with the observation that he had seen someone advertising a "programming job" on the site rentacoder.com (I refuse to link to these scumbags) that exactly matched one of my programming assignments. I reciprocated by finding similar ads for assignments at other schools and emailing the faculty involved. I just browsed their site, and here's what I found:

• Digital computer design HW C: I need to finish this project due 12 DEC 9 AM EST...
• Technical Paper on Interoperability between distributed heterogeneous data sources and applications: You are required to write the essay of about 2500 words to discuss the basic concepts,techniques, and challenges associated with interoperability... PLEASE ALSO REMEMBER THAT THIS MUST BE AN ORIGINAL PIECE OF WORK. All work will be plagiarism checked.
• easy java homework
• Develop classes in BlueJ: You are required to develop a set of classes for use in an Address List Organiser such as those found in PDA devices...

To get enough information to backtrack where these projects come from, I'd have to register as a developer on the site, as a login is pretty much uniformly required to access the various details. Maybe this is a business opportunity for someone: cruising the web for students outsourcing their homeworks.

Topics: computer games, plagiarism, outsourcing.

[+/-] show/hide the rest of this post

Mirecki resigns as department Chair

There's not much information out there, but Paul Mirecki has resigned his position as Chair of the University of Kansas' Department of Religious Studies. For those of you who aren't keeping score, Prof. Mirecki:

1. Is faculty advisor for an atheist/agnostic student group at KU.
2. Makes informal posts to a Yahoo! group created for that KU student group.
3. Was planning to teach a spring semester course on Creationism/Intelligent Design as mythology.
4. Sent a message to the student group plugging the course as something they might be interested in taking. That message mocked fundamentalists and stated they might not like the treatment of Creationism as mythology.
5. Cancelled the class when the message was leaked and creationists in Kansas had a cow.
6. Was beaten by two men early one morning; those men mentioned the course he was planning to teach.

You can read the AP news article from the title link above; you can read the KU press release (with link to resignation letter PDF) here. I personally don't see bugging Prof. Mirecki to find out anything more; his personal decisions are really none of my business. I do find the comment by his Chancellor that his emails to the student group were "repugnant and vile" troubling from an academic freedom point of view. I wonder if that Chancellor would say the same thing if Mirecki's comments were about atheism or a non-Christian religion. Mirecki's area is Religious Studies, and so his comments are relevant to his scholarship and shouldn't be subject to summary criticism by University officials without due process by Mirecki's peers.

Update: John Wilkins has a thoughtful post about this over at Evolving Thoughts.

Topics: religion, creationism, intelligent design, academia.

Culture wars heating up?

OK, it's probably just a case of a couple of rednecks doing the sort of thing rednecks do (attack people with whom they disagree). But it's difficult to deny that the rhetoric used these days by the "traditional values" pundits (including those who swirl around the manufactured creationism/"intelligent design" "controversy", for example) is often an incitement to violence. In this case, a Religious Studies professor at the University of Kansas, Paul Mirecki, sent out a private email mocking the creationist crowd and describing his plans for a religious studies special topics course that would "...be a nice slap in their [fundamentalists'] big fat face by teaching it as a religious studies class under the category ‘mythology’". Because of the email, he was forced to apologize and cancel his class. Because it's important for an instructor's mind to be a blank slate before teaching a class, I suppose. I guess I'd better think twice about emailing out information on my upcoming algorithms class; wouldn't want the wrong people getting the idea I have already formed an opinion on red/black trees.

P.Z. Myers at Pharyngula says that he may have some additional details to post about this soon.

Making lemonade

I've blogged before about the decrease in Computer Science enrollment. One of the side effects of decreased enrollment is decreased class size. In fact, I've got a class coming up next quarter -- an algorithms/data structures class -- that will almost certainly have fewer than 10 students in it. This raises the question: how would you teach such a class differently, given that the number of students is small enough that there is no need for a formal lecture approach?

I'm still in the early stages of thinking about this, but this is what I've got so far:

• Our classes meet twice a week for 2 hours each time. Have a seminar for one of those periods, focusing on the topic du jour: balanced trees, hash tables, etc. Make a student responsible for presenting his or her best understanding of the topic, leading a discussion for whatever points are unclear or not covered in sufficient depth. Do "mini lectures" for especially unclear matters.
• Instead of having programming assignments targetted at each topic, have one or two larger projects. Do them as a group. If enrollment is small enough, we can just have one group; I could be a member of that group, too. It's a fairly easy substitution, for example, to substitute a search tree or hash table for a database in a web application.

That's all I've thought of so far. I'd welcome suggestions. Please! Topics: teaching, computer science.

TiVo + iPod = Networks + Lawsuits = Stupidity

Just the mention that TiVo was planning to make it easier for owners to transfer video from their boxes to iPods was apparently enough to freak out some TV networks (see the link to a iLounge article, from title). Right now, people can transfer programs to their PCs and watch them there or burn them to DVD. Or, someone with a VCR can tape a TV show. What exactly are the networks trying to prevent? As one of the commenters to the iLounge article says:

Look, I WANT to watch yoru [sic] shows. STOP TRYING TO STOP ME.

Calculator math

I guess I can laugh while reading the post at Tall, Dark, and Mysterious, linked from the title above, because my wife and I plan to live in our daughters' high school classrooms. Seriously. Right now, our younger daughter is in a K-6 program that requires at least 80 hours of parent participation each year. We expect to continue this, resorting to bribery if necessary. And can anyone explain the appeal of a graphing calculator, anyway? If you really have a problem that requires an electronic aid to make graphs, it seems to me the a calculator will not be much help; you'll need something more like MATLAB.

I also plan to subject any boy who comes within 10 feet of either of my daughters to a polygraph and urinalysis. There will be time enough for dating in their 30's...

NSF Grant proposal: Draft Complete

It's amazing what you can do when you single-mindedly neglect your other responsibilities (translation: I've got a lot of grading to catch up on). In this case, what I have is a first draft of the grant proposal. The project description is 22 pages of 12-point type (as opposed to the NSF limit of 15 pages of 10-point type), double spaced. It will eventually get edited down to the right size, hopefully without me having to reduce the point size (after all, it's all about making things clear and easy to understand). The final organization even came out pretty much the same as I anticipated. Since I started out writing this by outlining the plan of what I wanted to do, the budget is worked out, and it can now be circulated within the campus, at least, for comments and approvals. Locally, the most important things are what I commit the university to support (things like course releases, or teaching specific courses in specific time frames) and what my budget looks like (since the university assumes fiscal responsibility).

The feedback I get, including two or three letters of support, will figure into the editing of the project description. Then, there will be the final "compression" down to 15 pages.

Topics: research, grant proposals.

Thanksgiving cooking

It's the day before Thanksgiving, and I'm ready for some serious cooking. I like to slow roast the turkey; I think it comes out much juicier and more tender (falling of the bone). As usual, we went to the supermarket, bought $25 worth of groceries to get the discounted price on the turkey, and then asked for the biggest they had: 24 pounds. At 300° F, this bird would take six to eight hours or so to cook. Instead, I want to cook it overnight at about 190° F: the desired internal temperature when it's done (to ensure all bacteria are killed). The rule of thumb I have from Let's Cook It Right is three hours at reduced temperate for each hour it would have cooked at the higher, conventional temperature. So, assuming I'll roast the turkey for an hour at the start and two hours at the end at high temperature, and using eight hours as the cooking time, I get a total cook time of 18 hours. I'll start it around 10PM tonight. What fun!

If you're going to the trouble of cooking turkey for Thanksgiving, you simply must make your own gravy. There's simply no comparison between home-made gravy and store bought; it's worth the trouble. Here's a recipe modified somewhat from the November 2002 issue of Sunset:

Giblets, neck, and liver from turkey
2 onions, peeled and quarters
2 carrots, peeled and cut into 1-inch chunks
3/4 cup celery
approx. 2 quarts broth
1/2 cup corn or potato starch
Salt and pepper to taste

Chill liver to add later. Put giblets, neck, onion, carrots, celery, and 1 cup water in 6 qt. pot. Bring to boil; reduce heat and simmer 15 minutes. Uncover and stir over high heat until liquid evaporates and contents are browned and begin to stick (15-20 minutes).

Add 1 qt. broth and pepper; stir to free brown bits. cover, reduce heat, and simmer, stirring occasionally, until gizzard is tender (1 to 1.5 hours). Add liver; cook 10 minutes longer.

Strain mixture into a bowl. Remove the meat from the neck and chop the meat, giblets, and liver. Discard the bones and vegetables. Measure stock and add broth to make 1 quart. Combine everything into unwashed pot.

When turkey is done, remove rack and bird; skim fat from drippings. Add 2 cups broth to roasting pan and stir over low heat, scraping brown bits free. Strain mixture into pot with stock and bring to a boil.

In small bowl, blend starch and 1/2 cup water. Add to stock and mix until boiling, 3-5 minutes.

Topics: cooking, holidays.

[+/-] show/hide the rest of this post

Creek Running North: Another career-limiting post

Chris Clarke writes one of the most literate and eclectic blogs out there, and the post linked from the title above is hilarious! I was also impressed by his pruning poetry, though I think December is a bit early to prune fruit trees, isn't it?

NSF grant proposal: Organization

In this entry, I will summarize how I'm organizing my proposal. The primary reason for me writing this is to help me think out the proposal; I'd appreciate any comments you might have. If you find this useful in some way (even as a negative example), good for you. The purpose of a grant proposal is to convey to the reviewers a clear and precise understanding of what you want to do, how you will do it, how it extends the state-of-the-art or the state of our underderstanding of the natural world, and what additional benefits the work will have (what the NSF calls "broader impact"). Call me naive, but I don't view this as a sales job, but rather as an exercise in clarity. The starting point for this is the set of instructions for writing grant proposals; for the NSF, this is the Grant Proposal Guide (GPG), NSF 04-23 as of this writing.

If someone writing a proposal can't even follow explicit preparation instructions, how can you trust him or her with money to do work that is merely promised? This is the question that has run through my mine when I've been reviewing proposals. The NSF GPG contains explicit instructions on the organization and content of a grant proposal. If the NSF program you're applying to says to prepare the proposal according to the GPG, then read the GPG and do what it says. Think of it this way: you are not awarded points for organizational creativity. This is an exercise in efficiently transmitting information to a group of people who may be in fields only tangentially related your own and who will be reading a couple dozen proposals. Have your proposal stand out because it gave the reviewers what they wanted without extra effort on their part. Not everyone does this.

The full proposal should present the (1) objectives and scientific, engineering, or educational significance of the proposed work; (2) suitability of the methods to be employed; (3) qualifications of the investigator and the grantee organization; (4) effect of the activity on the infrastructure of science, engineering and education; and (5) amount of funding required.
    -- NSF GPG

There are some basic requirements regarding formatting. In my case, I use LATEX, and so all of the basic formatting (margins, etc.) is taken care of for me and I can concentrate on content, in this case of the Project Description (everything else is either simple stuff, like a brief bio, or derived from the Project Description).

The Project Description is limited to 15 pages, not including references (which are a separate section). So, I'm not going to present an exhaustive review of the relevant literature, which is a relief, since that would be a very large book in my case, covering neuroscience, information theory, and nonlinear dynamics. As it is, I will need to be brief to cover the basics and still have room for a complete description of my research plan, methods, and broader impact. These are the key parts: what are my objectives for the proposed work, how these fit in with my longer-term objectives, how these fit in with the field in general, specifically what do I propose to do, and how will I do it. I need to address the matter of broader impact "as an integral part of the narrative"; in other words, not as an afterthought tacked on because it's required. In my case, and after some thought and reorganization, the sections will be:

1. Motivation and Background
1. Biological Significance and Motivation
1. Effects of Noise and Jitter on Neural Responses
2. Nonlinear Dynamical Significance and Motivation
3. Biocomputing Awareness
2. Goals and Objectives
3. Detailed Project Plan
1. Physiological Neuron Model
2. Simulation and Analsysis
3. Training and Outreach
4. Project Schedule

Note that the goals will appear twice: once at the beginning as an introduction to the entire proposal and again, in detail, after I've covered enough background and motivated the work. In this case, the background is biological, mathematical, and pedagogical. I need to justify that this is an important biological problem and that the mathematics to be used (models and techniques) are appropriate and interesting. I also need to show that introducing students at my school and in local community colleges and high schools to biological applications of computers will benefit them and CS by showing that it's not all hacking away at a computer by yourself: that computers have an impact in a range of interesting and exciting fields.

Anyway, this has probably gone on long enough. I need to make a pass through the document, get everything in the right place, write introductory and transitional material, and trim down what I already have on methods and preliminary results to the essentials.

Topics: research, grant proposals.

[+/-] show/hide the rest of this post

NSF Grant: Background

In this series of blog postings, I'm going to summarize my process of going from early draft to final product in preparing a grant proposal for the National Science Foundation's Mathematical Biology Program. We'll see if there is any value in doing this after the fact. In this post, I'll go over the background behind the proposal, including both a vague outline of the science (non-technical, so I'll dispense with references) and the basics of my proposal preparation process (such as it is) to date.

It's often said that preparing a grant proposal can take six months to a year of work. I'm sure that some people actually spend that much time focused purely on developing a proposal, but that doesn't seem a very attractive way to spend my time. Instead, I've spent my time (well, yes, more than 6-12 months) actually doing research. At this point, I've got a firm understanding of the biological, dynamical, and computing/theoretical basis for my work, I have preliminary results that demonstrate that I can do the work and that this is at least a feasible line of inquiry, and I enjoyed the process. One of the reasons you go into academia is to do research; why spend maybe 1% of your life merely working on proposing to do research? Of course, if you need a billion dollar spacecraft to do your research, then things (such as your planning time horizon) are a bit different. Computer scientists are, by comparison, cheap dates.

So, what's my research on? I want to understand brains as computing devices. Now, this is a tall order, so I've personally "settled" for trying to understand how very small networks of nerve cells (neurons) do their thing. You may have heard of all sorts of exciting results in the neurosciences, and how researchers are closing in on the secrets of the brain. Couple this with the first, real non-industrial commercial/near-commercial robots, and it really looks like we're close. I'm not so confident of that.

First of all, progress in robotics is misleading, because most of it is in sensors, actuators, and processing power. Little of it derives from any deep understanding of biological information processing. On the other hand, it is certainly true that we have entered a golden age in the neurosciences, with an amazing array of tools that allow us to gather all sorts of information about brain, network, and cell function. You may have seen "pictures of the brain working": functional magnetic resonance imaging (fMRI) that shows how active different areas of the brain are while a person does some task. This is often accompanied (for non-technical consumption) by an article that talks about how much this tells us about how the brain works. OK, I'm contrary by nature, but in my opinion this tells us very little about how the brain works. Yes, it narrows the focus of our inquiry from the entire brain to maybe 10% of it. Yes, it tells us something about which areas are active at the same time, and even sequences of activation in time. But we're still talking about the activity of hundreds of millions, even billions, of neurons. To me, this is just the peeling of the outer skin of the onion.

I don't want it to seem that I'm saying that these investigations aren't worth it or that they don't produce great data; they do. It's just that the brain is so incredibly complex: the most complex object in the known universe. A research data flow of petabytes per year over decades may only scratch the surface of what we need to learn before we understand how the brain works (assuming we're capable of assimilating so great a flux). This complexity may very well extend to the smallest level: while many researchers consider individual neurons to be simple devices (in a computational sense), this is really just an assumption. The fact is that a good simulation of a neuron, including its shape and the interactions of its internal molecular machinery with its external electrical and chemical activity, is a job for a hefty supercomputer. If this structural complexity shows through to a neuron's computational complexity, then we're suddenly dealing a brain as a complex network of billions of supercomputers.

There's so much complexity in nervous systems that entire aspects of them are almost ignored, or at least given second-class status in the search for understanding. Just one example: I've mentioned nerve cells, or neurons, as making up nervous systems and brains. But there's another class of cells that's actually more numerous in our brains: glial cells. They're usually dismissed as serving only structural and physiological functions: scaffolding and waste disposal. But they can produce external electrical and chemical activity. What does it do to our view of neural computation if glial cells play an important role?

Anyway, my research focus for this grant proposal is on error correction coding: looking at how characteristics of the output of one neuron could be used to allow other neurons to recover faster if an error occurs. Think of it as the neural equivalent of a CD still being playable despite the fact that it's scratched. This is small enough scale for me to be able to wrap my mind around it. Whether this is reflective of the true complexity of the subject matter, or merely the complexity of my mind, is another story.

So, back to the grant proposal. I know what I want to do, I can explain why I think it's important, I can relate it to biology and mathematics, and I have a lot of material I've already written about the subject (four conference papers and a journal paper). Writing should be a piece of cake, right? I give myself three months to do it, working part-time, of course (I still have teaching and a journal paper to finish writing while I'm doing this). More on this to come...

Topics: research, grant proposals.

[+/-] show/hide the rest of this post

In defense of academic blogs

Rebecca Goetz, a history graduate student with an "academic" blog, has an article in The Chronicle of Higher Education (linked from title) rebutting the two articles written there pseudonymously by "Ivan Tribble". And what she says makes perfect sense to anyone in academia with an actively maintained web presence: blogs (and the like) are fun, they are an outlet for thoughts that are either not ready for formal (reviewed) publication or not intended for that kind of venue, and they allow us to network with colleagues who we are unlikely to have met in any other way. I have some additional thoughts about academic blogging, and why it shouldn't be controversial:

• I guess it never occurred to me that some people would look askance at blogging because computer scientists have had on-line presences for such a long time. Go to any CS department's web site, and you'll almost certainly find links to faculty sites with personal sides (photos of family, discussions of recreational activities, semi-academic hobbies, etc). Even before the web, academics (grad students and faculty) participated widely in both academic and non-academic oriented USENET newsgroups.
• Blogging is enjoyable, and it's enjoyable writing. Anything that increases the amount of writing we do seems worthwhile to me. The fact that the audience is not as specialized as one's academic peers (to whom one writes in refereed publications) is even better, as it demands a level of clarity and transparency of explanation that might even improve one's "serious" publications.
• One academic fantasy that many of us have involves something like a little cafe filled with relaxed colleagues with whom we can discuss research or other interesting topics: a society of the mind -- like "Cheers", but with stimulants. There may even be actual physical places like that. But a blog, by virtue of being a bidirectional communication medium, can serve this function, too. Yes, it's not the same thing, but the coffee's cheaper and you can contribute your part of the conversation at midnight while wearing pajamas.

I must confess to a case of blog envy: I wish my own blog had more academic content. Now, I don't blog about my teaching and professional experiences (except as abstractions) because I'm not doing this anonymously -- it would be unprofessional to comment on students and colleagues thus. But what about research? My excuse has been, "You can't handle my research." Seriously, I'm not sure that anyone beyond a circle of a few dozen people or so care about my research (if that many).

On the other hand, I keep feeling that a blog can be useful in helping "real" work. Now, at the time of this writing, you would note on the left hand column of this site a set of 43 things entries that include "Finish research proposal draft" and "Finish journal paper". By making part of my to-do list public, I have this hope that I'll meet these deadlines. But I think this blog can be an even greater help, so I'm going to try to experiment with blogging my grant proposal preparation. After all, once it's funded (note I didn't say "if"), it becomes public information. And it's a continuation of research I've already published: I want to disseminate these ideas. So, I'll give it a try: blogging my grant proposal preparation for a more general audience (more general than the folks who would be reviewing it or reading my publications, that is). Now I've said it, and now I'll either do it (and, by way of prerequisite, actually finish the proposal) or face the embarrassment of the ten people or so who read this blog knowing I'm all hat and no cowboy.

Topics: academia, blogging, grant proposals.

[+/-] show/hide the rest of this post

LibraryThing: I'm in the top ten

If you haven't found LibraryThing, and you're into books, or you're just anal like me (I started using 3x5 cards to catalog my books when I was about 10 years old), you simply must check it out. The title above links to the "science fiction" tag in LibraryThing. I'm in the top 10 of quantity of science fiction books. We'll have to see how long that lasts.

Ernie's 3D Pancakes: We're doomed

The title above links to a great quote at Ernie's 3D Pancakes that points out that CS departments are in danger of becoming irrelevant to science and math as those latter departments teach their own computing courses. Why? The quote implies that we computer scientists spend too much time navel-gazing and too little time talking about how CS can impact life (and science and math). In my opinion (expressed as a comment to the linked post), part of the problem is the short shrift given to non-majors courses in CS departments. If we want other departments to look to CS departments for educating their students about computing, then we need to deliver something compelling and non-trivial. That will take hard work.

Of toaster ovens and clock radios

Sometimes, accomplishing even the simplest things can can be a great task. One of these chores can be finding an "ordinary" appliance, the fashion these days being to drive every product towards either the low or high end of the spectrum. A case in point is our recent searches for a replacement toaster oven and a new clock radio.

Previously, we had purchased Black and Decker toaster ovens, because they were reasonably priced and had the features we wanted, primarily being the ability to toast bread (with the occasional cheese or other topping), heat small servings (those too small to justify the use of a regular oven), and clean up fairly easy (primarily a slide out tray and a door that dumps any crumbs on it into the tray rather than onto the counter). But Black and Decker's products seem to have sped their way to the cheapest end of the appliance heap. (I wonder if there's a connection between licensing the company name rather than doing the manufacturing and this?) The most recent toaster oven of their's that we owned had metal parts that fatigued, presumably due to heat cycling, and their current products seemed even cheaper, so we did some more extensive shopping around.

We quickly found that there are almost no "simple" toaster ovens around anymore. If it's primary feature isn't rock-bottom cheapness, then it's an array of dubious features that we'd use once in a blue moon. I suppose if you cook a lot of frozen pizza, then you might want one of the very large ovens that abound that can hold an entire pizza, but that's not a selling point for us. We probably searched for a month before we found a Hitachi toaster oven at Fry's at an acceptable price (more than we originally wanted to pay, but we could live with it, given the unit's quality). We still have to wipe the crumbs off the counter after using the oven; it seems that none of them have the door hinged any more so that the crumbs are dumped into the oven.

More recently, we went on a search for a new clock radio. Our old radio hasn't been receiving radio so well lately: I believe that this is a result of its aging analog tuner. It was about 13 years old, so there was no reason to be upset. I've got some homework for you: try to find a clock radio with a digital tuner. Despite the fact that they come with CDs and a thousand other features, clock radios still by and large come with analog tuners. You'd think that, given the word "radio" in their name, clock radios' tuners would be the subject of major refinement. Why then the analog tuners? And God help you if you don't want a CD player. Often, clock radios being electronics, the products have raced to both the feature laden and cheap ends of the spectrum simultaneously, and so you end up with clock radio CD players that cost $25 and yet have the look and feel of something that should cost $5. Eventually, we found a Sony clock radio on Amazon.com with a digital tuner that was, again, of acceptable price.

I tell you, it's enough to make me not want to go shopping anymore. Oh, wait, I don't like shopping anyway. Well, another reason, then.

Topics: consumer, appliances, quality.

[+/-] show/hide the rest of this post

When will the first direct-to-iTunes TV series happen?

I've been thinking about the failed bid by fans of Enterprise to raise money to essentially buy an additional season of the show after its cancellation was announced. Now that I've seen the movie Serenity and all the episodes of Firefly I've missed on the Sci-Fi Channel, I'm wondering what the economics of a direct-to-iTunes TV series might be. At $2 a pop, how many of each episode does one need to sell to pay for a series like Firefly? If George Lucas could go it alone to make Star Wars, how about a TV series?

Garden photo of the week

Hops
Originally uploaded by stiber.

We were away from our house for a year while I was on sabbatical at the University of Florida, and in that time, our golden hops plant (Humulus lupulus 'Aureus') took over a section of the garden. Luckily, there's not much to worry about there, as it dies back to the ground each year.

A very sparsely populated map

Courtesy of Pharyngula and Crooked Timber, Frappr! provides a map that readers of this blog (both of you) can add themselves to. It's pretty neat, in spite of that exclamation point. Please feel free to add yourself to my map.

Garden photo of the week

Fava bean seedling
Originally uploaded by stiber.

We plant our fava beans in early October each year for harvest in late spring. The new seedlings are just coming up now. It only occasionally drops below freezing here in the Seattle area, so the plants have no trouble (they're supposed to be hardy down to around 20 degrees Fahrenheit, I believe). They'll flower in early spring, filling the garden with a wonderful, sweet scent. In June, shortly before it's time to plant warm weather crops, they'll be ready to harvest.

We like to harvest the beans when they're large and thus need to be peeled after cooking (this is after shelling; the beans themselves have a thick skin). Our usual approach is to quickly blanch them in boiling water, let them cool, then slip them out of their skins. We then saute them in olive oil with garlic, chopped onion, chopped tomato, and whatever seasonings strike our fancy.

I've got a problem

I guess I need to be extra alert now that zombie season is upon us.

You made it. Barely. Congratulations! You scored 48%!

Whether it was the fact that you could run faster, or were just plain lucky, you made it out alive. Even you aren't sure why. But you're sure as hell not going back, or risking your ass for anyone else from now on.

My test tracked 1 variable How you compared to other people your age and gender: You scored higher than 7% on survivalpoints

Link: The Zombie Scenario Survivor Test

Fast reaction to college gender gap

I'm at a conference on engineering education the second half of this week. A good portion of the conference focuses on how to increase participation in engineering by under-represented groups, which includes women. This, of course, has been a matter of concern for years, and the National Science Foundation has taken an especial interest in this, which includes requiring that all grant proposals address the broader impact of the research, of which a major component is to "...increase the participation of women, minorities and people with disabilities in science, technology, engineering and mathematics (STEM)." So, it was especially ironic for me to read the USA Today article linked from this post's title.

Colleges that want to compete for the dwindling pool of men should emphasize male interests, such as sports, he [family therapist Michael Gurian] says, and offer more male role models.

As that article states, female college enrollment has reached an all-time high, with significantly more women than men earning degrees. This is a very recent phenomenon, but already people are saying that something needs to be done about it. I especially like the quote above, since of course so few colleges place much emphasis on sports (for example, we never hear about college sports in the mass media). Funny, but I believe that men far outnumbered women for many years (technically, centuries) before there was any call to do something about that. Meanwhile, enrollment of women in engineering (especially computer science) continues to drop.

Topics: education, academia, engineering, women.

Bill Gates surprises students as "stand in" professor

It seems Bill Gates is touring colleges these days, giving pep talks to students about how exciting technology careers are. The message is spot-on of course: technology careers are incredibly exciting and rewarding. I'm mostly curious why his tour doesn't include the closest (and at times largest, depending on enrollment fluctuations) CS degree program to his office. If I were an egomaniac, I'd think that he's not happy with my previous open letter. If I were a realist, I'd assume he wasn't aware of it...

Dear Mr. Gates,

I'm very sorry that my open letter annoyed you; please take it in the manner in which it was meant: as a helpful suggestion.

Humbly yours,
Mike Stiber

[Thanks to Daniel Lemire for pointing the news item out to me.]

Topics: Bill Gates, careers, information technology.

OrganicHTML for blog

OrganicHTML for blog
Originally uploaded by stiber.

This is what my blog looks like to OrganicHTML. It reminds me of ikebana; I'll take that as confirmation of my own inner beauty.

Accuracy versus precision

Some (perhaps most) people confuse the definitions of accuracy and precision. They are not the same things. Apparently, this is something that statistics majors learn; I know that it is an important part of the scientific computing curriculum. Not understanding this is why many people say things like, "you can prove anything with statistics," when they should say, "you can use a large amount to seemingly precise data (the real precision of which you may be obscuring) to distract the audience from realizing that your statement is wildly inaccurate."

Sometimes, analog is better

These people need a clock. Not to mention a grip on reality. But they could start with the clock.

Garden photo of the week

Grape Leaf
Originally uploaded by stiber.

Nature is giving fair warning: be prepared to rake leaves. Here in the Seattle area, this means while wearing boots and waterproof jacket and pants. At least I now have the hops cleared away from the compost bin.

Blogiversary

This blog is one year old. Time to go to sleep.

I never noticed

My computer was hacked into. Ten years ago. At the time, my main workstation of choice was from Sun, with a Mac used as a secondary machine. I actually didn't initially notice that my Sun workstation was compromised, and even afterwards my first reaction was to just contact the cracker and ask him to be cool about it. Yes, that's old school thinking, though it probably wasn't too far off the mark at the time. The break-in, of course, was the result of amateur sysadminning (by me) and easy availability of "root kits" that allow even novices to break into unhardened machines.

That's no longer the case, as the Infoworld article linked above discusses. Computer break-ins, and software development towards that end, is becoming more professional and more targeted at generating monetary damage. On the other hand, I've moved on to using Mac OS X as my primary work environment. As far as I can determine, there is still no "real" malware for Macs (by "real", I mean software that can do its thing without user intervention -- that doesn't require the user to install the software himself). Maybe it's the benefit of being on a platform with relatively low market penetration. I suspect, however, that there are fundamental architectural differences between Mac OS X/Unix and Windows -- primarily the monolithic nature of Windows that creates single points of failure for security -- that make it much easier to break into Windows machines. I even don't have much of a spam problem, other than the need to periodically check my spam trap to see if any good messages got erroneously flagged.

That's no reason for complacency, however. Never give crackers the keys to your computer. Use the software firewall that comes with OS X (under "Sharing" in System Preferences) and only allow connections that are absolutely necessary. Only turn on network services (also under "Sharing") that you really need, and even at that, only those that use encrypted protocols, like ssh. Use a hardware firewall at home. Have separate sets of "secure" and "insecure" passwords, and never send secure passwords over unencrypted channels. Never use your computer password for a web site. Change passwords periodically, on the assumption that you'll occasionally (by mistake) send a secure password unencrypted across the net. Don't install software if you're not confident of the honesty of its source. Set "Software Update" to check for updates automatically, and install security updates promptly. Have I forgotten anything? Check out this macCompanion article for more.

Topics: computer security, Mac OS X.

Spam, DOS, and cell phones

I've always been surprised at the lack of spam on cell phone SMS. For the most part, everyone who has the ability to receive short messages on their phone has an email address, and there's no fundamental reason that spammers couldn't send messages to lists of cell phone numbers. I assume that the phone companies employ spam filters, so at least long messages and messages that contain HTML get rejected. But maybe the spammers have decided that short messages to cell phones doesn't pay in terms of money received (for one thing, there's no connection between reading the message and getting on a web browser to give the spammers your money). The story linked from the title raises another issue, which is the launching of denial of service (DOS) attacks on cell phone networks. By sending messages from the internet fast enough (easily achievable with a single computer), voice traffic can be halted -- a result of the dual use of network control channels (the ones used to set up voice calls) to also carry the short messages (because their bandwidth requirements are expected to be low).

The take-home message, as always, is that a cell phone is a luxury -- there's no substitute for a land line, switched network telephone in an emergency.

On the death of meritocracy

According to my naive way of thinking, one of the fundamental precepts of modern university education is that of meritocracy -- that students can control their futures by excelling at their studies, and that being better at things that matter will be the key to success and happiness later in life. Now that it has been a few days and it still seems that President Bush has nominated a female version of Michael Brown to the Supreme Court (OK, and former lottery commissioner instead of horse show official), I have my face rubbed in the ugly reality: who you know really does matter more than what you know. Choose your path through life based on doing things that will let you meet the right people, rather than provide you with the best educational and life experiences. And pick your parents wisely (sorry, kids).

Links: Daniel Drezner has some interesting links, while Michael Bérubé provides some comedic relief.

Serenity

I always had mixed feelings about the TV show "Firefly". On the one hand, it had great chemistry among its characters (something that "Enterprise" sadly lacked). On the other, I've never been a fan of "horses and spaceships" space opera. In any event, the point was moot, as the show was inexplicably cancelled before any reasonable person would expect a large audience to grow.

So, I was very curious to see the new Serenity movie, and I must say that it's one of the best SF movies in recent years. (Notice that SF doesn't include Lord of the Rings.) In many respects, I'd put it up with the best of Star Wars: it has a great feeling of concrete reality to it, the characters' interrelationships are interesting and the viewer cares about them, there are actually reasons behind the things that happen (someone thought about more back story than appears in the movie, though I wish someone had spent about five minutes thinking through the movie's "geography".), and there is a point to the whole exercise. In some ways, it's better than Star Wars: more mature, three-dimensional characters, a palpable feeling of risk (that characters aren't necessarily safe just because they're main characters), and a ending that is satisfying beyond the old standby, "those bad guys blowed up real good".

I think I'll buy the DVD version, which is something I don't do very often (excepting the kids' movies).

Links: review at Wired, Internet Movie Database, official movie web site.

Topics: movies, science fiction, space opera.

Petition for scientific support for Darwinism

The link above is to a petition on an archeology web site asking for signatures from scientists whose work depends on evolutionary science. They originally called it a "four day petition", to show how many signatures of scientists they could get in that span of time, given that the Discovery Institute had a petition that took four years to get 400 signatures. The four day petition got 8000 signatures.

It's still going strong, so sign away. How does evolution affect a computer scientist? In my case, I work in the area of computational neuroscience, studying (among other things) synaptic transmission across the inhibitory synapse at the crayfish slowly adapting stretch receptor organ. Because of common descent, the neurotransmitters and ion channels in crayfish neurons are also in human neurons. For these, and a number of other reasons, we expect that the things we learn about information processing in invertebrates will also apply to vertebrates, mammals, and humans.

As another example, I'm making evolutionary computation the theme of the introductory computing course I'm teaching this quarter. The success of stochastic optimization algorithms like these provides strong mathematical support to the idea that the combination of randomness, inheritance, and selection can produce solutions to complex problems.

Do what I say...

I always tell my students to think before coding: that time spent designing and documenting their programs is time well spent. And, I mostly follow that advice. And then there's the occasion that I don't, and I'm reminded why documentation is such a good idea (actually, I'm not reminded so much as driven crazy by the fact that code I once understood is now utterly incomprehensible to me). A case in point comes from my expert systems class, which I am preparing to teach. (Lest you think it's a bit late, my school is on the quarter system, which truly sucks, but which means class doesn't start for a few days yet.)

In this class, I use the JESS expert system shell, which is written in Java. One of its very nice features is that it can easily be integrated into other Java software, and that's what we so in this class, using the expert system to build a player for Wumpus World. Wumpus World is an elaboration of the old text adventure, Hunt the Wumpus, in which players wander around, looking for gold, avoiding falling into pits, and avoid (and/or try to shoot) the Wumpus. In this case, we use a Java version of Wumpus World that includes a server that supports multiple client players across a network. The original code included a very simple example player. The students' task is to develop a more sophisticated player using JESS.

Now on to my stupidity. I spent some time puzzling out some of the Wumpus World code and figuring out how to interface JESS to it. Two years ago. I always intended to document what I figured out, but "didn't have time" to do it right away, let it sit on my desktop for a while, then things came up, and yada yada yada, it's two years later and I haven't a clue how the code works. Now I wish I had spend what would have been perhaps an hour or so writing up some documentation. I remembered that there were two versions of the code, corresponding to the two ways that JESS can be interfaced to Java: with the server communications encapsulated as an object that is manipulated from JESS code or with the JESS engine as an object controlled by a modified Java player. That's about it, until I had invested another half day working out the details. We all need to be reminded of the benefits of good work habits from time to time. I wonder if my return to meticulous documentation will last longer than my post-dental checkup flossing regimen...

Topics: software development, documentation, expert systems.

Trials and Tribble-ations

I just couldn't resist this opportunity to use the second Star Trek tribble episode title in a post (having used the first here). The Chronicle of Higher Education's "Ivan Tribble" is back, to remind us that there are some departments we really don't want to work in (i.e., his). He's once again out to save academics from themselves, telling us not to blog because it might reveal something that someone else doesn't like. My advice: worry more about your professional writings' impact on your career. Not that I blame Prof. Tribble for using a pseudonym...

Much better commentary about this matter can be found here, here, here, and here.

Topics: academia, jobs.

IT employers catching a clue?

Cynicism knows no bounds for Dilbert aficionados, but it just may be that employers are finally starting to wise up to what their IT workers want. As the linked Computerworld article discusses, people rarely gain a feeling of satisfaction from salaries or bonuses.

Compensation is more a dissatisfier than a satisfier
   --Brian LeClaire, VP & CTO, Humana Inc.

Far more important is that employees feel that they are valued by and valuable to their organization, and this includes feeling secure that they won't be fired as soon as the company decides that outsourcing their job will save a penny. The other thing they want is reasonable working hours. But (the main point of the article), to know what people really want, you need to do one special thing: ask them.

Topics: profession, careers, information technology.

CS enrollment: no worries

As I've written about before, I think all the fuss about low enrollment in computer science is overplayed. Enrollment will pick up as the job market improves, and in fact it already seems to be improving, as the El Paso Times article linked from the title above states. Regardless of outsourcing and all the other worries, computing employment within the US will still grow robustly over the next several years. Like other engineering-related fields, CS goes through periodic booms where enrollment shoots up dramatically as the field becomes hot, then drops precipitously when the hype cools off. It always recovers because the field is integral to the development of new and improved products and processes. As long as such development is an important part of our economy, demand for computer professionals will remain strong (neglecting short-term fluctuations).

So what does this mean to you if you're thinking of a college degree? First of all, follow your interests. Who knows what the employment situation will be like in four years? Or, for that matter, ten or twenty years? But, hopefully, you will be around and be happy. Doing work you don't like is not a recipe for happiness. Second, right now is the best time to major in CS. Enrollment is bottoming out, so you'll be part of the smallest graduating class in a long time. By the time you graduate, most of the computing professionals who lost jobs will have found new ones (or moved on to other employment). Just like many other areas of life, it sometimes pays to be a contrarian.

Garden photo of the week

Pear basket
Originally uploaded by stiber.

Nothing can match the taste of home-grown fruit. In particular, asian pears really should be ripened on the tree, but at that point they bruise too easily to ship. So, you'll never be able to get asian pears at a supermarket that are as good as these. Go plant a tree!

Garden photo of the week

Rudebeckia fulgida 'Goldsturm'
and Cimicifuga ramosa 'Brunette'
Originally uploaded by stiber.

I've got a garden again! It was mostly neglected during my absence, but the advantage of living things is that they often can take care of themselves. I've certainly got a lot to do to clean things up, but parts of the garden are amazingly mature and filled in now. My garden photo of the week is from my perennial border, with contrasts in color and similarity in the spiky rudebeckia flowers and cimicifuga leaves.

The "after" picture

Mascot after the trip
Originally uploaded by stiber.

Here's what 6000 miles can do to an antenna ball. Now that I'm back home and at least can see the light at the end of the unpacking tunnel, this will once again become an active blog.

6100 miles

And roughly 300 gallons of gas. I must admit that there were times I doubted the wisdom of the drive, and wondered it might not have made more sense to sell everything, including the car, and just fly back home. Luckily, there were no mishaps or mechanical difficulties on the way. Nope, the disaster was waiting for us at home, where the folks renting our home decided to violate their lease and keep a cat. I'm very allergic to cats. We hoped that a thorough cleaning and steam cleaning of the carpet would remove the bulk of the allergens. Then the cleaning folks said the fateful phrase, "Your know, you have fleas." And now, the battle lines are drawn, the carpet is sprayed, we've purchased yet another $36 can of flea spray, we scratch, and we vacuum (every day). Welcome to our return trip disaster.

On the dinosaur trail

Fossils
Originally uploaded by stiber.

T-rex jaw
Originally uploaded by stiber.

After the museums in Washington, DC, the museums on the Montana Dinosaur Trail may be underwhelming. But, the various paleontological field stations won't be. There, you can see the real fossils, both those still in jackets awaiting extraction and cleanup and those already prepared. You can see all of the equipment used to prepare them and talk to someone who actually does that work. This is a great experience, very different than what you get at a museum. Federal and state highways in Montana are usually undivided and low traffic, and have a speed limit of 70 miles/hour, so with a little extra care, the field stations are as easy to reach as destinations on the interstates.

Daniel Lemire's blog: Met with Michael Stiber today!

Bison and Capitol
Originally uploaded by stiber.

I'm catching up with a little blogging while I should be sleeping, here from a Super 8 motel in... oh, yeh, Bismarck, North Dakota. During my recent conference stay in Montreal, I met with Daniel Lemire. He's better looking than his blog photo would seem to indicate. And perceptive, too.

Where shall we eat?

Quebec
Originally uploaded by stiber.

It's always fun to wander around a new city looking for a place to eat. Well, as long as you're not hungry. If you're hungry, you'll almost certainly end up paying too much for mediocre fare at some touristy place. In that case, it's much better to ask someone who lives in the city for the name of a restaurant that they like to go to.

We didn't find such a place in Quebec city, but in Montreal, try:
Au Petit Extra, 1690 rue Ontario Est, Montréal, Québec H2L 1S7, (514) 527-5552. Dinner will cost around C$20-25 (face it: you won't get a decent dinner in downtown Montreal for much less than that).

Restaurant Europea, 1227 de la Montagne, Montréal, Québec H3G 1Z2, (514) 398-9229. Their 9-course dinner costs around C$70, but you can get a 5-course lunch for less than C$25.

We still don't know why hotels and motels in eastern Canada are about twice as expensive as those in the US (US$80 or more for Travelodge-level quality, which isn't saying much). We never noticed that in British Columbia.

Niagara Falls

Horseshoe Falls from below
Originally uploaded by stiber.

American Falls
Originally uploaded by stiber.

Ignore the run-down area on the American side and the shlocky shops and mediocre restaurants nearby on the Canadian side and concentrate on the Falls themselves. Access to the tunnel under the Falls on the Canadian side is definitely worth it.

Afterwards, take a drive to the canal linking Lakes Erie and Ontario and watch the ships move through the locks. There's an information center near lock 7 and a museum near lock 3. The drive between the two lakes is about a half hour.

NY State Thruway

NY State Thruway
Originally uploaded by stiber.

An amazingly transparent blue sky with fluffy white clouds, on the way to Niagara Falls. One of the things that makes the driving worthwhile. Are we there yet? Who cares!

Trip mascot

Trip mascot
Originally uploaded by stiber.

A gift from a friend in North Carolina, this is not intended as an endorsement of any company, just as a smiling cowgirl.

The best museum

Enterprise
Originally uploaded by stiber.

Maybe the best museum in Washington, DC (and certainly the most under-appreciated) is the National Air and Space Museum Udvar-Hazy Center. While it has various educational attractions like its downtown sibling, the Udvar-Hazy Center is more like an art museum for aircraft and spacecraft, packed with everything from the earliest attempts to build aircraft to the Concorde, an SR-71 Blackbird, and the space shuttle Enterprise. It's out at Dulles airport. When you go there, be sure to take the 90-minute guided tour, which is like a docent's tour of an art museum.

An open letter to Bill Gates

Dear Mr. Gates,

I read with interest a recent Computerworld article, which discusses your worries with respect to declining enrollment in computer science degree programs in the United States. As a computer science faculty member in the closest CS department to Redmond, this is a topic which is also close to my own heart.

I agree that the recent precipitous decline in CS enrollment is something that should give one pause. I would go further to say that the longer-term drop in the percentage of CS degrees awarded to women is far more troublesome. Furthermore, I believe we see eye-to-eye regarding the perception of CS careers as unpleasant and isolating as being at least part of the cause of both. Where I part company with you appears to be that you believe that this perception is unfounded and I see it as evidence that college students are getting, at least to some extent, an accurate read on the state of computing careers.

Let's face it, unless things have changed drastically in Redmond while I've been away this past year, your technical employees (and those of other companies; this is not unique to Microsoft) put in far more than 40 hours per week. It doesn't matter how interesting that work is; I submit that there is something wrong with an industry that expects its workers, as a permanent state of affairs, to work more than the accepted standard work week. And I think students agree with this and are voting with their feet.

I understand that changing this would have an impact on your company's productivity. However, I also understand that you have been pushing, along with other industry leaders, for increased immigration quotas for technology degree holders. Don't you see that this might be viewed in an adversarial fashion by prospective CS majors? That they would perceive the primary motivation behind this a removal of pressure from technology companies to provide real fixes to CS workers' quality of life? It's true that immigration provides great benefits to universities, companies, and US society, as some of the best and brightest from around the world come here. But there is a difference between bringing in some people to enhance an organization and wanting to bring enough to materially affect the work environment.

In closing, I think that you have an opportunity to set an example and show that your interest here is not merely Microsoft's bottom line -- that you really have a concern for the long-term health of the computing profession and the student pipeline. Why don't you make the 40-hour work week official Microsoft policy, and back it up with measures such as official comp time? Why not evaluate managers poorly if their teams consistently put in more that 40 hours/week on average? Why not put other teeth behind this, to ensure that hours are accurately accounted and that comp time is actually taken? Just because an employee feels that life away from work is important doesn't mean that that employee isn't serious about his or her career. Those are the people who are more like your typical user: the people who may understand how to, as you say, "make things simpler".

Topics: careers, computer science.

[+/-] show/hide the rest of this post

Google Moon - Lunar Landing Sites

Yet another example of how inventive a company Google is. Go to the link from the title above. For fun, zoom all the way in.

Welcome Skeptics!

The thirteenth Skeptics' Circle is being held at Respectful Insolence, and I am participating via the article linked above. I hereby provide this post for your commenting convenience, as the above-linked post already has a few long ones.

A day of mourning for all engineers: James Doohan 1920-2005

Although he may only have played one on TV, James Doohan's portrayal of Montgomery Scott on Star Trek was an inspiration for many of us when we were forming our visions of what we would do with our lives. If you're an engineer, the next time you need to make an emergency design revision, I hope you'll add a comment in the documentation: "This one's for you, Mr. Doohan".

Motel blogging

Washington Monument and Capitol
Originally uploaded by stiber.

I guess there's no such thing as a real vacation anymore, as more and more hotels and motels offer high-speed internet. No rest for the compulsive...

Myth of for-profit schools' CS degree programs spreads

I blogged previously about a report from the American Association for the Advancement of Science (AAAS), which, among other things, states that two for-profit "universities" produced the most CS degrees in 2001. The report has now been picked up by MSNBC (link from title above). I pointed out a number of problems that this report has, in my opinion. The biggest one is this: the two schools mentioned don't even offer CS degrees.

Topics: academia, computer science, information technology.

Traveling light

Fully loaded car
Originally uploaded by stiber.

An interesting thing happened while packing for our return trip from my sabbatical. We knew we had bought a thing or two while in Gainesville. Then we started packing. The car-top carrier is from our trip to Gainesville a year ago. The luggage rack mounted at the back of the car is new. The things on the rack came with us. But, even with all that stuff moved to that rack, and even after shipping something like twelve boxes of books, the interior of our minivan is still just above the critical density required to initiate self-sustained nuclear fusion. Let's hope the transmission holds up...

On the road again

Well, all good things must come to an end. In this case, it is the end of a sabbatical. Most of the pain of packing is done -- the minivan is as dense as a neutron star. Now comes the drive from Gainesville to Seattle, with a stop in Montreal for a conference. Expect intermittent blogging, hopefully without any stories about en route transmission rebuilds (that was part of the trip at the start of the sabbatical; Chrysler transmissions leave a bit to be desired).

U.S. losing lead in science and engineering?

I'm trying to remember if there has ever been a time in the last 25 years when there wasn't some study or article warning about the shortage or lack of US engineering graduates and the dire consequences of this for the US economy. In this case, the article linked from the title is from Reuters and the information source is a study conducted by the National Bureau of Economic Research, which sounds like a government agency, but is really a private organization. Right away I'm suspicious, for why would people name an organization that way unless they wanted to convey the false impression that they are a government agency? Do they take the same approach with their studies? However, to be fair, my cursory perusal of their web site did not leave any impressions of something underhanded; they seem like a serious organization with a number of high-level researchers. So, I'll leave the issue of their name as a minor point; perhaps someone reading this knows more (assuming anyone is, of course)?

For example, the author of the study holds a named Chair in the Harvard Economics Department. The report would cost $5, so of course I just read the Reuters article and not the full report. I assume Reuters just grabbed some bullet points from its executive summary or press release, anyway (the article reads that way). The gist is that, despite its dominance of engineering and high technology industry and research for the last 50 years, the United States' share of world engineering production is shrinking, and, urgent action is needed to ensure that this does "not undermine America's global economic leadership".

Note that this is talking about all engineering graduates, not just CS, and so it is comparing a currently stable US output with rising outputs in other countries. This is not surprising, considering that:

The study said deteriorating opportunities and comparative wages for young science and engineering graduates has discouraged U.S. students from entering these fields, but not those born in other countries.

So, what is to be done? Produce more graduates to compete for fewer opportunities and lower wages? It doesn't seem that the report suggests this action; at least, the Reuters article doesn't mention it, and the wording is strange, in that the action urged above is to prevent the decreasing share from undermining US economic leadership, not to prevent the share itself from decreasing. The report's abstract has the following closing sentence:

To ease the adjustment to a less dominant position in science and engineering, the US will have to develop new labor market and R&D policies that build on existing strengths and develop new ways of benefitting from scientific and technological advances in other countries.

This seems purposefully vague. Is there something more concrete in the report? Is this code for something that would be obvious to an economist, such as increasing immigration quotas? Anyone want to know badly enough to spend $5 to find out?

Personally, I am skeptical that this will be as much of a crisis as is advertised. Of course the number of engineering graduates in China, for example, will increase greatly in the years to come, and so of course the US share of graduates will decrease. However, the cost of living in major Chinese cities already rivals or exceeds comparable US cities. For example, new condos in the outskirts of Guangzhou can cost well over US$100,000. Chinese engineering graduates aren't stupid. They didn't work like crazy to beat incredible odds to get their degrees to live in third world conditions someplace in the boondocks -- they want comfortable lives in the big cities. I don't imagine that the situation is all that different in India. So, the cost advantage of offshoring will be transitory: enough to jump start high-tech industries in developing countries, but insufficient to do permanent damage to the US engineering profession.

Topics: engineering, profession, careers, globalization.

[+/-] show/hide the rest of this post