Date: Wed, 27 Apr 2005 16:44:07 -0700 From: Norm Matloff To: Norm Matloff Subject: CNet: "Can Johnny still program?" Lines: 320 To: H-1B/L-1/offshoring e-newsletter Earlier this month I posted an irritable critique of an article in CNet News on the results of this year's Association for Computing Machinery (ACM) intercollegiate programming contest. I was upste not just because it was a one-sided piece which quoted only lobbyists, but also because the article was written by Ed Frauenheim, a journalist whom I've highly respected over the years. Most reporters blindly accept self-serving statements from the industry and education lobbies. Ed has been so good at questioning the industry lobby that I've been quite disappointed in several of his pieces during the last year or so which have taken people from the education lobby at face value. There was a similar article in the San Francisco Chronicle the day after the CNet piece appeared. In my critique, I had speculated that it was no coincidence that two articles appeared at the same time, quoting the same lobbyists saying the same things. The circumstances suggested that the two articles were "plants," i.e. had come about as a result of calls or lavishly printed "educational packets" from lobbyists. Ed later informed me that that was not the case; he had simply seen a discussion on the ACM contest on Slashdot (www.slashdot.org), a programmer bulletin board. The author of the Chronicle article informed me that the genesis of her article had been that her editor had seen the CNet article. I apologize for my speculation, and have posted a corrected version of my critique at http://heather.cs.ucdavis.edu/Archive/ACMProgContest.txt But the essence of my criticism remains intact: The two articles quoted the same lobbyists saying the same things, without any hint that the quotes were self-serving and without any attempt to seek comment from the other side. Ed mentioned to me that he would be writing a followup article, which I'm enclosing below. He has cleverly -- or defiantly? :-) -- parodied the title I gave to my critique in this e-newsletter ("Why Johnnie can't...wait, hold on a minute"). CNet has graciously invited me to adapt my critique on Ed's first article into a column which they will publish. I've been snowed under with a ton of tasks to do, but I will definitely take them up on that. My piece will also comment on the article enclosed below. and also on a related piece by CNet commentary editor Charley Cooper, at http://news.com.com/Can+the+U.S.+still+compete/2010-1071_3-5672106.html I'll save my detailed comments on the enclosed piece for what I submit to CNet, but I do wish to make some quick comments now: Again, I have no problem with the contest itself. It's a fine intellectual exercise. But the "poor" showing by U.S. schools in the last couple of years doesn't mean much. As I pointed out before, and is pointed out in the article enclosed below (I am referred to in the article in the anonymous phrase "Some argue"), the number of teams worldwide who participate in the contest has skyrocketed; that alone would statistically cause the U.S. absolute rankings to go way down. And as both the article here and I pointed out, teams in some Third World countries spend far more time preparing, since victory means so much more to them. (I'll have some specifics of that in my column.) Patterson's comments to the effect that top programmers ought to get accolades like winners of the Super Bowl do are really very silly. Not that I think that wouldn't be a nice idea, but the point is that it hardly explains the American "decline" in this contest. After all, back when Americans were winning this contest, they didn't get invited as guests on the David Letterman Show either. Nor does Patterson's comment that American CS curricula haven't changed in 30 years make sense as an explanation. First of all, he's wrong. Back when he was in school, CS courses he cites were much more theoretical. An operating systems course, for example, would have a substantial math content, maybe even a bit of queuing theory. Nowadays the students in an OS course are involved in heavy programming work. So the curricula HAVE changed, and more importantly have changed in the direction of being intensively-programming oriented. So, contrary to Patterson's hypothesis that the U.S.' poor showing in the programming contest is due to curricula, the curricula have actually become MORE programming-oriented. Patterson wouldn't know that, because he's a hardware guy. (He's also a PR guy. He didn't invent RISC; Cray and IBM did. He merely popularized it. Same for his RAID project cited here.) Moreover, the top foreign universities emphasize that they use American textbooks. Patterson notes that university CS enrollment nationwide has plummetted, largely due to publicity about offshoring. Both Patterson and past ACM presidents have refused to discuss the problem which goes hand in hand with offshoring, the H-1B work visa problem. They say that ACM, as an international organization, can't get involved in an American political issue like H-1B. (As I've reported before, previous ACM presidents have said this too.) That's total garbage. If ACM is supposed to be so neutral, how come this ACM president is publicly bemoaning the alleged lack of U.S. competitiveness? He does that bemoaning both in the two CNet pieces and in the Chron's, and urges the U.S. to spend more money on research in order to keep up with the other countries. If the ACM is supposed to be so neutral, how come it maintains a Public Policy Office in DC? (http://www.acm.org/usacm/faq.html) And again, if you look at their Public Policy Office's agenda, you'll see that research funding is one of the major themes, with the message being, "Give research money to our U.S. universities, so that China and India don't take over." That's "neutral"?! Ridiculous! In other words, this line of "ACM is an international organization, so we can't address the H-1B issue" is just a phony excuse. ACM looks international on paper, and they pay lip service to that notion, but in reality ACM exists to feather the nests of the people who really run it--the academics like Patterson and Foley who want to get more research money, more foreign students (again, cheap labor, deliberately planned that way by the National Science Foundation as I've detailed before), etc. to maintain their empires. Norm Can Johnny still program? By Ed Frauenheim http://news.com.com/Can+Johnny+still+program/2008-1036_3-5675770.html Story last modified Tue Apr 19 04:00:00 PDT 2005 If David Patterson had his way, the president of the United States would congratulate top code jockeys just like the commander-in-chief applauds the Super Bowl champs. That would send a message about the importance of technology smarts and skills, argues Patterson, a computer science professor at the University of California, Berkeley, and president of the Association for Computing Machinery, a group that runs a major student coding contest. "(Our presidents) meet the winners of the football championship, right?" Patterson says. "Gee, wouldn't it be wonderful if the presidents would meet the winners of the programming contest? Wouldn't that be a better world?" After U.S. students earlier this month made their worst showing in the 29-year history of the ACM International Collegiate Programming Contest, Patterson and others are wondering whether the United States does enough to encourage programming talent. The top U.S. school finished in a tie for 17th place. Students from China's Shanghai Jiao Tong University took the top honors, continuing a gradual ascendance of Asian and Eastern European schools during the past decade or so. The last time a U.S. institution won the world championship was in 1997. Some argue the results don't necessarily mean much, given the way foreign schools may put more emphasis on the contest. What's more, the number of entrants has mushroomed, from fewer than 650 teams in 1994 to more than 4,100 this year. Patterson, though, thinks there's more to the U.S. decline--viewed by some as a sign the country's tech leadership is in trouble. ACM's leader knows a thing or two about creating important technology: He played a key role in the development of so-called reduced instruction set computers, or RISC, and was involved in a Berkeley networking project that led to technology used by Internet companies such as Inktomi. CNET News.com recently spoke with Patterson about ACM's contest, the state of student tech talent in the United States, and how outsourcing is affecting the field. What's measured in the ACM programming contest? Patterson: The problems are not simply, as the title sounds like, 'Write a program to do factorial fast.' It's more problem-solving than that. One of them had to do with a cell phone tower. You had to cover this many people, so where would you place your cell phone tower, or something like that. Gee, wouldn't it be wonderful if the presidents would meet the winners of the programming contest? What are the origins of the contest? Patterson: Since ACM has been around so long, it started off as a local programming contest and then it expanded to be nationwide and then international. Education was a big part of the ACM mission. Since computers were brand new, one of the big challenges was going to be to teach people how these use them. So this was kind of a natural thing to do for this volunteer organization. Somebody thought it would be nice to have a programming contest. What's happened in the last 10 or 15 years is information technology has been spreading through a bunch of these countries, many of which do not have great economies and find information technology very attractive. It's not capital intensive. It's a nonpolluting technology. And nations think of themselves as good at things. If you think you're good at math and science--like the Russians, I think, do, and Indians and Chinese--if they think, 'Gee this is something we're good at,' IT becomes a target. As result of your story, I've gotten feedback about how seriously this is taken. I'm told in Russia...the contest couldn't be taken more seriously. That's one of the questions I had: Is it taken more seriously in other places? Patterson: Yeah. As far as I know there's no steroid that you can take to make you a faster programmer. (But) you had the feeling it's almost like East Germany was with the swimmers--it was national pride how well they did in the Olympics. Well, apparently Russia takes it so seriouslythat success in the programming contest affects the funding of local schools. Those that are more successful in the programming contest get more funding. The programming contest is now like (soccer's World Cup). It's not just a national competition. You get to measure yourself against everybody in the world. Do you think that the U.S.'s poor showing--the poorest showing so far--is a reflection of us not taking it as seriously? Patterson: I've been thinking about that. The United States is used to being No. 1 in everything. If we were fourth and it was, 'Oh, the other guys are just trying a lot harder,' that wouldn't be as big a deal. But I suspect that--given that we're 17th--it's more than that. It's not just like, 'Well, this one country is taking it really seriously and so we're never going to beat them at that.' You know, we're 17th. There's a lot of teams from a lot of countries ahead of us. So I think it's more serious. Why did the U.S. do so poorly? Patterson: First of all, I think that'd be a great study. That'd be a great study to see what's happening at these other places versus here. >From the ACM perspective, this is great. The contest is getting more popular, and people are taking it so seriously. It's impressive. The Russians winners--they won it I think a year or two ago--they got to meet (President Vladimir) Putin. They got to meet the leader of the country. Wouldn't that be wonderful if that were true in the United States? What happens with our presidents? They meet the winners of the football championship, right? Gee, wouldn't it be wonderful if the presidents would meet the winners of the programming contest? Wouldn't that be a nice place? Wouldn't that be a better world? That's a good point. Patterson: So I think a sense of national pride (influences how different schools perform in the contest). Some of it is a sense of laziness. We've always dominated the software industry, the computer industry. The United States has always dominated it. 'Why do we care how some amateur contest turns out. We know we're the best in the world at this, so where's the problem?' I think there's some of that. As far as I know there's no steroid that you can take to make you a faster programmer. Filtering it down to the level of the colleges, they may not be pushing it much? Patterson: Yeah. So, I think the question would be, what would it take for us to do well? Suppose this was seen as something that deserved more attention. It does in some sense measure us against how good the rest of the world is at these things. Would more-focused attention bring the United States higher up in this competition? And is it a simple thing or are they teaching differently? (Are) these other countries teaching more effectively? I know as a faculty member I was kind of struck that at least the titles of the courses are the same as when I was an undergraduate student. I took a compiler course. Berkeley teaches a compiler course, you know, 30 years later...Are they actually teaching things in different ways in these other countries? Are they being more successful that way? The other thing that's happening, absolutely, is a decision in these countries to increase their research funding in information technology. They are fairly significant increases, even in countries that don't have a lot of money. They decided that this is a good bet for some of reasons I said--it's nonpolluting, not capital intensive, 'we're the type of people who would do well at that.' And this gets down into the colleges? Patterson: Yeah. I graduated from a big public university and I've spent my life in a big public university. I believe in the big public universities. And part of the big public universities' model is you are able to attract pretty successful faculty, pretty successful people who could be doing a lot of things in part because there's this research side of it as well. So if these faculty are pushing the state of that research, those ideas get into classrooms. And those ideas improve the content of these courses. So, If I'm getting you right, you're looking at this contest result as possibly a reflection of this symptom of flat or declining funding of research, and that may be not energizing the curriculum or energizing the schools. Patterson: Yeah. That kind of goes together. The computing industry is doing less research than they used to. I think over my career what's happened is the computing industry has really driven up performance and prices down. It squeezed the margins out of almost everything...It's amazing what we're doing. But we've concentrated so much in removing margin out of everything that there's not the money around that there used to be to be able to do the research--that funded the Bell Labs and the Xerox PARCs in the past. And then with cutting back on research funding from the government--if both those things are cut back, there's not as many vehicles for pushing the IT research as there was 20 and 30 years ago. I think (the U.S. schools' contest showing) fits in that context. Do you think that, partly as a result of (the research funding situation), the field isn't attracting the best and brightest? Or as many of the best and brightest as it did? Patterson: In the last couple of years, the echoes of the dot-com bust and then this outsourcing stuff has really affected high school seniors' decisions. ACM is working on a study of outsourcing. I don't think the reality is as bad as people imagine. People think, 'Look at our wage scale; look at the weight scales of some of these other countries--it's impossible for the U.S. to compete.' I don't think it's true. I think the reality won't be as dire as people are deciding on their own. But right now in the United States it's affecting what people are choosing to major in. For me at the very height (of the Internet boom), it wasn't such a wonderful thing. We had people taking computer science courses, and they didn't like computer science...There were people who were just doing it for the money. I was just happy for those people to major in something else--become pre-meds or whatever they do. Show-and-tell So some downsizing is fine. But I'm worried right now whether it's actually worse than that, that people are thinking IT--because of their fears of the outsourcing--won't be a factor. And to me it's, 'Boy, we're only 50 years into this field, and the opportunities in this century are just astounding: all the things we didn't do right plus all the new opportunities. It's going to be an exciting field.' Bill Gates gives a talk like that. I don't know how often I agree with Bill Gates, but I absolutely I appreciate him doing that.