Epsilon XR is a Division of Epsilon Systems Solutions, Inc.

The Evolution of Training Technology


“We would take large groups of students and divide them into three groups: a web-based group, a face-to-face group, and a blended group; and we would look at outcomes. When we looked at the data, the group that did the least well was the group with face-to-face. The group that did by far the best was the blended group. So, after one round of experimentation, this got super interesting to us. All of a sudden, we could add this kind of electronic means of knowledge transfer and suddenly really change education.” Murray Goldberg, CEO of Marine Learning Systems, joins us to give insight on the evolution of learning technology.

Murray Goldberg was a tenured faculty member conducting research on the effectiveness of web-based learning in the department of Computer Science at the University of British Columbia. In 1997, Murray founded WebCT, which grew to be the world’s leading LMS serving 14 million students in 80 countries at 4,000 universities and colleges. WebCT sold to Blackboard in 2006 for $200M. Since then, Murray has created a new company, Marine Learning Systems, to address learning in the maritime and other skill-oriented industries. Marine Learning Systems is experiencing rapid growth and now counts among its customers more than half of the world cruise segment, as well as ferries, workboats, coast guards and others.

Goldberg has won numerous prizes including the UBC Killam University Teaching Prize, the 2000 National IWAY Award for outstanding contributions to information technology, and was named as the recipient of the New Media Hyperion award new media in IT. In 2004, Goldberg was granted an Honorary Ph.D. from Southern Cross University for pioneering work in advancing Ed Tech. globally. Also in 2004, Goldberg won the national Manning Awards foundation Principal Award, a $100,000 prize honoring the country’s most outstanding innovator. Most recently, Goldberg was named one of the top 15 Canadians in digital media by Backbone Magazine.


Host – Steven Maggi: They say you can’t teach an old dog new tricks, but apparently with technology, you can; and mainly in the maritime business – which I think of “Gorton’s Fisherman” statue and what have you. But actually, the maritime industry has learned a lot from our guest today, Murray Goldberg. Murray began his career in the academic world at the University of British Columbia. He got into technology – we are going to talk about all that today – and now heads up a company called “Marine Learning Systems”. Well, Murray, was it kind of a surprise to you?


Murray Goldberg: I really thought I was going to be a faculty member for the rest of my life, because as you mentioned, I was a faculty member – I was a computer science faculty member in fact. My research area was online learning and web-based learning, looking at blended learning and effectiveness and student satisfaction. But, it was part of my teaching and I loved teaching. I never expected to ever leave that.


Steven: But you enjoyed it so much that you actually developed some software. Talk a little about that, because that was back in the early part of these online courses and that kind of thing.


Murray: It really was. So, I am old, and an old dog – as your introduction went – applies not just to the industry that I am working in now, but it applies to me equally well. We are talking way back in, I think 1995-1996, when I was starting this research. It was some of the earliest research on web-based learning and blended learning, because there really wasn’t much web-based learning or blended learning in existence at the time. The tools weren’t there, people weren’t doing it, but it was kind of an emerging thing. So, we did some experimentation. We would take large groups of students and divide them into three groups: a web-based group, a face-to-face group, and a blended group; and we would look at outcomes. It was really just something we could do with computing at the time. I didn’t have any huge designs or certainly didn’t have any predispositions about how I thought it was going to come out. But when we looked at the data in doing this experimentation, it was actually very surprising to me, because the group that did the least well was the group with face-to-face – the group that had me standing up in front of them teaching them – which was a little disheartening. The group that did by far the best was the blended group. So, after one round of experimentation, this got super interesting to us. All of a sudden, we could add this kind of electronic means of knowledge transfer and suddenly really change education. I mean, we’ve been teaching the same way for 1,000 years or more than 1,000 years, and all of a sudden, we had a tool which had the potential to really upend outcomes and effectiveness.

Anyway, so we did that, and one thing led to another. We built some technology to facilitate the continued experimentation, and that technology ended up becoming one of the first learning management systems. So now, we are talking late 1996, early 1997, and so I left the university to start a company around this. The company was called “WebCT”. Now by this time, there actually was a lot of pressure in the community – especially in higher education – to look at putting courses online and implement blended learning, and so there was a tremendous built up demand. So, despite everything I think I inadvertently did to try and kill the company, we actually grew quite large. We grew and grew to the point where we had about 14 million users in 80 countries, and about 4,000 universities and colleges around the world using WebCT, and as I say, really just because it was the first commercially successful learning management system for higher education. It was a very exciting time.


Steven: Coming from the world of traditional academia, it had to surprise you, but did you start to see as you got into it, “Wow! This can really take what I was doing before and bring it to a whole other level!”?


Murray: Absolutely, and in a number of ways. One was just scale, effectively. We have a scale problem in training and education in general, if we are looking at traditional means, right? You need somebody standing up in front of a group, and the larger the group gets, kind of the less effective it becomes. An instructor can only be with one group at one time. So first of all, there is the potential advantage of scale, in that if you build these technologies and you build these course materials, the scale is more or less endless, certainly not for interaction, but at least for that initial phase of knowledge acquisition – you know, the first part of the blended learning – so that was very interesting. And I guess the other part of it was, what I mentioned before, which is really not that much had changed in education for a very long time. And so, this was the first evidence that we had that applying technology could actually change outcomes, and in a really meaningful way. So, I think it opened a lot of people’s eyes to “Wow! That’s interesting! That’s a fantastic first step! What’s next?”. Interestingly, not that much was next for probably the next five or eight years. Everybody was waiting for the next shoe to drop. Now this box of potential had been opened up, and everybody is thinking, “Wow, that’s fantastic! This is going to change everything, and the pace of change is going to be remarkable!”. Well, actually not that much happened. We refined what we did over the following 10 years, and it’s actually been up until maybe the last three or four years that we didn’t see much. Now, we’re starting to see some really new and interesting stuff come about.


Steven: Do you find the reason for that is because maybe they were looking before just for technology to improve and “All right, we’ll wait for that and then we’ll figure out how to incorporate that into the program”, as opposed to “How can we work at this together? What does the end user really need?”; and then, “How do we make the new technology work for that?”?


Murray: I think that was a big part of it, and it expanded a little bit, because it wasn’t just the end-user that we were really looking at. I mean, you have to remember that our institutions – just in the same way that our teaching hadn’t really changed for 1,000 years or more – our institutions hadn’t fundamentally changed for 1,000 years or more as well. Universities were universities. They were all built on the semester model or the trimester model. They all assumed classes were in person, and there was a little bit of a strut there. So, you had some people who were experimenting around the edges – you know, MOOCs started appearing, the massive online education stuff – others were looking at how to restructure course curriculum and credentialing in light of these new technologies. And so, yes, it was absolutely the case that people were struggling with “Ok, how do we now take this technology that we’ve developed and best apply it for the advantage of the student?” but also “How do we fit it into this world context that really hadn’t changed for a very long time?”.


Steven: I know you did this whole thing with the British Columbia Ferry Services, and you came up with some interesting numbers and so forth. What did you learn and how did those numbers run out?


Murray: Yes, that was really fascinating to me because my background, as you know we’ve described so far, was really in higher education. It wasn’t in the training – the corporate training world – it was in the universities and colleges. But you know, every bit of research that we did, and frankly all the subsequent research – and there’s been so much of it – attests to the fact that blended learning really works. It’s a huge advantage over just single model learning, and face-to-face in particular. One of the biggest outcomes of all that research was this effect is discipline agnostic. So, what works in higher education should work elsewhere, but it wasn’t really applied a lot elsewhere – certainly not in the first few years of its emergence.

In 2006, what happened was – and I’m based in Vancouver here – we have a large ferry operator off the coast called BC Ferries. They unfortunately had a sinking, and as a result of that sinking, they really did a lot of self-examination and decided that they wanted to become a world-class training and safety organization, and by the way, it’s not like what they were doing before this time was bad in any way, it was actually quite standard for most maritime organizations. They were a good organization, but accidents do happen. Yet, they decided that they didn’t just want to be good, they wanted to be the best of the best of the best. And so, now of course we are into late 2006, early 2007, and a lot of the research around blended learning and its effectiveness has already been published. So, they decided that what they wanted to do was do a blended learning pilot, and it was meant to replace their typical training model, which was a lot of job shadowing. So, if you take somebody in a position of like a deckhand, let’s say, and they are new to the job. He or she will come on board in a job shadowing kind of training model, and they will spend three weeks, four weeks, five weeks, even six weeks following somebody around, somebody more experienced. They will learn the ropes that way. Now, we probably don’t have to go too deeply into the various pieces of that model that are flawed, I mean, obviously they are not getting a lot of standardized education – it is not a good way to measure that – there’s lots of problems with that. But, nonetheless, it was a pretty standard way to educate people in maritime. There’s lots of good reasons, by the way, that it should be a way to educate, because it may be that you and I as deckhands, we need a lot of the same fundamental information, but as soon as you step into your role on one vessel and I step into mine on another vessel, our jobs are very different. So, there’s lots of reasons why job shadowing was kind of a norm.

Anyway, so what happened was BC Ferries hired me as a consultant, and hired a lot of other people as well – it certainly wasn’t just me – and we created a pilot, which was a blended learning replacement for that job shadowing model that I talked about before. The pilot worked out very well, and so they decided that what they needed to do was to roll it across the organization as part of a larger safety culture initiative called SailSafe. So, in order to do that, they needed technology, they needed a learning management system. And it turned out that upon investigation, there were no learning management systems that could really address the operational and training context of a safety-critical maritime operator like them. So, they turned to me and said, “Well, build us one”. So, okay, I spent about the next year or so building one for them. Once it was built, it got rolled across the organization step by step by step through the various roles, through the various departments. It took about six years, and what we did was we looked at how training was proceeding, and we looked at how accidents were affected, how insurance claims’ costs were affected, and these kinds of things. BC Ferries, because they had done this deep introspection, they decided that whatever they did to improve the situation, they were going to be very transparent about so others could learn from it. And the results were actually astounding.

So, after six or seven years, when things were completely rolled out – and remember, this isn’t only the blended learning; this is blended learning together with other safety culture initiatives, because it was a combined effort – a few things happened. First of all, accidents went down by about 60%. Serious accidents went down by the same percentage. And that’s interesting, because those people who have been involved in similar situations before know that it’s often the case that minor accidents will go down through an initiative like this, but not serious accidents. Here at BC Ferries, serious accidents went down about 60% as well. Insurance claims’ costs went from about 3.5 million dollars a year, to about $800,000 a year; this was really shocking. Days lost due to accident, where a one-day loss is one employee not showing up for work once in a year because he or she has injured themselves on the job, went from about 12,000 a year – it’s a large organization – down to about 7,000 when the program was completely implemented. And, most interestingly, it didn’t stop going down after that, so even though the training and the safety program were completely implemented by that time, it seemed to have created kind of a positive feedback cycle that continued to drive down days lost due to accidents. The latest number that I have is now down to about 4,000 days lost due to accident in a year. And again, why is that happening even though everything was implemented and more or less finished being implemented a few years ago? Probably because when you change a safety culture like that, and you change a training culture like that, the people who are in the organization for whom that does not resonate slowly leave the organization, and that’s not a bad thing. And you create this reputation in the industry for really caring about professionalism and really developing your employees, and so for those who are looking for work in the industry, that reputation is known, and they are going to start applying to BC Ferries if that’s the kind of model that most resonates with them. So again, it creates this very nice positive feedback loop. So, that was a fantastic kind of experience, and in fact, it was the thing that took me in this surprising direction, which was starting a company around learning technologies for the maritime industry.


Steven: Well Murray, you’ve got a product you’ve created, SkillGrader, that is a revolutionary product for the world of assessing skills. Can you come back with us and discuss this maybe in more detail?


Murray: I’d love that. Anytime at all, Steve.