Monday, August 25, 2014


OER, ERP’s, and an Idea

Libby Nelson has a thought-provoking piece in Vox about why textbook prices keep climbing so quickly.  It’s worth a read, not least for the point that, as with tuition, prices do not necessarily equal spending.  As prices have climbed progressively higher, students have become more vigilant about finding alternatives, whether through rentals, used purchases online, or other, more ethically ambiguous means.  

Unfortunately, many of the alternatives to purchasing -- other than just going without -- are more available to students who have more social or economic capital from the outset.  The thicker and more educated your network, the easier it will be to find friends who will lend copies.  Some campuses have relatively robust rental programs, but many don’t.  (From an institutional perspective, it would be easier to control book costs if departments would select standard texts across every section of a given course.  But they prefer not to.  As long as each professor can choose her own instructional materials -- a stipulation in our collective bargaining agreement -- certain potential economies of scale are off the table.)  And some of the cost-saving mechanisms that can work tolerably well for high-enrollment introductory courses aren’t nearly as effective for lower-enrolled, more specialized classes.  

The blindingly obvious answer, I think, is an industry-wide drive to make high-caliber Open Educational Resources available, starting with the developmental and freshman levels and working our way up.  Platform-independent OER offer the promise of sustained affordability, and of better accessibility for students with disabilities than many commercial publishers offer now.  

A serious push towards OER would require a significant chunk of sustained external capital.  Even if the resources are free to students, they’re still costly to produce and maintain.  Publishers have no interest, since OER violate their business model.  But much of the infrastructure of publishing would still have to exist.  As many of us discover painfully in graduate school, manuscripts don’t edit themselves.  In highly technical subjects, the level of formatting and proofreading involved is not to be sneezed at.  

A similar dynamic is at work with ERP systems.  ERP systems are those back-office IT systems that manage class schedules, student registrations, academic records, and the like.  A few major commercial providers dominate the market, and from my non-technical perspective, they all basically suck.  But no single college (outside of maybe Harvard, which probably doesn’t feel the need) has the resources to build its own.  So we wrestle with vendors who keep buying each other.

I had great hopes for Kuali, the open-source embryonic ERP, which was why the story that it’s becoming a for-profit vendor itself was so disappointing.  Rather than developing as the badly needed alternative that it could, it’s likely now to become another variation on Banner.  

In both cases, the missing element is a deep-pocketed consortium dedicated to solving a problem on the ground.  For-profit publishers won’t just give it away; they’d go out of business.  And for-profit ERP providers have an interest in a cycle similar to textbooks: lots of updates, high maintenance, and ever-higher cost.  

Non-profit consortia are notoriously hard to organize.  They tend to fall prey to the “free rider” problem, in which each individual (or college) makes the calculation that its own contribution is far less crucial to the success of the collective than to its own budget, so it’s individually rational to let others carry the weight.  When enough people do that, the collaboration collapses.

If only there were some sort of large piles of money sitting around, with some sort of social purpose, but without a mandate to make a profit.


If only such things existed, they could form the foundation of broadly beneficial investments.  That’s a good word: we could call them “foundations.”  Foundations with deep-pocketed, tech-savvy benefactors could choose to make the initially unglamorous investments that would save millions of students significant money for decades to come.  By virtue of their structure, they could get around the free-rider problem.  

Last week Paul LeBlanc proposed a national college degree that would form the baseline against which colleges would have to show value.  I’m thinking this is the least controversial, lowest-hanging fruit in that proposal.  It’s a prerequisite to the NCD, but it could also stand without it.  Cutting the costs of back-office operations and textbooks may not stir the blood, but it could make real and sustained differences both for existing institutions and for emergent or potential ones.

Or, we could just keep publishing rankings and talking about pipelines.  

I was not impressed with the Vox piece about textbook prices. The 3-year cycle has nothing to do with competition and everything to do with destroying the used book market. And since modern computer typesetting was developed and spread in the form of open source software over 30 years ago, there has been no need to "type set" equations.

Those theoretical physics articles in the ArXiv? The typesetting of the equations was usually done by the author. It is easier than it looks. If you know what you want to say, you actually just write it like you write a sentence. For examples, pick a random pdf file or two on this page:

Besides, the equations in many intro math books have hardly changed in 50 years in some cases, and even the figures are often the same except for the colors.

That picture of equations in a physics book in that article? It comes from a book published in 1965 with the equations done on a typewriter.
What you are saying is that a contribution to an OER will be more valuable to your institution than a published book or journal paper. Why not request evidence of contribution to an appropriate OER when considering applications for TT roles and promotions? If this became widespread among CCs then the OER movement would really take off.
Given all the complaints there are about textbook prices, perhaps there is an opportunity here.

Suppose some enterprising soul bought the rights to textbooks that are currently out of print, and started printing then cheaply. Presumably these would be older texts that lost out to more popular (and maybe better) rivals.

If these cheapo texts were available at a hefty discount off what's charged for the currently popular textbooks, would colleges catering to cash-strapped students be interested?
@CCPhysicist - I agree that technology's made things much easier, but can you imagine teaching a first year course using a physics textbook that was copyedited to the same standard as the average arXiv paper? You'd spend more time sending out errata than teaching concepts.

While technology can mean that some of the mechanical parts of the process are hugely streamlined, were still stuck with the reality that putting together useful teaching materials is just incredibly labour-intensive.
Here's an example of someone who has taken the first steps in Johan Larson's proposal. A professor acquired the rights to his out-of-print text, updated it, and posted it as free PDFs online.
As far as I'm aware, he hasn't taken the step of making a cheap print version available. I actually use the book as a refresher in a graduate course. I'm not going to have my graduate students pay (at least not much) for a sophomore-level book to supplement the graduate-level materials, but it's good to be able to provide them with a free book to refresh themselves on the fundamentals.
@Johan Larson: You've just described Dover publications. Some of the best textbooks I have are Dover books, and I still don't see why so few people mention them whenever people complain about textbook prices.

I mean, we're talking about $10-30 for a textbook (new) instead of $100-300. And the Dover books depreciate far less, too.

The one problem with Dover is they have few texts that are more introductory--their mathematics section, for example, starts at Calculus and goes from there.

There's also some preexistent free books (see for example) floating around.

I think the greater problem is that while these books are there, there's no will to ask professors to use them. If, for example, a bunch of colleges had a policy that required professors to keep their book costs below ~$50 per class (perhaps making them have to ask for a waiver for pricier options, having to argue the pedagogical requirement for doing so) then perhaps we might see some real motion in that direction (and some complaining faculty, but what else is new?).
It's entirely possible that mere mortals can produce an open source College Algebra text.

An enrollment management system (especially one that it integrated with other systems, such as fin-aid and course management) is almost as sophisticated as an airline's reservation system.
I agree with Anon at 12:38 - make contributions to OERs count as much, or more, during tenure discussions as journal pubs and college professors will suddenly have time to put out decent products for such things. This would make a ton of sense at a teaching-focused school in terms of "what it matters to your students that you are good at", but probably would require a lot of people to get behind and push since how tenure is awarded is pretty entrenched.

Realistically, you also need someone who has some release time to curate and edit this thing for each course or course series on an ongoing basis, particularly for the first five to ten years. In math, at least, I'd expect to find a ton of errors the first year any page or chapter is used and probably an exponential decrease each year that nobody does a major re-write on that page but does go through and fix the errors that have been found so far. Once you had the content more or less stabilized and corrected the errors relentlessly for five or ten years, you're probably still going to need someone to babysit it somewhat, but that might even be a small enough task to count as someone's service rather than requiring a course release.

I base this on the month I spent writing daily "explain it like a book would, but more approachably and then here are some problems" worksheets from scratch for a high school geometry course while a long term sub at an alternative school. The school did not give students textbooks "because they would just lose them and they refuse to do homework anyway", and the worksheets I'd been left by their regular teacher for the rest of the year were all handwritten problem sheets with no notes on how to do the problems or what I should teach them before handing any given sheet out (they did tend to have answer keys, also handwritten, and some with an answer crossed out and another added in its place, which didn't reassure me as to the quality of the overall keys). I eventually decided it would be less work to just work backwards from the things they were supposed to be able to do by the end of the term according to the course standards and write my own worksheets that actually explained things and on which I knew which things were being taught which day. Anyway, I ended up with a goofy typo probably every other day, which was fine because the students knew I was writing these things myself and if I didn't catch them they would. I suspect if I'd had that as a regular job I would have done quite a bit of revising on several of them the next year, creating more typos in the process, but that it would have settled down by the third or fourth year, with most changes being typo-fixing. That's where I get the 5 or 10 years to stability figure from.
I've stopped using "mainstream" textbooks in my math courses, both due to costs and the constant appearance of new editions.

Fortunately, there are some great alternate options out there. For my calculus course, I could choose from

(a) a free text developed at a SLAC:

(b) a free, somewhat older, text made available by MIT OpenCourseware

(c) a low-cost ($35) option developed by the Math Association of America

Any of these are useful models for other disciplines...
I've been pushing for the adoption of OpenStax Biology for our intro for bio-majors course since last Spring. I wish I had a situation like Matt described where I was given the right to choose any textbook I wanted for my course. I'd go straight OER. And when students realized they had to spend $0 on my section and $175 on the other sections, their complaining and enrollment decisions would put pressure on my colleagues to follow.

Instead I have to convince my entire department, and my dean, that open textbooks are an acceptable alternative. In a best-case scenario open textbooks are one year away from improving my students' lives. In a worst case scenario I don't even know how far away they are.
Anonymous@5:48AM -

Yes, I can imagine how bad the average can be, because I have refereed papers as well, but my point was that assertions about the difficulty of "typesetting" math are based on ignorance. I hit a couple of papers at random on that page, thinking about posting one, and the first one I clicked had 148 pages and 551 non-trivial equations! (1408.5660) Someone wrote all of that in LaTex for the heck of it. Writing the equations in a calculus book is trivial by comparison.

It is a challenge to copy edit and review a book, but no different than the challenge of reviewing (let alone test driving) a new textbook. There are some pretty lame for-profit books out there, while there are some VERY nice OER books available, some being used at my college.

I haven't found any for the class I teach, but could some of us revise one of the algebra-based physics books to do an adequate job of introducing calculus-based physics? I think so. Heck, there is an entire freshman E+M book that is copyright free (vol 2 of the berkeley physics series) because NSF paid for it. Not a bad place to start. Better than waiting for copyrights to expire or hoping prices come down.
Are you talking about something like this?

Free, backed by a foundation, and taking advantage of a tablet platform to include models, animations, and video as well as pictures?
I'll also add that A. P. French's MIT books (also $20-30ish) are some of the best freshman/sophomore textbooks I've seen. I've kept the ones I got from my classes and got rid of the regular textbook we had (I don't even recall opening up my main freshman physics textbook much--I got enough just using course notes and the AP French books on mechanics and relativity).
I like the idea of using older books, as other commenters have suggested. I think this is very doable in most math classes, at least until you get to very advanced material -- certainly up through calculus. A new math book is not necessarily better than an old one (it is often worse) and I bet the same is true in other disciplines.

Could a similar approach work for ERP systems? Would it be possible to do without them, or at least to use simpler ones with less functionality? I'm sure much of what they do can be done manually -- that certainly holds for the ways I interact with them. How did previous generations handle things like scheduling and records? Maybe it's not necessary to always be innovating and changing everything -- sometimes change is for the better, but sometimes it is for the worse. I found teaching easier before I had to use Blackboard and the like.
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?