Openness: Finally, We're Getting Somewhere

Update: The Sannier vs. Wheeler smackdown is available on video here. Microsoft Silverlight is required.

As I sit on the flight home from the EDUCAUSE conference thinking back on the themes of the week, it is clear to me that various flavors of openness have finally arrived in a big way in higher education, with more on the way. Certainly, a critical mass of universities are recognizing the value of open standards and accepting that open source is a viable and important approach. Open access and open educational resources are coming on too, although they are much earlier in their adoption cycle. Further down the road, open governance and open business practices are looming on the horizon.

One of my big tasks of the week in my role as an Oracle employee was to promote the adoption and value of the IMS Learning Information Services standard, which I have blogged about several times before. (This is a probably an appropriate point to remind everyone that the opinions I express on my blog are solely my own and do not necessarily reflect those of my employer.) “Promotion” consists of both recruiting potential adopters in the form of software vendors and open source projects and also evangelizing university stakeholders about why they should demand better support for the standard from the software that they adopt. This is often a bit of a hard sell. For software developers in general, supporting a standard often requires diversion of resources from the development of sexier features, and it almost always requires making a compromise to do things a little differently than you would do them if you were developing the functionality in isolation. For software vendors in particular, support for a good, working interoperability standard often means a reduction in the services revenues that customers pay you to make the darned things work together. And for customers, the standards pitch is often a bit of a bank shot. They don’t care about standards support in and of itself. They do care about the benefits that standards can provide, but you have to make the connection for them. So all around, it can hard to get traction for a new standard, particularly if it is one that entails any significant complexity.

I was pleasantly surprised at how well our message was received this week. To begin with, the movement toward adoption by projects and vendors has been terrific. Every LMS product or project that has significant market share in North America is now moving toward compliance with the standard. Some are going to get there faster than others, but every one has promised they will adopt, the majority have begun development, and a couple have started or even finished testing. Facebook application provider Inigral has also provided a great boost for adoption by being an early and high-profile non-LMS adopter of the standard. They have demonstrated that the value of the LIS interoperability standard goes well beyond provisioning into Learning Management Systems. On the producer side, Sungard made some very strong public statements about their intention to adopt the standard sooner rather than later. These statements have been backed up by regular attendance by both functional and technical representatives in the IMS working group. With both Sungard and Oracle on board, application developers who might consider developing software that consumes LIS data can be assured that something approaching 80% of the North American higher education market will have systems capable of providing that data. On the university side, my colleagues and I spoke with quite a few university folks who really understood the value and importance of their standards and were ready to hold their vendors’ feet to the fire in order to ensure that the promised support for the standard really happens. Some of the progress here is because LIS hits an obvious pain point for them, but part of it is also due to increased general awareness that standards matter. As campus IT gets more deeply involved with integrating a wide range of systems, they simply cannot afford to not have interoperability standards at this point. It’s too darned expensive without them.

On open source, one of the early highlights of the conference was a no holds barred smackdown debate between Brad Wheeler of Indiana University and Adrian Sannier of Arizona State University. Both of these guys have achieved notoriety for their respective actions as CIOs. Brad was one of the Sakai founders and now spearheads Kuali in general and Kuali Financial System in particular. Adrian went another way at ASU, with a highly publicized rapid deployment of Peoplesoft (outsourced and externally hosted) in which he pushed the university to change their back office business practices to follow the way the software works rather than customizing the software. He also was an early and high-profile adopter of Google Apps university-wide. Both of these guys are very smart, very articulate and very good verbal pugilists. The debate was spirited, to say the least. I’m talking Godzilla vs. King Kong here. I’ll link to the audio when EDUCAUSE posts it.

One of the most fascinating aspects of the debate was the ground that it didn’t cover. Within the first two minutes, the two men agreed that open source is obviously practical and important for universities. Everybody is running Linux and Apache in their data centers these days. They even agreed that it makes sense to consider developing open source tools for functions that are close to the core university mission, such as LMSs for teaching and learning or grant management systems for research. Where they differed was on where to draw the line.

Adrian argued that, unless universities have invented a way to compensate their faculty without currency, their fundamental financial practices are really not much different than those in any other organization that pays people, and that it makes no sense to develop something like a university-specific financial system when you could instead ride “the infinite growth curve” of the rest of the world. Brad replied that (a) in fact, university payment processes are quite different from those of other organizations and therefore need special software support, and (b) the small number vendors in the higher education ERP market have practical monopolies by market sector, and this was bad news for customers who want leverage to extract more value from their vendors. And so it went. This was a very sophisticated debate, and it was really a refined articulation of the same conversation I heard elsewhere throughout EDUCAUSE. Suddenly, nobody is asking whether open source is safe anymore. That debate is (finally, blessedly) over. The question people are asking is where does it make the most sense for their institutions.

Reflective of this new atmosphere is the way that the various private source vendors are suddenly trying various strategies to attach themselves to the “openness” label. For example, Jenzabar just announced that they are “opening up” their ERP product through what they call a “collaborative source” initiative. But Blackboard is probably the most high-profile of the sudden promoters of “openness.” They announced integration with Moodle at this conference, having previously announced integration with Sakai. (For my analysis of the business strategy behind the move, see this earlier post.) They have been putting a lot of energy into the IMS Learning Tools Interoperability (LTI) standard, significant energy into Common Cartridge, and there were signals this week that they are ready to step up their efforts to support LIS. (In fairness, I should note that ANGEL Desire2Learn, and Sakai have also been very heavy participatants in various standards efforts for a long time now, albeit with somewhat less fanfare.) John Fontaine was all over the conference wherever an openness conversation was going on, working very hard to make sure that Blackboard was participating in those conversations.

But the most interesting move by Blackboard at the conference was to host their own panel discussion on openness. Moderated by Inside Higher Education’s Scott Jaschik, the panel included Blackboard CEO Michael Chasen, open source and open standards evangelist Chuck Severence, and the CIOs from Princeton and Syracuse. (My apologies to these two gentlemen; I don’t seem to have their names. I’ll update with this information when I get it.)

Update: The panelists were Michael Morrison from Syracuse and Serge Goldstein from Princeton. I should have remembered at least Michael’s name, but I have a horrible memory for both names and faces.

To its credit, Blackboard imposed no boundaries or limits on their panelists or moderator. That said, the panel selection was…let’s say relativelyfavorable to Blackboard. Both CIOs were honest, outspoken, and not afraid to voice their criticisms of the company. But fundamentally, they were both reasonably satisfied Blackboard customers. And Chuck, while he is his own man and always speaks his mind, has a consistent track record of courting Blackboard toward more open behavior. There’s nothing wrong with any of that, but it did limit the spectrum of challenges that Blackboard was likely to receive from the panel. Another limiting factor (and this doesn’t appear to be Blackboard’s fault) was that the presentation was not clearly labeled in the conference program. Most of the usual open source advocates were missing from the audience and therefore the conversation.

Naturally, the issue of trust came up early. Chuck brought up the fact that Blackboard course cartridges currently use a modified version of the IMS standard. This is understandable in and of itself, since the old versions of these standards were not really adequate to meet the needs of current-generation learning systems. (Common Cartridge is supposed to fix this problem.) The real problem is that, in principle, Blackboard would have the right under DMCA to prevent people from modifying those cartridges to work in other systems. Chuck said that Blackboard has made private assurances that they would not do this, but called on them to make a public declaration to that effect. Michael Chasen responded that it would be impossible for Blackboard to make legal statements on every single request of this type and that there needs to be some level of trust. Leaving aside the question of whether Blackboard should make a point of making a public legal statement on this particular question, Chasen has a point on the larger issue. There has to be some measure of trust

In which case, the patent suit looms large. Large portions of the educational community view the filing and assertion of the patent as a grave breach of trust. In order to heal that breach so that the trust necessary for “openness” can grow, Blackboard needs to do one of three things. First, they could choose not to assert the patent anymore. There is no indication that they are willing to do so. Failing that, they need to convince people that they are right in pursuing the patent litigation. After two years of making their case, anyone who was persuadable has now been persuaded. For the rest of the world, if Blackboard is not willing to stop asserting the patent, then they at least have to convince people who they want to trust them that the suit is somehow a special case, and that there are clear and bright lines that Blackboard would never cross. I can’t evaluate how effective Michael Chasen’s argument was in this regard because, as far as I can tell, he didn’t make one. He simply asserted that Blackboard would never do anything to harm education. For people who are convinced that they have already done something to harm education, his assertion without a compelling rationale and in the face of what they consider to be evidence to the contrary is not going to be persuasive. Until they are able to articulate some bright lines that skeptics would find convincing, I fear that John Fontaine will continue to have a tough reception in these openness meetings, despite his best efforts.

The week wrapped up (for me, anyway) with the Openness constituent group meeting. Patrick Masson did a terrific job of facilitating. He had three tables set up–one for “open technologies”, one for “open education”, and one for “open governance/business practices”. Participants self-selected their seating at these tables based on interest. He got each group to articulate a definition of their area, benefits, enablers, and barriers. Then we made comparisons across groups. As Patrick had intended, it became clear that (a) there were some critical common threads of philosophy and approach, but (b) each group is at a different level of evolution, with open technologies being a fairly well-understood and nuanced area of discussion, open governance being fuzzily defined and not widely known, and open education being somewhere in the middle. Going forward, the group is going to continue to try to surface commonalities and help higher education figure out how to approach openness in more systematic and fruitful ways.

All in all, I believe it was a watershed conference. I predict that next year we will see more programatic exploration of how and where to adopt various flavors of openness. On the software side of things, this conversation will be facilitated by the growing interest in service-oriented architecture, which is pretty much at the same place in the conversation this year that open source was three or four years ago. Both REST-style and SOAP-based flavors of SOA will definitely impact the practical range of choices avaiable to universities regarding mixing and matching integrated components available under all license types.

The best is yet to come.