A Potpourri of Games

In 1999, my Ph.D. student Andreas Junghanns and his wife Manuela hosted a party at their apartment in Edmonton. They invited a few faculty and graduate students over, all members of the GAMES (Games, Analytical Methods, and Empirical Studies) research group. We enjoyed an evening of playing a variety of games, eating, drinking, and socializing. Needless to say, a good time was had by all.

Andreas and Manuela had a nice idea, so it had to be copied. Several months later I hosted a GAMES party. Soon, other faculty members in the GAMES group joined in. For almost 15 years now we have had several GAMES parties a year, trying to average one per month (excluding the July and August holidays). Any excuse is a good reason for having a party: a visiting GAMES researcher, meeting new GAMES students, celebrating a success, or just because we need a break. Whatever the reason, GAMES parties are popular amongst the faculty, staff, and students who are interested in games-related research (and some who just enjoy playing games).

One of many tables at a typical GAMES party.

The parties include any form of game. During any given party, you may encounter classic games (such as chess), newly published games (often from Germany), physical games (such as billiards), and video games (no shortage of choices here). The play is non-competitive; everyone participates for the fun of it. Thinking and strategy games dominate the selection; the higher the luck factor the less likely it will be played. A typical party has between 20 and 40 attendees, dominated by the graduate students. There is always plenty to eat, drink, and think about.

Michael and Shayna Bowling: May 2014 GAMES party hosts.

Michael and Shayna Bowling hosted the GAMES party held in May. During the week, Michael is a passionate, highly successful artificial intelligence researcher and the leader of our Computer Poker Research Group (CPRG). But during their leisure hours, Michael and Shayna are enthusiastic game players and game collectors. They have a large collection of games and bring their latest additions to every GAMES party.

A sample of the Bowling's game collection.

And what of Andreas and Manuela? They are both back in Germany and doing very well. When visitors from Alberta drop by, they host “small” GAMES parties. GAMES group alumni occasionally host parties around the world, including Iceland, Denver, and Silicon Valley. GAMES parties have now gone international!

What's a party without food?

MH370 – The Largest Maritime Search?

On March 8, 2014 Malaysian Airlines flight MH370 took off on a routine trip from Kuala Lumpur north to Beijing. Shortly thereafter, the plane disappeared from radar screens and, other than a few satellite pings, no trace of it has been found. Two hundred and thirty nine people are missing and presumed dead. There are several clues as to its whereabouts that surprisingly led south, to one of the most inaccessible regions of the Indian Ocean. Even a massive international effort involving dozens of boats, airplanes, and submarines has failed to unravel the mystery of MH370. Commentators have called this the “largest maritime search in history.” But is it?

Readers of this blog know of my interest in polar exploration. So it should be no surprise that I find a connection between MH370 and the biggest polar search in history.

On May 19, 1845 the ships Erebus and Terror departed England under the command of Sir John Franklin. Franklin and his 129 crewmembers set forth on what was expected to be a triumphant journey through the North-west Passage. Discovering a navigable route through the Canadian labyrinth of islands and ice-clogged passageways was one of the great geographical mysteries yet to be resolved. On July 25, 1845, just west of Greenland, a whaling ship pulled along side Franklin’s ships. After that, not a word. At the time, that was not a problem. After all, this expedition was expected to be away for a year or more.

As Andrew Cohn writes in Lost Beneath the Ice (2013):

But when nothing was heard from Franklin by late 1847, the [British] Admiralty worried. … Tens of thousands of worshippers filled churches across England, and the efforts to find Franklin became a cause célèbre. A massive international effort was launched... It ‘was the greatest activity the Arctic would ever witness’… Over the next dozen years, some forty ships and 2,000 officers and men would join the search for Franklin, making this ‘the longest and most expensive search and rescue operation ever undertaken.’ Between 1848 and 1853, some twenty-eight expeditions on sea and land were sent to the Arctic.

The search area was literally the entire Canadian Arctic, at the time regarded as one of the most inhospitable places on Earth. Expeditions probed from the Eastern Arctic, trying to retrace Franklin’s path. Others entered from the Western Arctic, expecting him to have completed or come close to completing the North-west Passage. Yet others explored the northern reaches, thinking that he might have gone into the mythical Open Polar Sea that supposedly existed around the North Pole. The British public was hungry for news. But each expedition took many months to return home, and most took several years. There were no telephones or Internet to keep the public informed. The drama in the Arctic was played out in painfully slow motion. 

In 1850, the first trace of Franklin was found. On Beechey Island, close the geographic center of the Canadian Arctic, the Franklin’s 1845-46 winter camp was found. But no note was discovered – nothing to indicate where he went once the summer came, the ice melted, and his ships were free to continue their journey.

It wasn’t until 1854 that evidence emerged as to the fate of Franklin and his men. John Rae, on a routine mapping mission, encountered Eskimos with stories of white men who died of starvation many years before. Rae returned to England with the stories and artifacts from the lost crew. The public had to wait until 1859 to get most of the story. Francis Leopold McClintock led an expedition that was able to reach the site of the tragedy (King William Island) and recover the only surviving document of the expedition. The single page of paper told a sad tale of being hopelessly trapped in the ice and a starving crew forced to abandon ship in a desperate attempt to march south. None survived.

It took 14 years to discover what happened to Franklin’s expedition. Over the course of that time, most of the blank areas on the map of the Canadian Arctic were filled in. A navigable North-west Passage turned out to be illusionary, at least for ships in the nineteenth century.

The Franklin mystery was eventually solved, as will that of MH 370 eventually. The search for MH 370 has been ongoing for only two months. With today’s sophisticated equipment, we can only hope that the search won’t take as long the Franklin search.

Coursera By the Numbers

In a scant 2.5 years, the landscape for teaching and learning at postsecondary institutions has changed. It began in September 2011with a Standford University course in Artificial Intelligence (Sebastian Thrun and Peter Norvig) -- the first “massive” MOOC had 160,000 registrants and 20,000 course completions. There is a raging debate as to the academic value of MOOCs. However, there is no disagreement that MOOCS have shone the spotlight on teaching and learning in a public way and to an extent that has never happened before. And that is a good thing.

A major research university, like the University of Alberta, has the mandate to do research and teaching. At most such institutions, the former receives all the attention (and all the funding) while the latter is low profile (and often underfunded). Given all the recent media attention given to MOOCs, teaching and learning, many institutions have begun to invest heavily in transforming and even moving away from the traditional classroom experience. “Blended”, “flipped” and “experiential” are some of the popular buzzwords that describe (not-so-) new teaching techniques starting to be widely deployed.

Coursera was created almost two years ago by Stanford University professors Daphne Koller and Andrew Ng. Most people associate the company with providing the computing platform for offering MOOCs. However, many instructors are using the vast amount of content that is being put up on the Coursera site as a resource for the blended/flipped delivery of their local courses.

I am writing these words at the second annual Coursera conference, in London, United Kingdom. The venue is the University of London in a building that is literally beside the British Museum. When there’s a 15 minute break, what better way to pass the time than by walking next door and browsing (and it’s free!). How cool is that!

Daphne Koller, London 2014

Daphne and Andrew gave a tag-team presentation on the current state of Coursera and some of the company’s future directions. Some of the takeaway numbers include:

·      7 million registered users,

·      33% of users are from the United States,

·      33% of users are from emerging economies,

·      25 million course enrolments,

·      7,466 years of video watched,

·      64 million quizzes taken,

·      2.5 million peer-graded assessments, and

·      1 million course completions.

A big discussion point is the completion rate of MOOCs. The first Stanford MOOC turned out to be an exception: 20,000/160,000 = 12.5% of registered students completed the course. The latest data from a variety of sources, including multiple MOOC providers, shows that roughly 5% of course registrants complete their course. To many observers, the completion percentage is an indication that MOOCs are fundamentally flawed. However, a reality check is in order. Since there's no cost to registering for a course, many people sign up to see what it is about before making a big time commitment (so called window shopping). Coursera has broken their completion rate data down as follows:

·      5% for those who register,

·      63% who register and say that they are committed to complete the course, and

·      90% who pay ($39-$69US) for Coursera’s “Signature Track” (essentially a “credit”)

The lesson is obvious: you get what you pay for. If you plunk down money for a course, you are very likely to complete it. Not a surprise.

Andrew Ng, London 2014

The other major knock against MOOCs is the monetization model. If courses are free, how does one recoup the costs? Coursera’s Signature Track is generating revenue, mostly for Coursera but a portion goes to the participating institutions. This revenue stream is catching on. It took nine months from the start of Signature Track to generate $1 million of revenue. It took three more months to reach $2 million. The next three months brought the total revenue to $4 million. Of course, one should be careful about extrapolating this trend! Clearly there is a market for students getting a piece of paper verifying that they passed a course.

So far there are no surprises at the Coursera conference. Attendees include a mix of people ranging from university presidents and provosts at one end to interested students at the other. The conference is all about hearing the experience of learners, instructors, and administrators. Everyone has an interesting story to tell.

Perspectives on Dino 101

Last September, the Faculty of Science and the University of Alberta launched Dino 101(dinosaurs!) – our initial foray into the world of MOOCs (Massive Open Online Course). As many of you will know from my previous posts, Dino 101 was to be a high-quality online course that would be offered for free to anyone in the world (credit towards a university degree required a fee). Now that the first offering of the course is complete, it is time to look back at what was accomplished and the lessons learned.

The course took almost a year to design, build and deliver and was the result of the efforts of close to 40 people in the Faculties of Science, Arts, and Education, and the University Digital Strategy group. Over 23,000 students around the world took the course including 400 University of Alberta students (for bureaucratic reasons, we had to have a cap of the number taking it for credit). Feedback on the course was overwhelmingly positive. From start to finish, Dino 101 was a memorable experience for all those involved with it.

From the beginning, we knew we could produce the best course in this domain – with our international strengths in paleontology and local expertise in educational design, assessment, and in building a rich interactive digital learning environment. As a result, we did several things that were well received, including some that go beyond what can be done in the traditional classroom:

• We had a talented graduate student deliver most of the course material (Betsy Kruk). She served as an aspirational surrogate for the course’s sage, Dr. Phil Currie. Our intuition was that students would better relate to a presentation by one of their generation, and the feedback seems to bear this out. Of course, Dino 101 was Phil’s course and he oversaw all the content and did some of the presentations.

Dr. Phil Currie and Betsy Kruk.

• On-location shooting at dinosaur digs in Alberta – including the bone bed in Edmonton and at the World Heritage site of Dinosaur Provincial Park and the superb Royal Tyrrell Museum.
• Engaging interactive software applications. Interactivity is critical for keeping learners engaged.

3D Fossil Viewer (using real scans of bones in the University collections).

• Original artwork was created specially for the course (created by Jan Sovak, internationally renowned paleo-artist).
• Created the idea of “credit by proxy” to allow non-University of Alberta students to earn the equivalent of a University credit.
• Built an effective online testing tool, critical for us to evaluate the learning performance of the thousands of course participants. No one wanted to do manual marking for 23,000 students!

The University's new testing tool.

From my point of view, many things were accomplished by the course offering. Some of them include:

•  Produced an excellent and novel science course, as judged by the students taking it. The student feedback was gratifying.
•  Enhanced the international reputation of our acclaimed paleontology group and opened the program to the world – allowing people to experience science and ancient life in a rich and open learning environment
• Showcased the research and teaching excellence of the Faculty of Science.
• Received provincial, national and international media attention (e.g., BBC News, Huffington Post, and National Geographic).
• Raised our academic profile. Over 50% of the students taking the course had not heard of the University of Alberta before.
• Marketed the course in novel ways (at least for MOOCs), finding interesting and fun ways to exploit the power of digital platforms and social media.
• I hope that for some people, we created or re-enforced their passion for paleontology/science/learning and that this course will impact their future educational and career choices.

Some of the lessons learned were difficult. First, we discovered just how much work is involved in producing a high-quality MOOC. Yes, we learned a lot and can use this experience to streamline the development processes for our next MOOC. It doesn’t change the fact that a high-quality MOOC is expensive and requires a large team of talented people to bring a course to delivery.

Second, it took a while but we eventually “learned” the formula for a successful MOOC. The following picture, created by Jennifer Chesney, Associate Vice-President, University Digital Strategy, shows that the academic side (pedagogy) is just the tip of the proverbial iceberg. It is all the other things (below the water line) that add up to produce a high-quality MOOC. Nothing happens without a topic that has universal appeal (market value), the credibility to talk about it (international reputation), and the passion of the instructional and production teams to do something extraordinary.

The formula for creating a high-quality MOOC.

Finally, we uncovered several challenging aspects of  University of Alberta administrative policies. The University has rules/guidelines that were right at the time they were adopted, but have not evolved to accommodate the new reality of massive online courses. Hopefully, these bureaucratic obstacles can be quickly removed.

In January, we launched our second instance of Dino 101. The material was further streamlined and enhanced (including adding in Phil Currie’s discovery of a rare baby dinosaur. Going forward, we plan to offer the course once annually, probably in September.

I am very proud of what has been accomplished with Dino 101. The Faculty of Science and the University of Alberta are on the frontier of a new age in teaching and disseminating science.  The project represented the strong collaborative efforts of a great team of people, representative of our entire campus, who did amazing things.

What’s next? We are currently filming a second MOOC, part of a joint effort in the Faculties of Arts and Science. Two others are in the planning stage. All of them will be every bit as good as Dino 101. In addition, we are analyzing the data gathered from Dino 101 and will turn the insights gleaned from the course into research papers and into recommendations for better practices going forward.

Many people doubt that MOOCs are the future. However, the momentum towards putting more course material online and freely available is unstoppable. There is no turning back. We do not know how this will play out, but the University of Alberta needs to be a leader in inventing the future of online teaching.

My Mantra

In the early years of my academic career, I was insulated from the politics of the university and the complexity that is the research funding reality in Canada. Once I was sufficiently far enough along in my career, I started assuming leadership positions. Quickly I observed a roadblock that limited what could be achieved. I have seen this pattern over and over again, to the point where I started becoming vocal about it. The problem is simple: division instead of unity. Let me give three examples from my career.

Canada is largely decentralized in its administration. Ottawa has responsibility for some areas (e.g., national defense) and the ten provinces for others (e.g., education). So when you get an area that overlaps national and provincial interests, problems ensue. Starting in the mid-1990s I became a passionate advocate for creating high-performance computing infrastructure for researchers. Early on, it became obvious that there were problems. Ottawa had one idea of what was needed; the provinces had others. The end result was what you would expect: friction that continues to this day. The reality is that if everyone could come together and work as one, at least in the high-performance computing area we could reduce duplication, simplify processes, create a national strategy, and give better value to researchers. But, hey, we would reduce or eliminate some (very important) empires, and that would be a terrible thing to do.

In Canada, we have a variety of organizations that support some aspect of information technology for researchers (e.g., one organization is responsible for research networking; another is responsible for the high-end computing you put on the network) and a plethora of  funding agencies that provide money for IT infrastructure, software, support people, and research grants. Heaven forbid that we simplify the Canadian system and merge some of the organizations to create critical mass (instead of critical mess). We could reduce duplication, simplify processes, create a national strategy, and give better value to researchers. But, hey, we would eliminate some (very important) empires, and that would be a terrible thing to do.

As Vice Provost and Associate Vice President for Information Technology at the University of Alberta (I had the longest title of any administrator – size matters!), I was responsible for the institutional IT strategy. But there were over 50 independent IT groups across campus, each with a separate agenda, budget, and vision. The reality is that most of these groups did not have sufficient funding to accomplish their goals; they were continually in a reactive mode putting our fires, rather than being in a proactive mode and providing new value to their stakeholders. Many of these groups were unsustainable, especially in an environment of repeated budget cuts. It made sense that there needed to be some unity of purpose.  If everyone could came together and work as one, at least in the IT area we could reduce duplication, simplify processes, create an institutional strategy, and give better value to faculty, staff, and students. But, hey, we would reduce or eliminate some (very important) empires, and that would be a terrible thing to do.

How do you solve the problem? First, you need to recognize that there is a problem. This is can be harder than you think since some groups will deny that there is a problem as a way of protecting their independence. Second, you need to open a dialogue between the groups. Unfortunately, dialogue can be perceived as a threat. Finally, you need a solution. The obvious route is to merge the groups, but the political or social reality may make this impractical. There are other ways that may be as effective (or more effective) to achieve the goal of a united voice. For example, creating a vision that everyone can buy into, or creating a greater level of communication or cooperation, might achieve most of the benefits. Regardless of the solution chosen, effecting major change is rarely easy.

My frustration with the fragmentation that seems to surround the groups that I interact with led me to create a mantra that has served me well over the years:

United we grow

Divided we status quo

The phrase is simple and gets at the heart of the matter. To many people, the status quo is acceptable; resistance to change is palpable in so many areas. But the world changes and organizations/processes need to evolve. Often the only way to effect substantive long-term change is to have a united purpose and achieve critical mass. Do you want one hundred small villages or one powerful country? Which will offer better value for the stakeholders in the long run?