Archive for July, 2012
We’re in the final week of the Computational Freakonomics course at Oxford, and students are looking for data. Several of my students are diving into the Guardian’s impressive open data journalism site. Helping them look around, I found this interesting article relating funding to teaching quality. The findings are all for UK institutions (comparable to US? Similar issues?). The “teaching scores” are not course-specific, but at the end of the three year undergraduate degree, what did the graduates think of the teaching at the institution? I wonder if the influences are the same as on other course surveys. The graph below was one of the most interesting: Higher funding was related to better teaching and student-to-staff ratios.
In the chart below, we seed how teaching scores relate to the expenditure per student and the student staff ratio and how expenditure per student and student staff ratio relate to each other:
Very interesting report from Neil Brown. Here’s the question I’d like to know: So what are students intuitions about computing as they enter the classroom? Are they suppressed or supplanted through instruction? My guess is that it’s different for computing than for science. We live our lives for many years, 24 hours a day, in the real world before we enter school. That’s a lot of time to invent science hypotheses about the world. Not so much for computing. While we may increasing live our lives in a computing world, it’s a constructed, designed world — a world in which the computer science is explicitly hidden. I bet that students only make up theories about computing in times of break down, when they have to invent a theory to explain what went wrong. How often does that happen? What theories do they develop?
The paper title here says it all: Scientiﬁc knowledge suppresses but does not supplant earlier intuitions. A consistent theme across the research described in this post is that when you are explaining science to pupils, you are not adding totally new knowledge, in the way that you might when explaining a lesser-known historical event. When you explain forces to someone, they will already have an idea about the way the world works (drop something, and it falls to the ground), so you are trying to adjust and correct their existing understanding (falling is actually due to gravity), not start from scratch. The paper suggests that the old knowledge is generally not replaced, but merely suppressed, meaning people carry their original misconceptions with them forever-after.
Is there really a shortage of computing professionals? Is it really in the industry’s best interests to bring more women into computing? On the Blog@CACM piece I wrote (before all the comments got wiped), one of the commentors told me that I’m wrong about any shortage of computing majors (and thus, industry’s need to recruit more women), and that I should instead learn from Norm Matloff, who argues that industry is faking the shortage. But a real indication of shortage is rising salaries.
Starting salaries at leading companies for average computer science grads from top schools range from $75,000 to $100,000, plus signing and relocation bonuses worth $5,000 to $15,000, according to venture capitalists and recruiters. New hires may also get small equity grants, with stars getting additional cash bonuses or larger grants worth as much as 1% of the company.
The letter from University of New Mexico CS professor leaving academia is depressing. It rings all too true. Beki’s response buoyed me up, and made me think (yet again, as I did earlier this week) that I do appreciate my job — and being a teacher for me (like Beki) is the best part of it.
An article written by a faculty member who is leaving academia for Google has been circulating on Facebook. I agree in parts with what he is saying, but I am not going to leave academia (I should add that I am a tenured professor, like—I believe—the author of the original piece). Here, for what its worth, are my thoughts. I’ve organized them around some of the same sub-headings.
In our summer camps, two of the most popular activities have been Scratch and Pico Crickets. Unfortunately, the company has been bought out by Lego and is being dismantled in favor of their WeDo, which isn’t anywhere close to the same thing. I’m excited about Hummingbird — I hope that it captures some of the Pico Crickets excitement.
While educational robotic kits traditionally have focused on the technology itself — the building of a robot — Hummingbird treats robotics as just one element that can be combined with craft materials and text to communicate thoughts, feelings or ideas.
“We want students to become inventors of technology rather than users of technology,” said Robotics Professor Illah Nourbakhsh, whose CREATE Lab developed Hummingbird for a project called Arts & Bots. “Hummingbird feeds a student’s natural curiosity about technology by enabling her to incorporate robotics into something she is making that is meaningful or useful.”
The results often amount to kinetic sculptures that use sensors to detect changes in their environment and respond with movement and/or light displays. A cardboard dragon that turns its head and tries to bite anyone who comes close is one example. Students in West Virginia built a working replica of Star Wars’ R2D2.
I’m not really that upset or surprised over the argument that the government did (or did not) fund the Internet. I do realize that it really matters, but it’s a complex issue. The Internet was most successful because of a government and business partnership, which means that there’s always going to be a question of who did what.
As a computing educator, I am more concerned that the article in the Wall Street Journal was so full of conceptual errors! Hyperlinks have nothing to do with the Internet. Ethernet is not the Internet. As my colleague Christine Alvarado said to me on Facebook, the WSJ piece is a symptom of a problem that even educated Americans do not understand the computing in our daily lives.
Crovitz then points out that TCP/IP, the fundamental communications protocol of the Internet, was invented by Vinton Cerf (though he fails to mention Cerf’s partner, Robert Kahn). He points out that Tim Berners-Lee “gets credit for hyperlinks.”
Lots of problems here. Cerf and Kahn did develop TCP/IP–on a government contract! And Berners-Lee doesn’t get credit for hyperlinks–that belongs to Doug Engelbart of Stanford Research Institute, who showed them off in a legendary 1968 demo you can see here. Berners-Lee invented the World Wide Web–and he did so at CERN, a European government consortium.
Cerf, by the way, wrote in 2009 that the ARPANet, on which he worked, “led, ultimately, to the Internet.”
As for Ethernet, which Bob Metcalfe and David Boggs invented at PARC (under Taylor’s watchful eye), that’s by no means a precursor of the Internet, as Crovitz contends. It was, and is, a protocol for interconnecting computers and linking them to outside networks–such as the Internet. And Metcalfe drew his inspiration for the technology from ALOHANet, an ARPA-funded project at the University of Hawaii.
Ian Bogost’s piece (linked below) on Georgia Tech’s involvement in Coursera is biting and to-the-point. ”The fundamental problem isn’t one of cost containment, it’s one of funding—of understanding why the cost containment solution appeared in the first place. We collectively ‘decided’ not to fund education in America. ” Why is Georgia Tech doing Coursera? Why are any of the other schools doing this? He argues that nobody knows, that everybody is doing this because they are trying to position themselves as a member of the elite, as being in the lead. It’s a defensive posture.
Are Universities under attack? De-funding is a form of attack. Why do we have universities, then? What do Universities exist for? Why did we collectively decide not to fund education? Maybe decision makers don’t understand what we do. And the question at hand: do MOOCs replace what we do?
I’ve been thinking about this, while living at one of the world’s oldest and most influential universities. Teaching is not all that they do at Oxford, though I do think that they are particularly good at real education and not just imparting knowledge. The issues of what Universities are for were raised at the C21U launch almost a year ago. Educating students is only part of what Universities do (and there is some question about whether MOOCs education or simply train). But when it comes to education, a research university can provide a unique learning experience.
I love teaching at Georgia Tech’s Study Abroad program at Oxford. The location is amazing, but that isn’t the greatest value of the experience for me–and I hope not for the students. I love the opportunity to interact with students intensively (in class, at meals, on the street, and even in the pubs), to spend every day in the classroom, and to grade everything myself and get a sense for how everyone is doing. All of us GT faculty are here to teach. There’s a community of scholars. I meet weekly for dinner (and often over breakfast) and conversation with a group of similarly minded GT faculty who really care about teaching and students.
For me, the experience informs my research. The intensive interaction with a small number of students is my opportunity to try out new ideas (like worked examples with self-explanations and pixels in a spreadsheet) and inform my intuition about whether or not they might work. It’s the first stage of design-based research: I’m trying to make something work, with small numbers, when I can really see what’s going on. This is more than teaching for me — it’s an intense, immersive, research-informing experience.
I believe that the students are getting something unique out of this, too. Excuse me for being immodest here: This is what I’m good at, and what I’m trained for. This is why I’m a professor. I’m a good teacher, but I also have decades of experience as an education researcher. My students know that I’m trying new ideas out with them. I tell them (in both of my courses) about what I’m trying and why I’m trying it and about my research agenda. Even those students who are “just” taking a first-year-level intro to computing course are hearing about the research context and how it informs what we’re doing. My colleagues who do not do education research also wrap their courses with their research context. Every course is infused with the passion of a scholar who talks about what they study and why they think it’s amazing and fascinating.
This is education that a University can offer, uniquely. My students are learning knowledge and skills and perspectives, in a rich and intense and personal experience. It doesn’t always work so well, I admit. I can’t do the kinds of things I do here at Oxford in our enormous courses in Atlanta. And this kind of education isn’t for everybody. Turadg told us that we need a variety of learning systems for a variety of needs. I definitely have students who are going through the paces and aren’t interested in taking advantage of the whole experience.
I’m damn sure that there is no MOOC that can replace what is going on in my classrooms this summer. Now, society can decide that what I’m offering isn’t worthwhile, or is too expensive, or can be offered to too few students, or may even not as work as well as I hope. Maybe that’s the real danger of MOOCs — it offers something for free (to the students) that seems as good as what a good University education could be, or as good an education as members of our society need.
Maybe what we in Universities ought to do is show people more often what it is that we do and explain why. We need to be able to show people why what we’re offering in a University is better than a MOOC and is valuable to the greater society.
Institutions like mine are afraid of the present and the future yet drunk on the dream of being “elite” and willing to do anything to be seen in the right crowd making the hip choices. The provostial email also notes, “It also is significant that Georgia Tech is a founding member of this group.” Group membership is a key obsession of university administration, and it’s why they take systems lik the US News rankings so seriously. Of course, all such structures are partly fictions we invent to structure our lives and society. The Ivy League isn’t a natural law or a God-given lineage.
In this respect, Coursera’s clearly got the upper hand among institutions that fancy themselves elite: once they get a critical mass on board, the rest don’t want to appear left behind. Given the recent drama at the University of Virginia, whose president was fired partly for failing to blindly adopt online learning only to be re-hired after a PR-nightmare only weeks before UVA announced their participation in Coursera anyway, you can see how Presidents and Provosts across the land might be ready to sign on for defensive reasons alone.
I mentioned awhile ago that some undergraduates built for me a new tool for converting from images to spreadsheets, and back again. It allows us to do image manipulations via spreadsheet tools like Excel. More importantly, it exposes the data abstractions in picture files (turning JPEGs into columns of x,y and RGB), and makes the lower level data malleable.
I’m using this tool in the Media Computation course that I’m teaching this summer. Normally, CS1315 (the course I’m teaching) includes labs on Word, Excel, and Powerpoint, but there’s no sense of “lab” in these compressed courses. And I bet that most of my students know a lot about Office applications already. So I asked them at the start of class: What did they want to learn about Office applications? Several students said that they’d like to learn to use formulas in interesting ways in Excel.
I’ve come up with a homework assignment where students do Media Computation using unusual Excel formulas (e.g., using IF, AND, and COUNTIF). I lectured on Excel on Thursday in support of this assignment, and it was rough. Things that I had worked out in Windows Excel failed or worked differently when doing a live coding session in MacOS Excel (e.g., the FREQUENCY function worked differently, or not at all — hard to tell). Fortunately, we figured it out, but I got a new appreciation of how non-portable the edge of Excel functions can be.
My students are working on this assignment this week, and I’ll let you know how it goes. Based on the questions I’m getting already, it’s challenging for the students. Excel functions are hidden, invisible when you look at a spreadsheet until you click on the right cell. Much of how you do things in Excel, the process, is invisible from watching the screen, e.g., shift-clicking to select a range. So, they’re having a hard time discerning exactly how I did what I did in class.
Maybe they’re learning a greater appreciation for doing all this in Python, rather than Excel.
Another piece, this time in the NYTimes, makes the claim that open education will have vast impacts on the global economy, especially in the developing world. Set aside that it’s very hard for any education interventions or reform to have economic impacts, it’s clear that any education effort has to be broad and touch many people to have an economic impact. Daphne Koller makes a comment in the Chronicle about Coursera “changing the lives of millions of people.” Does it? Will it? Do we have any evidence than any on-line site does? Notice that ALISON (an Irish system described in this piece) admits that the bulk of its learners are in the developed world. There are developing-world users of MIT OpenCourseware, but not a large percentage and those users are mostly just getting a piece of information, not doing long-term studying (as best we can tell from the usage statistics). The results on the new MITx course are just out, and they’re mostly serving US, India, and UK, with 7K students finishing.
What kind of usage would lead us to believe that an open education site is having an impact on the global economy? I completely believe that open education has the potential to have a huge impact. The question is: is it? What measure are we hoping to achieve that would indicate that we’re on the right track? If we didn’t achieve that standard, how do we need to change/improve the model so that it did?
I don’t need job stats or improvements in GDP to be convinced that open education is reaching that impact. Let’s consider the statistics given below. We have a worldwide shortage of 40 million college students, the article says, and it’s probably a much greater shortage in the developing world (e.g., if you count available job openings, you’re not going to count jobs that don’t yet exist but might if there was a dramatic improvement in education and entrepreneurship). How about if one of these sites had 1 million students (which still means it’ll take 40 years to address the shortage), in the developing world, each of which visit the site more than four times in a month, spends more than 3 hours on-line, and actually posts something (homework, feedback to peer students) at least once a week? That feels like a minimum to indicate real studying at a scale great enough to potentially have an impact. Can we find those kinds of statistics for any of the sites? Perhaps for all of the open education sites summed together? At 100K students per course, it’s conceivable that we could reach that goal — if the students stuck around.
If we’re not seeing that, is there something wrong with our models? Maybe there are other factors that we’re not yet identifying that prohibiting open education from having the broad reach that could result in an impact on the global economy.
This is good news for everyone, but it is particularly good for the vast number of people around the world whose job prospects are constrained by their skill levels and who lack the resources to upgrade them through conventional training. It’s a problem that a company based in Ireland called ALISON — Advanced Learning Interactive Systems Online — is helping to address with a creative model.
ALISON provides free online interactive education to help people acquire basic workplace skills. It’s not a megasite. It has a million registered learners, the bulk of whom live in the United States, the United Kingdom, India, Malaysia, the Philippines, Nigeria and the Middle East, where ALISON has 200,000 students. It is adding 50,000 learners each month, but the kinds of services it offers are likely to proliferate in the coming years.
To understand why, we only have to think back to last week, when the big news was the release of the June jobs report, which found that the unemployment rate had stalled disappointingly at 8.2 percent. As always, the story behind that number is more noteworthy than the political spin it gets. According to the Department of Labor, the unemployment rate for people in “management, business and financial operations” is nowhere near 8.2 percent; it’s only 3.8 percent. For workers in “installation, maintenance and repair,” it’s 5.3 percent. It’s workers in certain occupations — like “transportation and material moving” (10.3 percent unemployment) and “construction and extraction” (13 percent) — who are experiencing the most severe economic pain.
That’s because the skills of many workers are increasingly out of sync with the demands of the job market, and the gap is likely to grow, particularly given that only a minority of companies provide formal training to employees. This isn’t just an American problem, however. There are 200 million unemployed people around the world, 75 million of whom are youths, and many lack rudimentary workplace skills — the ability to use a computer, make a budget, communicate in an office environment. According to a study published last month by the McKinsey Global Institute, by 2020, the world will have a surplus of up to 95 million low-skill workers and a shortage of up to 40 million college graduates.
Historian Larry Cuban has a great perspective on the role of technology in learning. His book on “How Scholars Trumped Teachers” is one of my favorites on the history of higher education in the United States. Here is his take on what’s not working, and what is. This was essentially the point that Woodie Flowers was making at the ACM Education Council meeting — new online courses are great for training (getting access to information), and that may be what much of the first two years of undergraduate are about, but real education is more than that, and online courses are probably not enough. I found an interesting piece by Milton Friedman that talks similarly, about the citizenship role of education and the need for the government to support that.
What technology enthusiasts, however, forget, neglect, stumble over — pick a verb — are the multiple purposes of tax-supported schools in a democracy. They and many others futurists err — my choice of the verb — in equating access to information with becoming educated. Even worse, these very smart people ignore the crucial and historical purposes public schools have served in a democracy.
Tax-supported public schools have been and are social, political, and moral institutions whose job is to help children and youth acquire multiple literacies, enter the labor market well prepared, vote, serve on juries, contribute to their communities, think for themselves, and live full and worthwhile lives.
“Everyone pretends.” My favorite piece that I’ve read on Turing in honor of his Centenary. Ian has a wonderful insight into what’s powerful about Turing’s work.
But the computer itself reveals another example of pretense for Turing, thanks to his own theory of abstract computation and its implementation in the device known as the Turing machine. In the form Turing proposed, this machine is a device that manipulates symbols on a strip of tape. Through simple instructions like move forward, erase, write, and read, such a machine can enact any algorithm — and indeed, the design of modern CPUs is based directly on this principle.
Unlike other sorts of machines, the purpose of a Turing machine is not to carry out any specific task like grinding grain or stamping iron, but to simulate any other machine by carrying out its logic through programmed instructions. A computer, it turns out, is just a particular kind of machine that works by pretending to be another machine. This is precisely what today’s computers do–they pretend to be calculators, ledgers, typewriters, film splicers, telephones, vintage cameras and so much more.
Nice piece in Time on the lack of computing education in the United States. In particular, I like that Jane Margolis takes on the myth that students will just learn it on their own without support. That’s thinking that prevents broadening participation in computing
Not every kid has those advantages.
“There is this assumption that if you have this innate talent and you’re drawn to it, you’ll learn it on your own and you don’t really need it at school,” says Jane Margolis, senior researcher at UCLA’s Graduate School of Education and Information Studies and author of Stuck in the Shallow End: Education, Race, and Computing. “Kids that have a lot of resources at home, often with parents with a lot of technical know-how and access to software, people look at them and say ‘Oh, they just take to it.’”
In 2010, the San Jose Mercury News reported that the percentage of computer workers in Silicon Valley that were black or Latino stood at 1.5% and 4.7%, respectively. Girls Who Code, an organization that encourages teen girls to pursue opportunities in technology, points out that only 14% of undergraduate computer science degrees are earned by women.
What an interesting argument in this piece from the Atlantic. As we mentioned previously, Internet technologies cut into older markets if they impact the revenue stream. So far, MOOCs are not impacting the revenue stream. This article goes further to point out that even the for-profits aren’t generating the quality to be much of an impact on traditional universities. Maybe quality doesn’t matter all that much in higher education (as discussed previously), but it does matter if the sense of quality (or lack thereof) impacts success in the job marketplace. If the for-profit’s can’t compete with the non-profits in terms of getting jobs, and they can’t touch the non-profits revenue streams, then there really isn’t much threat from for-profits or MOOCs.
New innovations don’t disrupt old industries by merely competing with them. They do it by cutting into their source of revenue. The music industry ignored the Internet, tried to sue it out of existence, and then let Apple effectively monopolize digital music sales, which gave Steve Jobs the power to set prices at 99 cents a song. Journalism saw its ad dollars whittled away by the profusion of online media outlets and Craigslist.
As George Washington University’s David Karpf has noted, if the Internet is to conquer higher education, it needs to hit colleges in the pocket book. And so far, there’s no sign of that happening.
The simple truth is that nobody has figured out how to build a cheap, high-quality online university. Not even close. So far, the biggest investments in Internet education have come from the for-profit sector, and their results have been, to put it lightly, lacking. For-profit graduates have worse job prospects and earn less than their peers who attend nonprofit schools. A new study released this week suggests that many for-profit diplomas are literally worthless in the marketplace. This even holds true when you control for student characteristics like wealth. And so perhaps not surprisingly, their alums are responsible a disproportionate fraction of student loan defaults.
“The Snowbird Report and the NSF-ED report both make the point that the working environment may not be able to sustain quality: Lab and computing facilities are not being upgraded or expanded to meet the demand; salaries and graduate student stipends are unattractive; faculties have not grown; heavy time commitments to large classes and counseling destroy the intellectual atmosphere and deprive graduate students of proper supervision…On the other hand, there is in Congress sentiment that ‘all the universities must do is raise faculty salaries,’ and the problem will go away.”
No, that’s not from a follow-up to the below article. It’s a quote from Peter Denning’s 1981 Letter to the ACM, “Eating our Seed Corn.” Eric Roberts warned last year that we were going to end up in the same place as we were in the early 80′s (when Peter wrote the above words) and in the early 2000′s (during the dot-com boom). According to US News and World Report, we’re getting there — the flow of students, in a time of cutbacks at Universities, is going to hinder our ability to meet demand, and the relentless draw of industry with its higher salaries is going to make it harder to find faculty.
At some institutions, the computer science program faces a shortage of qualified computer science faculty to meet student demand, notes Gwen Walton, a professor of computer science at Florida Southern College. Walton, who spent more than 20 years working in the industry, says schools cannot compete with the salaries many professionals command in the job market.
“Computer science is one of the few fields where you can start with a very high-paying salary with only a [bachelor's degree],” Walton says. “You don’t go into [teaching computer science] for the pay.”