Monday, June 02, 2008

Open Source or Ad Supported Software and Content

The economic storm we’re now in is soon to become a deluge, according to Louis Uchitelle of the New York Times. The source of the problem is state and local government spending. To date in this business cycle, such spending has served as a counter force to the downturn. But tax receipts have lagged and the state governments are building up hefty deficits as a consequence. Surely spending cuts are just around the corner. For the usual Keynesian reasons, this bodes poorly for the economy overall, with the smart money waiting on the sidelines to see if we’ve past the bottomed out phase or not. But it is particularly troublesome for us who work at public institutions of higher education because it looks like serious budget cuts will be coming, if they haven’t hit already.

Within IT at like instiutions, I suspect things are particularly troublesome in this regard. Back in the early part of the decade when my university when through a previous downturn, administrative units were hit more heavily than were academic units and IT tends to be viewed more as as administrative function, even as most of the users are doing so in an academic context. There is the further point, made first by Nicholas Carr, that IT as it has matured has lost the ability to differentiate us as institutions, and hence the non-utility part of IT, if such a part exists, appears as if it can be dispensed with. And there is the still further issue that certain cost increases appear to be unavoidable – heating and cooling are notable in this category; everyone is aware of how the pump price of gasoline has skyrocketed and HVAC costs are obviously highly correlated with gasoline prices. So that is likely to make campus administrators more cautious about any other expense that seemingly is avoidable.

Budget cuts in IT is indeed a dreary subject, so instead of focusing on that let’s consider emerging technology and services, where by use of the term “emerging” I want to take the perspective of our respective institutions rather than the view of developers of such technology. In this sense emerging means there’s not yet institutional lock-in to the service and the future direction still seems remarkably flexible. Social networking in an academic context may be in that category. Likewise, so may video hosting, even at campuses that have supported video streaming for some time. They likely haven’t supported anything akin to YouTube and almost certainly have no mechanism to support student created videos made for an academic purpose.

Now let me take an aside and talk about my “roots” with instructional technology, formed mostly during 1995-96 when I was a grantee of the SCALE project here on campus. I had used the application PacerForum in spring 2005 and then switched to FirstClass thereafter. We referred to these applications as “off-the-shelf” software. First Class, in particular, was a poor man’s alternative to Lotus Notes and Microsoft Exchange and had a clientele both in business and higher education. Such software survives and evolves based on the needs and vicissitudes of the marketplace. Sometimes such software dies out. PacerForum, though quite admired by its users on Campus, didn’t or couldn’t make the switch to TCP-IP. The grantors from SCALE at the Alfred P. Sloan Foundation, former executives from IBM, had a distinct preference for us to use off-the-shelf software. While they never fully explained “the why” in this stipulation, the consequence was clear enough. It encouraged an approach of adopting pedagogy to the software and seeing what was possible that way, creating an emphasis on the teaching approach rather than on the software development, and demanding a degree of creativity on the part of the instructor to get a successful implementation.

Having also served as sponsor of some ed tech software development projects, wherein the developer’s primary goal, to do well by the users, seemingly trumped concerns about the business sensibility of the development, an admirable but undisciplined approach that oftentimes proved unsustainable after the grant money runs out, I’ve come to appreciate the approach with off-the-shelf software and still favor that approach ceteris paribus.

But, of course, things have changed. Most of the applications we might be interested with nowadays are Web based, accessed by a browser, rather than require a dedicated client as was the case 10 – 15 years ago. And many are “freely available” with a seeming explosion of new offerings in the past year. Many of these fledgling offerings hope to make it commercially. There seem to be two possible alternatives for that. One is to generate revenue directly, mostly likely via ad support, and see whether the operation can sustain in that manner. The other is to not worry so much about revenues up front but instead to focus on the experience of the users and grow based on their word of mouth, with the hope of viral growth and then to be bought out by one of the big players in the future and capitalize in that manner. Of course there can be hybrids of these two forms, such as where there is no sell out to one of the big players but where the ads emerge as a revenue aspect after there has been some substantial growth in usage. If you take those two extremes and allow for all permutations and combinations, that’s my current day analog to off-the-shelf software.

Likewise things have changed with campus software development. Home grown has been supplanted by open source, whether for LMS, ePortfolio, or any number of other applications. And to the extent that these efforts are meant as alternatives or tonics to the commercial provision of these applications, so much the better. But in the previous paragraph I was seemingly making an argument for commercial provision. So what gives?

Let’s consider the split between applications that the campus provides for users and applications that the users select for themselves. When I was getting started in the mid ‘90s, this divided nicely by the distinction between client-server applications on the one hand, and desktop applications on the other. There still is a difference between campus supported and user selected, but the notion of desktop applications is going by the wayside. Unfortunately, in my opinion, our thinking about campus provision of service hasn’t really accommodated that change. And the result is that we in IT take the scope of what should be provided by the campus as more broad than it otherwise might be.

Let me illustrate with a concrete example. I don’t want to make this an advocacy for the product, but I hope discussing it will help to bring out the issues. Consider the social networking application Ning, which is unabashedly ad supported. The feature of Ning that I want to take note of is that, like many other social networking applications, it employs the email address of the user as the login. Suppose that I, wearing my instructor hat, wanted to use Ning for my class and interact with them in that manner. Then I’d spend some time setting up the Social Network to my specifications and after that I’d invite my students to participate. I’d use the class roster that I can get from the grade book in the LMS to issue those invitations. The students would have to opt in and accept my invitation to have access to this class social network.

This is the first issue. Is my invitation coercive? Suppose the students don’t want to be part of that social network. Can they opt out by not accepting the invitation? Note there is no opt out for students in the campus LMS. They are in there by virtue of having enrolled in the course. So the opt out can’t be because they object to the pedagogic approach the software encourages and that I as instructor have embraced. Rather it would have to be because they don’t trust the commercial provider to protect their privacy, even when they can’t get into the social network without a password. The fear is that the vendor will data mine the individual information in such a way that might be detrimental to the student in the future. There may also be fear about copyright appropriation.

It is on these issues that my campus has either blocked or slowed down contracts with a variety of commercial vendors, who don’t seem willing to ensure the security of this information in writing. But I’m not the institution in my current position and in the above discussion I’m acting as an individual instructor. In that role I look to a different set of information to try to attain these type of assurances. Does the software work well? Are there other users more or less in the same boat as I who seem to be using the software successfully? Do I feel safe using the application? If I were a friend of the students rather than their instructor, would I feel ok inviting them to use it? In practice these questions are likely to be asked implicitly. When the answers are in the affirmative, meaning when I feel comfortable with the application, then I might very well embrace the application for my class.

The next issue turns from me the instructor back to the institution. What should the institutional stance be on this sort of instructor behavior? Possible alternatives include: go for it – active encouragement by the institution, don’t ask don’t tell – tacit acceptance of the practice by the institution but instructor beware because the liability might be on you, or you can’t do this – prohibition by the institution because until the institution has gotten the vendor’s promise in writing the students are being put at risk and the institution has in loco parentis. Which of these three we actually do might depend on to whom at the institution you pose the question. For software that has an admission charge (an Island in Second Life costs real dollars) the institution may be able to exercise control over campus dollars being spent on such software. For software that is freely available to the end user, however, there is no mechanism for institutional control up front that I am aware of. So de facto I believe we have don’t ask don’t tell, with the only mechanism for the institution to get involved after the fact being for a student in the class to lodge a complaint. But, if I had my druthers, we’d see an activist promotion by the campus, which would send an explicit message to all instructors to the effect of, “Go ahead with the use of commercial software that is externally hosted for your courses as long as you intend to use that software for legitimate academic purposes. We want to encourage you to innovate in your classes. An important way to maintain quality in the classroom is via your innovation and one way to innovate is for you to embrace emerging technologies. We understand there is some liability risk in doing that and we the campus will bear that risk for you as long as your intent is to promote the learning of your students.”

Now let’s take it another step. We’re in the world of don’t ask don’t tell where there is commercial provision of social network software that might be used for academic purposes. Absent the activist promotion described at the end of the previous paragraph, how does the availability of the commercial software impact the campus decision to host its own social networking software for academic purposes? Consider this question in light that many are arguing something like, “The students spend all their time on Facebook and if we want them to engage in their studies while online we have to provide something equally as compelling.” And with that let me return to how I led off this piece, talking about the economic downturn. How can the campus in good conscience embrace new applications of this sort when there isn’t sufficient budget to support what’s already a production service, when IT spending appears discretionary, and when other clear needs on campus are being given short shrift? Though consistency is the hobgoblin of small minds, if you agree with the sentiment expressed in this question, might you reconsider your thinking about instructors choosing to employ externally hosted social networking software for their classes?

* * * * *

Let’s switch gears and now let’s think of the instructor as content creator. In days of yore there was the “textbook model” under which faculty who wrote textbooks would have copyright of that work unless and until they sold that to a publisher, and such work would normally not count for promotion, tenure, or salary review except perhaps if it attained sufficient visibility to have some light pointed at the faculty member’s department and host institution. But textbooks are a dying breed and online content that has resale value is subject to a severe appropriability problem. Where do the incentives come from nowadays to get faculty to write good and interesting online content? Should those incentives be provided by the institution, perhaps in the form of some bonus payment for delivery of such content? Should faculty be marketing themselves much in the same manner as musicians who tour, viewing the online content as in part promotion for the live presentation, which ends up being the revenue generator? What about the alternative of faculty content displayed on Web sites that are ad supported, where some of the ad revenue bounces back to the faculty member, that in lieu of a royalty payment?

Many may feel uncomfortable with that suggestion and I’d be among them. There is an ethical issue regarding instructors making money off their students. I know on my campus there have been cases where the instructor is one of the co-authors of the textbook and in a position to require the textbook for the course and where colleagues of the instructor have viewed the behavior as less than exemplary if not downright uncollegial – it’s ok to make money off of students at other campuses if instructors there willingly embrace the textbook, but for our own students charging them for the content is immoral since so much of the content is an outgrowth of the teaching experience that was had on the campus and hence the instructor should feel a debt to the campus for providing such a productive environment. (A similar argument, attributed to Herb Simon, is made here for why on ethical grounds society should have a 90 percent flat tax, though Simon recognizes the incentives from such tax would be disastrous.)

Nowadays, however, many of the large gateway courses (the ones that would generate enough eyeballs so that ads might actually prove revenue generating) are taught by adjuncts who are on year to year contracts, rather than by faculty on the tenure track. And these folks tend to get paid less salary-wise. In the class system among instructional personnel, the adjuncts are definitely treated as second class. Further, because their teaching is often the main part of the job, the online content they make while teaching might quite possibly be considered “work for hire” and therefore the copyright may be held by the home institution rather than by the instructor himself.

My experience is that such instructors tend to be extremely cautious about their position – they don’t want to do anything to jeopardize their circumstance. Thus in spite of their focus on teaching, that makes them very slow to innovate with technology, especially if any of the innovation appears to be labor saving and hence possibly leading to their own (unplanned) obsolescence. The point is that the incentives are quite different for such adjuncts than they are for tenure track faculty. Might it be ok for adjuncts to produce content on ad supported Web sites? Might that give these instructors incentives to product good content or, at least, incentives to produce content that would attract eyeballs?

If the instructor does produce good content and profits from it as a consequence, what should the institutional stance be? It’s all well and good for the institution to create an open repository for such content for those instructors who decide to contribute there. But when the repository gets few contributions, can the institution look down its nose at commercial alternatives that might rebate some bucks back to the instructor?

The traditional argument against commercial interests sponsoring research is that it leads to bias in the outcomes and ultimately the parochial interest trumps the broader need, producing a perverse outcome. But a substantial aspect of argument is that the sponsorship remains hidden and so others are not able to factor that in when evaluating the results of the research. Ads, by their very nature, are overt. A different sort of argument would have to be made to argue against this incentive for the creation of instructional materials, namely that the students are not mature enough to evaluate the bias and hence that in an environment with ad supported content one is likely to witness a mild form of brainwashing rather than true instruction. Perhaps.

Yet I think the argument either needs to be more subtle or it is simply wrong. Many of us have come to trust Google as a search engine. It produces results quickly and efficiently and most of the time we find what it is that we were looking for. But, depending on what keywords we used in our search, Google might spit out a bunch of sponsored links – ads targeted by the words used in the search. We understand that’s Google’s way to make money. They’ve pioneered the approach. Do we think their search results are skewed as a consequence of the ads? Wouldn’t a widespread belief that the results were skewed kill the goose that laid the golden eggs? Likewise, wouldn’t an instructor who has a reputation as a good teacher do much harm to said reputation by deliberately biasing materials in order to generate additional ad revenue? What magnitude of revenue would it take for us to scratch our heads in answering that question?

Let me close with the following observation. We need to talk about this more, a lot more. We’re not ready to provide answers to these questions but the eager beavers among the faculty are ready to embrace commercial ad supported applications, purely because many of these applications are more functional and user friendly than other alternatives. So these faculty will act and then we’ll react, quite possibly inappropriately, meaning not in the best interest of our institutions.

The budget pressures might spur this sort of dialog. Sometimes being in a tough spot causes one to be more pragmatic in approach.

No comments: