Ideal CIO

Our CIO is leaving us for another job. My boss wanted to give my name to someone on the search committee seeking staff input for what we need in a CIO. This is just an organization of my thoughts and conversations I have had.

A couple recent conversations netted these:

  1. Big picture. There are over 100 people working on implementation. You should spend your time deciding what are the goals and priorities not meddling in the details of making it happen. Sure, reports on progress should percolate up, but quality people were hired to make it work, so let us.
  2. What do you bring? What experience do you have? Someone from another kind of environment may lend to a very different operating style than exists here. We operate in higher education but not academia. We work with universities and their users, but we are not ourselves academic. The “needs of many outweigh the needs of the one.” The CIO should fit into our culture and needs.

My own thoughts on the kind of leader I would like:

  1. Students first. We work in higher education, so you need to love students and want what is best for them. If the choice comes down to what is best for students vs what is best for faculty / staff, then the student choice should be the winner every time.
  2. Under promise and over deliver. Failing to meet a deadline even a “super tentative” one makes people unhappy. True, you will survive the first and n times, but eventually these failures will catch up to you. (True, we can sometimes save your ass by getting something out there and refining over time, but that early delivery sucks and the clients know it.) I learned under promise over deliver my sophomore year of college, so it surprises me how so many people at the top level can throw down an unrealistic arbitrary date before we even have looked at specifications or planned what is involved.
  3. Empower automation. My ideal CIO will have me replaced with a shell script. Well, lets me replace myself with them. My goal always is to replace everything monotonous I do with shell scripts and wherever possible schedule them. Why should I spend any time a week on tasks by hand a script can do for me? The first thing sacrificed in insane deadlines is the automation with the hopes of circling back around later to build it. Unfortunately, the success of meeting the deadline by sacrificing some things is more commitments which means no time to circle back around for automation. But, also distance from the initial development makes doing it later harder as what seemed trivial then gets forgotten and complicates the future revisiting. This is also where under promising helps as items like this stay on the plan so it gets done right the first time.
  4. Open architectures. Haters gotta hate. Salespeople gotta sell. That sometimes means glossing over the crappiness of their product(s). Really they cultivate relationships to feel needed. Just like I want to test drive a car before I buy it, I want to stand up the product and have end users clicking every button, administrators running every job, and engineers running it for some portion of the community in a pilot to ensure it works as expected. Pre-sale we learn how wonderful software is and how it will change our lives for the better. We learn just how terrible software is and no different from what we had prior running it for real the first six months.
  5. Need vs want. Our clients will say they want the buzzword of the day, but really they want a specific aspect or another thing entirely who might be somewhat related. Buzzwords create confusion because everyone intends something different by them. They know these are important concepts and use them to sound relevant. You need to tease out their vision and share your own so that when we can deliver yours, they get the thing they expect.
  6. Risk assessment. If you want to do something big, then we are going to tell you what are the risks. It may sound like “No,” but really we are laying out the groundwork for what needs to be overcome for success. Think of it as part of the planning. Pay attention to what is being said as you might learn what seemed easy enough will take far more effort than originally assumed. Use it to set expectations for #2.
  7. Introverts. This is information technology. Some people who work here like computers better than people. (Easier to understand. Consistent.) Extroverted activities are fine, but do not force us to attend. An extrovert is more like to be CIO, but they need to understand we are not all that way.
  8. Real costs matter. Good services improve efficiencies such that people can do more than the past. Bad services bog people down such that they spend more time to do less. One-size fits all products tend to fit into the bad services category. They try to cater to every users’ needs such that they have too many buttons with wrong terminology. Too much time has to be devoted to training just so people can feel comfortable using it. Too much time has to be devoted to support issues so people can figure out how to do basic mandatory tasks. Because really they only ask for help on those issues and curse us for all those tasks they give up on trying to accomplish because it was too hard but not mandatory so ultimately not worth the effort.

The search committee should thoroughly investigate the hire before announcing it. From past experience they will interview, select 1-4 finalists they bring for us to meet, and offer to one. A piece of advice: Any finalist and definitely whomever gets the offer should be vetted beyond just the resume, references, and background check. Do Google searches on them and their last few employers. Investigate further anything publicly negative about them. Even better, talk to people there who are not listed on the references because we the ground level will find out any dirty laundry within days of the announcement. If they are leaving because of a scandal, then you need to know before we do.

(Technically I wrote a draft of this a month ago, but I did not schedule it to publish until later because of this part. A few more items were added since. And decided to publish password protected.) Less than six months into the now outgoing CIO’s reign, I was very unhappy. I was thinking long and hard about how bad of a fit this was to me. The grumbling came from everywhere around me. The last time the people around me were this discontent, I jumped ship and landed here. So I was preparing to jump ship again. In the end, my enormous respect and loyalty to my boss (who had no idea) saved me from going. The difference between last time and this was the great boss.

AMLS Rec Engine

Now anyone can tap into the Amazon Machine Learning Service (AMLS)?

I’d love to see Ellucian’s Banner product make smart recommendations based on AMLS. Students register for university classes through Banner. I could see Banner leveraging AMLS to figure out that students who like certain classes might be interested in other classes. There already is a market trying to improve retention by detecting students who have trouble on certain classes or tests or questions get flagged for extra attention. I could see AMLS helping the other spectrum. Students who take a certain class might be interested in certain cooperative education opportunities, clubs, or campus events. Or vice versa, such as students who join the environmental protection club might be interested in an elective class about the impact of biological pest controls vs pesticides.

Or maybe AMLS going beyond class recommendations is more appropriate for Ellucian’s Luminis portal product.

Higher education is full of opportunities for connect students to things about which they are oblivious. Flyers on various walls easily miss students. Enormous daily emails go straight to junk mailboxes. Students (and staff and faculty) deserve smarter ways to connect to the things that will make their experience better. Done right, I could see AMLS filling that need.

The Loss of Tech Support

I found a statement in Twitter is your IT support interesting:

For reasons I won’t go in to, I haven’t been able to get [a WordPress install with the FeedWordPress plugin] done at the Open University, despite trying since last July. I’ve spoken to people at others unis and it isn’t isolated to the OU, it seems to be this low-level, experimental type of IT support is increasingly difficult to find.

Do you know who I think the culprit is? The VLE. As universities installed VLEs they became experts at developing enterprise level solutions. This is serious business and I have a lot of respect for people who do it. The level of support, planning and maintenance required for such systems is considerable. So we developed a whole host of processes to make sure it worked well. But along the way we lost the ability to support small scale IT requests that don’t require an enterprise level solution. In short, we know how to spend £500,000 but not how to spend £500.

(For those of you non-British/European readers, VLE are Virtual Learning Environments which are often also called Learning Management Systems on this side of the Atlantic.)

It is true the higher education IT has change with online class systems, but I think that part of the symptom and not causal. Chief Information Officers, Chief Academic Officers, and presidents all get recognition for big things. Enterprise level solutions are sexy because it is something that makes them look decisive and effective. Employees who report to them know this, so enterprise level solutions have the priority. Everything else fits into the dwindling extra work time.

What extra time?

The good news though is the small things have gotten much easier for anyone to go off on their own. At my last job, I sat as an ex-officio member of the Faculty Senate technology committee. One of the hot topics one year was a couple faculty members taught students how to use the LMS adopted by another college system in the state. It was two courses. Should we spend $20,000/yr and take up a significant amount of my time running a second LMS? Or should they continue to pay $800/yr for Blackboard to do it? The answer ultimately was to continue with Blackboard. Now days, they probably would be directed at CourseSites. At the time my to-do list was several pages long and hundred plus hour weeks were not uncommon just to keep top and high priority items timely done. The ETA for anything not top or high priority was over a year.

I prefer working with innovative technologies. Custom solutions that require creative thinking and problem solving make me feel like I accomplished something special. They give the biggest rush. Enterprise level software is steak and potatoes, so it is the core. The enterprise is the minimum. I just wish I more time to devote to achieve going beyond the minimum than I did. Well, do. This is a top level decision. Improve staffing and flexible team management so that people can spend time working on the things that make them happier.

Global Higher Education Trends

According to Trends in Global Higher Education (PDF), we should pay attention to globalization, massification,

Globalization is an interesting trend. As a college student, I enjoyed hanging out with international students and as an employer of student workers, half were international students. Exposure to different cultures, meaning values and perspectives and rituals and (the best) food was a great experience for me. It is harder to hate another culture when one has real friends among them. Such ties often become the basis of international diplomacy. But those students also mostly went home and are doing great things as part of the growing middle class.

Employers looking at post-secondary degrees as signals for middle class jobs drives massification. If this signal were terrible, then perhaps employers would seek an alternative. But I don’t think it means what most expect. The expectation is it means highly educated within the major. Instead, I see the bachelor’s degree as a demonstration of successfully navigating the world’s worst bureaucratic disasters. Having the tenacity, patience, and soft skills to deal with process failures all over the place. Secondarily, the degree means the ability to demonstrate some learning on demand to pass an evaluation.

Turning to look at how we here in Georgia compare to rest of the world, the crises facing least developed countries are constraints on research university budgets, constraints on student financial aid, increases in tuition, more part-time faculty, larger class sizes, a freeze on books & journals, construction, etc. This may not be solely a problem for least developed countries. Most of these are happening here in Georgia.

  • State funding was stagnant before the recession and been dropping since. Per student funding state funding has plummeted from about half to a quarter. The legislature and the governor have to make hard choices about what to fund. Higher education does not rank high enough compared to keeping people safe and healthy. Is there a Maslow’s Hierarchy of Needs for government funding?
  • A big source of financial aid here, HOPE, used to pay for all tuition for students who maintain a B GPA. It is lottery funded, but revenues were not able to keep up with the 10%+ annual growth of students. So now the awards are reduced for all but the top most students and may continue to drop.
  • My librarian friends lament about their severely reduced budgets for purchasing journals. Combine this with skyrocketing costs for these same journals and maybe by 2030 the research universities should just sell their collections and close the libraries?
  • The one positive is construction has not stopped. Though buildings are not built fast enough. (Some schools schedule class days to happen on the online class system I help run because they lack the classroom space.)

Even when the Georgia economy fully recovers, the lost ground is unlikely to be regained. But there is also increasing pressure to improve graduation rates and the number of graduates. Interesting problems we get to solve.

One of Many

The Learning Management System (LMS) has been a despised technology by some ever since I started working with one, WebCT, in 1999. At the time it was deemed crappy technology that had to improve or die. So today in 2012, about 13 years later, I have to roll my eyes at the pundits writing about how the current technology has not significantly changed in a decade (really more than a decade) because it still offers the same tools and will die unless it adapts.

My first few years, 2006-2010, of working at GeorgiaVIEW, our active user counts doubled every 1.5 years. We plateaued at around 290,000 and grow a few thousand a year. Numbers of actions in the system still doubles every 1.5 year. That is insane growth. Growth unlikely fueled by people despising use of the tool. Right now, we are getting pressure to migrate Summer 2012 content for the Fall 2012 start in Desire2Learn1 because instructors roll over the classes from term-to-term. That speaks of long term consistent loyal use not occasional only as little as have to use. For something on the verge of death, it is hard enough keeping the users happy.

I am a database administrator not a faculty member (or dean or vice president for academic affairs or provost). It seems to me though no one would say, “When you teach a class, the white board in the room is the only tool you can use.” Instead, the push would be to add to the available tools in a neverending pursuit of finding better ones. So we see pressures to integrate the LMS with a variety of similar specialized services. Many are textbook replacements or supplementary services designed specifically for student needs. Others are social media. More and more the LMS is just a portal: a place to organize where students really go to learn.

Also, as an IT guy, I think it is important to have a plan B. Things sometimes fail. As a student I was always annoyed when the instructor had to leave the room for 20% of the class to go track down a piece of chalk because the remaining ones were too small to write. I applauded once in my junior year because the instructor happened to have a piece of chalk in her purse just for that contingency. Similarly, faculty members and even students should think about what to do when the LMS is not there. Heck, what should they do if everything the university IT runs like the web sites, email, portal, and network all disappear. It can happen.

When the university bureaucracy selects and administrates a tool, they will adhere to university policy which adheres to higher education laws. When a faculty member selects and administrates a tool, they should do the same. Unfortunately, that means the faculty member becoming familiar with policy and law. Another challenge is running into different interpretations. An example: a user following @VSUENGL1101 on Twitter could be reasonably expected to be a student at Valdosta State University enrolled in the subject English class 1101. Some say that violates the Family Education Rights and Privacy Act. Some disagree, so it is being debated. The law is old and did not likely anticipate social media, so naturally there is movement towards an update.

I doubt the LMS will simply die because there is something better. Instead it will remain one of many tools for years to come. Like the land line, television, JavaScript, still camera, WiFi, non-smartphone, and (God forbid) pagers.

Note 1: Desire2Learn objects to their product being called an LMS. They prefer Learning Environment on the grounds it integrates with so many other tools.

P.S. This totally is from a sustaining technology perspective. Guess I should write this from a disruptive technology perspective.

Black Box Magic

black boxes ttv

With a black box system a person working with it sees what goes in and what comes out. The machine’s decision making process is obfuscated. Theories are made based on incomplete evidence on the behavior. More data points on more situations confirming the behavior is my way of being more comfortable the theory is correct. Sometimes we lack the time or conscientiousness or even access to ensure the theory is correct. This leads to magical thinking like labeling the software in human-like terms, especially insane or stupid or seeking revenge.

With a white box system, a person working with it can see the machine’s logic used to make decisions. Theories can be made based on more complete evidence due to investigating the code to see what it is intended to do. The evidence is far more direct than testing more.

Systems today are so complex they tend to have many parts interacting with each other. Some will be of each type.

Then there are Application Programming Interfaces (APIs) which expose vendor supported methods to interact with a black box by disclosing how they works.

Proprietary systems tend towards a black box model from the perspective of clients. This black box philosophy depends on the experts, employees of the company, design the system so it works well and resolve the issues with it. So there is no need for clients to know what it is doing. Where the idea breaks down is clients who run the systems need to understand how it works to solve problems themselves. Sure the company helps. However, the client will want to achieve expertise to manage minor and moderate issues as much as possible. They want to involve the vendor as little as reasonably possible. Communities arise because peers have solved the client issues and getting an answer out of the vendor is either formulaic, inaccurate company line, or suspect. Peers become the best way to get answers.

Open source systems tend toward a white box model from the perspective of clients. This white box philosophy depends on clients to take initiative figuring out issues and solutions to resolve them. Clients become the experts who design the system so it works well. Where the idea breaks down is some clients just want something that works and not to have to solve the problems themselves. Sure the open source community helps. Companies have arisen to take the role of the vendor for proprietary systems to give CIOs “someone to yell at about the product”. Someone else is better to blame than myself.

Cases of both the black and the white box will be present in either model. That is actually okay. Anyone can manage both. Really it is about personal preference.

I prefer open source. But that is only because I love to research how things work, engage experts, and the feel of dopamine when I get close to solving an issue. My personality is geared towards it. My career is based around running web services in higher education. Running something is going to be my preference. (Bosses should take note that when I say not to run something, this means it is so bad I would risk being obsolete than run it.)

This post came about by discussing how to help our analysts better understand how to work with our systems. It is hard to figure out how to fix something when you cannot look at the problem, the data about the problem, or do anything to fix it. So a thought was to give our analysts more access to test systems so they get these experiences solving problems.

Photo credit: black boxes ttv from Adam Graham at Flickr.

D2L Drops Utah Lawsuit

Higher education is very litigious. Students sue universities over encroaching on personal beliefs. People who fail to get accepted sue over discrimination or reverse discrimination. Employees claim wrongful termination. Publishers claim copyright infringement. Software companies battling over patents. Equal funding of historically black colleges and universities to non-black counterparts.

We are involved in selecting a new LMS. The Utah Education Network is one of the few Blackboard clients very comparable to us. I track their performance issues because I expect when we reach the same size we could have the same problems. Their selection of Instructure Canvas caused quite a buzz. D2L going for the buzzkill with a lawsuit perhaps stirred up even more buzz. Unfortunately, it seemed all against D2L.

D2L milked being the victim in the Blackboard patent lawsuit. Now they want to be the victim in suing a tiny company over winning the UEN LMS contract. Some speculated this was all so D2L can get a hold of the documentation explaining why they lost to a company they consider inferior. (The letter explaining why they could not possibly be the real reason.)

Dropping the lawsuit was a good move for D2L, I think. What they would lose in the “court of public opinion” certainly outweighs anything they could gain through a lawsuit.

Anyhow, a good of this is others involved in a selection process hopefully will make sure to be more transparent in order to avoid a similar situation.

Apples to Oranges

My web hosting service, Dreamhost, happens to have a one-click-installer for Moodle. So I installed one for my own personal sandbox. In looking at the available roles, it suddenly occurred to me…. Comparing any LMS to another is like comparing an apple to an orange. The industry is like the Tower of Babel. Each product has its own jargon covering much of the same ground in absurdly different ways. How could you have an Internet if all the servers talked to each other so differently? Yet in technology created for higher education every system has a different name for the person who teaches a class or even if the name is the same, the capabilities differ.

Sure, there are some commonalities, even apples and oranges are both fruit, but the developers had different conceptual models in mind. The same word meaning different things really is quite annoying. Another example is a course in Vista is a container for the type of place where learning takes place whereas in Learn a course is where teaching takes place. Teaching in Vista takes place in sections. So… Vista : section :: Learn : course.

These different conceptual models are why the faculty get so irate about change. It is hard enough to have to learn new places to click and how to accomplish what you used to accomplish. For some period of time they have to have two vocabularies and maybe even years later they still cannot call it the correct term. (WebCT CE/SE called where teaching takes place courses and both former and current coworkers 6 years later still call sections “courses”).

One would think standards organizations like the IMS Global Learning Consortium would help solve this. Every product adhering to a standard should end up adopting consistent terminology, conceiving of objects similarly, and conceiving of processes similarly. This make comparing the two easier. Except the standard adoptions appear to be in the integration components or database not the main product.

I really feel bad for the instructional technologist who has to support more than one learning management system.

Also, selecting a new LMS seems like an insanely difficult task when trying to learn a dozen vocabularies enough to ascertain whether it has what you need.

Anya Kamenetz: DIY U: The Coming Transformation of Higher Education

The solution to the upcoming fall of higher education is openness of content and sell the services beyond the content. Lower the cost by sharing. Will be interesting to see whether higher education would actually go in this direction.

Instead what I have seen is colleges trying to figure out how much of a premium to charge completely online classes over the normal tuition. The boundaries were in-state and out-of-state tuition. Charging less than in-state tuition was seen as crazy because the assumption was no in-state students would take a class online when they lived close enough to show up in class.

Please Don’t Write Off the LMS Just Yet

Found the Educational Technology Trends 2010 quite interesting. Especially the part which predicts yet again (still?) the death of the LMS.

Both learning and learning content are moving away from traditional centripetal models, in which everything happens at set locations and is controlled at the institutional/publisher level (top-down), and moving toward centrifugal models that are learner-focused (bottom-up) and in which learning happens wherever a student happens to be. This means new platform models for learning (post-LMS), greater mobile access, more flexible e-commerce models, and a renewed explosion in generic online learning.

I suspect the relationship between the LMS and what is next is more like  LMS : post-LMS/PLE/DIY U :: book : Internet. The Internet has not as of yet killed off the book. We still have plenty of books available. Even books are shifting towards digital. The Internet just moved into greater prominence and changed how we think. Similarly, new platforms may result in something which we will think of as how students learn in higher education, but the LMS will still be around for a very long while. (Which is good because it means I still have a job for a while.)