{\huge D plus O Equals Do}

Our 500th post comes with a non-Federal proposal \vspace{.5in}

Barack Obama is the 44th President of the United States, and the 43rd person to hold that office. Stephen Cleveland counts as both the 22nd and 24th President since he served two non-consecutive terms, and he remains the only one ever called ``Big Steve.'' Our Barry, as he is sometimes called, likes to play up the `O' in symbol and speech, and even named his first dog Bo for his initials. As mavens of asymptotic notation we might like to call him ``Big-O,'' but Obama's many vociferous critics would read the O\/ a different way. O well.

Today we make post number 500, `D' in Roman numerals, and combine that with `O' to frame some remarks made by the President last week about higher education.

Obama unveiled proposals for funding higher education in a speech last Thursday at the University at Buffalo, my campus. It was the first speech by a sitting president at the university since 1853, when Millard Fillmore retained his chancellorship of UB all through his time in Washington, DC. About 7,200 members of the university and the public crammed into Alumni Arena for his 11:15am--noon performance.

After winning a ticket in the faculty e-lottery last Tuesday morning, I kept my family company as they waited in line several hours for the public offering of tickets. We arrived at 9am on Thursday to find the line to go through airport-like security already ringing the entire campus. The people-management was excellent, and we entered about 10:30. My wife and I found two seats together at the back of the rafters, while my kids found space closer down. It was a great atmosphere, a once-in-a-lifetime experience, and attendees were not disappointed, because the clearly-audible Big-O had quite a lot of substance to say.

Obama's Speech

Obama began by acknowledging local political leaders and recognizing the pressing concerns of those in attendance, including UB's football team being sent to face \#2-ranked Ohio State in the college opener. He enumerated how college costs have far outpaced family income, and could have made explicit that they have far outpaced family medical costs as well. This has caused an upsurge in student debt and defaults. He mentioned actions already taken by what he called his ``consumer watchdog,'' who is my fellow 1981 Marshall Scholar Richard Cordray, and who was finally confirmed last month after a two-year battle against Republican filibusters in the Senate.

Cordray used much of his five-time Jeopardy! winnings in 1987 to pay his loans from law school, but few others have been so fortunate. Indeed we joined many families waiting for Congress to continue a program of student loans at under 4% interest, after rates had reverted to 6.8% on July 1st. But action is needed at the other end of the spiraling staircase of costs, subsidies, taxes, and debt, and that is why the President came to a large public university to make his pitch.

The value of university education is not in question. Former UB president Bill Greiner liked to quip that the Master's degree had become what the baccalaureate was in the 1960's and the high school diploma was in the 1930's. My department has many students doing 5-year BS/MS combined programs or double majors, and while such talk of graduate education is getting ahead of the horses, it speaks the impetus toward college degrees as a universal goal all the more loudly. Yet the USA is not in a position to make it a universal free privilege as several European countries have done. We have a roaring undisciplined market, and Obama's proposals recognized that.

Three POTUS Proposals

The first proposal began with an observation that perhaps has been covered in papers on computational game theory and economics:

Right now, private rankings ... [encourage] a lot of colleges to focus on ways to -- how do we game the numbers, and it actually rewards them, in some cases, for raising costs.

Led by Secretary of Education Arne Duncan, who was also present and spoke briefly before Obama arrived, the plan seeks cooperation from universities and colleges on metrics that emphasize the position of students after they graduate: average debt in relation to costs, graduation rates, opportunity after graduation, success in the workforce. Of course many private sources offer such metrics, but the search is for a definitive index to which taxpayer assistance can be tied.

The third proposal was to tie the rate of paying back one's student loan to one's salary: ``Pay As You Earn.'' This brings us to the second proposal, which we mainly wish to discuss. This is to encourage schools to innovate in ways that keep teaching costs down, with the carrot being increased Federal assistance for such objectives.

We're all for innovation. And as you know we have sensed an upheaval coming to higher education in the near future, regardless of anything Washington does. After a couple ideas of credits based on speed of learning and for high-school courses, Obama indeed talked about online education and MOOCs:

Universities like Carnegie Mellon, Arizona State, they’re starting to show that online learning can help students master the same material in less time and often at lower cost. Georgia Tech, which is a national leader in computer science, just announced it will begin offering an online master’s degree in computer science at a fraction of the cost of a traditional class, but it's just as rigorous and it's producing engineers who are just as good.

Well perhaps a little more time is needed to judge that last part. Obama then went into that third proposal and spent the rest of the speech on student loans, but Dick and I want to stop right here and address the innovation part.

Does What You're Doing Count?

Innovation means doing something new. That's often hard, and neads leadership. We have a different idea that's easier to come by. It's using anew something you're already doing.

We see online learning as pushing toward a standardized curriculum, teaching not only 'to' a text but to a packaged set of lectures. We might feel some of our friends doing these lectures are delivering close to an optimal presentation of the material. We'll leave aside our feeling---for which Dick has taken much more time than I to view videos---that some others are far from optimal, and grant that some near-best rolling out of curriculum can be found. We will still suggest the opposite: that maybe the best way to improve your classes is to take time away from the curriculum, to what you yourself doing outside lecture prep.

I teach at a large public university, on the regular rotation for several 1xx/2xx/3xx courses, and I've taken care to tailor my lectures to lift the lower-middle of the student body. I've hence been reluctant to bring up topics that involve jumping ahead of present material. Sometimes I've even given self-limiting responses to advanced questions from one student in a large class, giving that student a longer reply in office hours, and/or posting an answer publicly with a this-is-FYI-not-on-the-exam label. I've come to realize that such jumping ahead works fine provided one thing: that it connects to something you yourself are actually doing.

An Example From Code

For one example, in our intro Data Structures course using C++ last spring, I covered the text's long chapter on arrays (C++ {\tt vector}) and linked lists, explaining using Big-O timing notation the situations where one might prefer one over the other. Toward the end of a lecture, I related that in practice {\tt vector} is almost always better even when the theory favors {\tt list}. I'd read an article to that effect, and this much-noted post puts the point even more strongly. I explained some things about modern hardware, such as caching and pipelined multiple execution, that were not in the text.

Now this has overtones of saying, Everything I Just Lectured On Is False, and I did worry about causing confusion. But what I think made it earnest is that I connected it to how and why I use {\tt vector} copiously and {\tt list} sparingly (for an outermost loop) in one of my own programs. It happens to be my chess-cheating testing program, and this caused some digression in class on how players cheat, but it could be any program needing good numerical performance. Only a few students were curious to see my code, and I showed just a couple the above article, but I think the potential it opened of making a personal connection was reflected in queries about other things by more of the class.

Teaching and Doing

We'll leave our readers to suggest other examples, but we have another thought on how it works. The lower-level the course, the wider the opportunity to make some personal connection. It might at first seem off-kilter, but sometimes the students themselves may find connections for you.

Dick and I have reflected on why college as an institution has done so well despite logistical hurdles on top of costs, and one answer we keep returning to is personal connections---to students and faculty---whose benefits are hard to quantify. I've shown my chess program once in our annual weekend Undergraduate Research Fair, but other times I've had conflicts and not everyone attends and most of those attending haven't had me in a course. Hence the idea of doing some of it in a course.

So Dick and I are suggesting: consider sometimes not teaching the curriculum. Break homogenization. After all, students can now go online to pick up offline some of the bits that were covered less. Perhaps the old cliché about teaching versus doing can be morphed into: ``Those who can, teach what they do...''

Open Problems

Related to what we're talking about are much-discussed topics of ``flipping the classroom'' and blending online content, amid concerns about effectiveness of the original online models. What makes universities effective, what balance with online learning will work best, and how does this depend on a school's own standing and resources?

This is the 500th post on this blog. I wrote the text but the content comes from both of us---indeed I used some of Dick's opinions. Dick adds:

Thanks to all who read us, all who comment on us, and thanks especially to Ken, without whose tireless efforts GLL could not continue. Our next big milestone is not the $latex {1,000^{th}}&fg=000000$ post, but the $latex {512^{th}}&fg=000000$---we are after all computer scientists.

Here is a cartoon that summarizes some of my concerns and thoughts about the future of innovation in delivery of learning via the Internet.