Captioning Videos

When I mentioned earlier in the week that I had made a bunch of videos for my online East Asia course last semester, a colleague challenged me on twitter regarding whether the videos were captioned. I responded that the ones I was talking about were not part of my OER, but just short messages to my students announcing what was next, explaining how to use tools like Hypothesis, etc. He responded that even so, federal and state law and our university system’s policies require captioning of any course video. 

I was not sure this is true, so I asked him to provide a reference to the statutes he thinks apply. He provided a link to a MinnState guideline document that seems to agree with his claim that the system I work within supports a broad interpretation of the law that seeks to require captioning of all videos, regardless of their purpose. Among the questions addressed in the document’s FAQ is an explanation that transcripts of videos are not sufficient, because the user experience is not identical. While I accept that this statement is true, I’m skeptical of its importance in all cases. For example, If I make a video version of a PowerPoint lecture based on a chapter of my OER textbook, the same basic content is presented in three different formats. My initial idea is to make the content accessible to students with different learning styles. I think, some students learn better by reading: they’ll use the textbook. Others may have time while commuting: a podcast might be useful to them. Others may miss one of my lectures of may really learn better watching a video: so I’ll make a video version of my lecture and post it online.  

The idea is to make my course content more accessible. Requiring me to caption these lecture videos complicates the process of making them and adds time. Let’s be honest: increased complexity and time are disincentives. Not only for someone like me who wants to make videos, but even worse for instructors who are on the fence about the shift to online resources or looking for a reason to reject the idea. Moreover, the type of aggressive virtue-signaling language that sometimes accompanies arguments or mandates of accessibility can be very off-putting. Maybe it’s me, but I’d prefer to be helped to make my content more accessible rather than told if it’s not 100% right out of the gate I shouldn’t bother. And if I was paranoid (who me?) I might suspect that although the proponents of universal accessibility have their hearts in the right spot, putting steep barriers in the way of instructors seeking to make a change seems to play into the hands of textbook publishing corporations.  

Don’t get me wrong. I’m going to try to make my videos with captions, even though that’s going to be more complicated and time-consuming. I’m going to try to find the least time-consuming and expensive way to do that and share what I find. If you know of any better, faster, cheaper ways, please let me know! Although I’m fairly comfortable with technology and have a lot of opinions I’m not shy about sharing, I’ve only been doing this OER and open ed stuff for about a year. Point out what I’m missing and I’ll work it in and make a video about it! 

Screen Shot 2019-12-29 at 9.53.19 AM

As I’ve begun looking at the tools available, there seem to be a several ways for me to type caption text into a video as I’m watching it. This isn’t optimal. If I’ve written copy and then made a video of it, the last thing I want to do is then retype it. I suppose I could cut and paste it a little bit at a time, but this is also a pain, especially if I’m captioning a lecture. So the best option seems to be finding a reasonably accurate, easily-correctable auto-captioning app. Again, there are a bunch of these available, but I’m not going to buy one in order to make my videos. Sorry. I paid for Camtasia, but there’s a limit to the amount of my own money I can afford to spend. I was disappointed to learn that Vimeo, the video-hosting service I’ve been using for years to avoid YouTube, allows you to add captions to videos but doesn’t have an automated process like YouTube’s. It supports automation tools like CaptionSync, but again, I’m not going to buy a subscription to an app that doesn’t even list its pricing online! So maybe it’s time to surrender to the dark side and stop paying $199 a year for an inferior tool.  

Screen Shot 2019-12-29 at 9.52.39 AM

Roughly six hours after writing the previous paragraph, I’ve removed over 150 videos from Vimeo, canceled my $199 per year “Pro” subscription, and loaded eleven lecture videos from my American Environmental History course onto YouTube. The auto-captioning seems to work remarkably well! Not perfectly, and I’ll need to spend a little time with each video, editing the captions. I’ve shifted the licensing on the videos from standard YouTube license to CC-BY, the Creative Commons Attribution open license that allows anyone to reuse or remix the videos. I don’t think it’s really possible to download the YouTube versions of the videos, so maybe I should load the originals up into Opendora (my system’s open archive) so that others who want to use the content can grab an mp4 file and mess with it.  

My next step will be to see if it makes sense to load videos like the lectures I just added to YouTube into an app like MediaSpace in my university’s learning management sandbox. I’m not actually teaching the course those videos go with again this semester, and I’m rethinking a lot of design elements of the video lecture format, so I’ll need to ponder on that a bit in the next couple of weeks as I prepare for next semester. I’ll report on the process as I go, so thanks for reading and I’ll see you again next time.  

Time Study

At the beginning of the semester I proposed a project to promote OER at BSU (which was not approved). In the process I was asked how much time I thought it would take and not having any data I grabbed a number from thin air. This prompted me to keep track of my time this semester, to see exactly how long I spent on each of the things I do at work. In addition to wondering how much time I spend working on OER I was curious about the more traditional triad of teaching, research/writing, and service. BSU is a teaching-oriented university, so I suspected the majority of my time would be spent working on my courses, but I wanted to find out for sure.

I used an application called Tyme 2 which I installed on my desktop and notebook computers, iPad and phone. It allowed me to create twenty tasks in three categories: courses, OER, and Professional Development (PD) which included both service tasks like advising and attending committee meetings, and also my work preparing the edits of my book for publication. I tracked each of these tasks over seventeen weeks from the last week of August to the end of the third week of December. While there may have been some slight overlap between OER activities and course activities, I think the results are pretty accurate.

So what were the results? Turns out I spent two thirds of my time working on my courses, for a total of 29 hours per week. I didn’t count one week (Thanksgiving) when I was at an OER conference and only did an hour of course-related work. If that week counted, my weekly coursework average was about 27.3 hours. Similarly, if I don’t count the conference week when I devoted about 85 hours to OER, my average time spent working on OER was about 8 hours weekly. My weekly PD total was 7.3 hours, made up mostly of work proofreading the first print of my manuscript and writing an index. My average workweek, not counting the conference week, was 44.2 hours.

Screen Shot 2019-12-25 at 9.42.44 AM

The total time I devoted to courses over the semester was 485 hours (including some prep for courses I’ll offer in the spring). The largest block of time went to a new course, History of High Technology (HST 2600, 142 hours). The smallest block went to East Asia History (HST 3419, 81.5 hours), which was another new course but was online. The difference between the courses can be mostly accounted for by the 40 hours of meetings of the in-person class and the time it took me to produce PowerPoint lectures for that class. I made short videos for the online class which were much less time-consuming. Even so, the online course prep seemed to be slightly more efficient. Compared to the two in-person courses I’ve taught before (HST 1305, 103 hours and HST 2925, 102 hours), the online course still seems more efficient. Student evaluations of all these courses were similarly positive, and I felt about the same about my effectiveness in each of the new courses (a good start but there were some things I could improve); so although the sample set is low I think there may be some significance to these results. I’ll be more aware of this efficiency question in the future.

Looking forward, I plan to continue trying to streamline my courses using technology (more effective LMS tools, Hypothesis, online assessment, etc.) and to explore the effectiveness of online vs. in-person delivery. I had 30 students in my online East Asia course and 12 in my in-person High Tech, so at a very raw, numerical level the lesser time I spent on HST 3419 was more effective. Early in the semester our CPD ran a brief session about efficiency in course design at the Deans’ request. It was mostly oriented around surviving higher course caps and just scratched the surface. As we work to reverse decreasing enrollment at BSU and struggle with increasing class sizes, I think effectively and efficiently delivering online courses is going to be key.

 

Takeaways from Milan

It has been a little over a week since I returned from the open education conference in Milan. Looking back on it, I think it was a valuable experience for me and a good introduction to the OER and open ed efforts being made by educators and policy-makers in Europe and Asia (there was only modest representation from Latin America or Africa). Much of the talk was oriented on social justice and equity and a great deal focused on the United Nations Sustainable Development Goals and on the recent UNESCO OER Recommendation, adopted at the 40th General Conference on November 25, 2019. Both these international statements can be easily related to BSU’s Sustainability Goals and focus on educational equity, and I’m going to begin referring to them more explicitly as I design my own courses and activities. The biggest takeaway from the conference, however, may be the rapid pace of change toward a much more decentralized educational system in which traditional institutions such as universities are decentered and greater student ownership and control over curricula and credentials become the norm. The three themes that seemed most prevalent in conference sessions and discussions were online education, MOOCs, and micro-credentialing.

Online education has been widely accepted as a solution for distance learning, but is increasingly seen as a way to enhance students’ learning experience. New technology enables interaction and automates many of the routine activities of running a class, freeing up both students and instructors to focus on the learning. For example, Moodle (one of the conference sponsors) demonstrated several new tools for collaborative learning and assessment. And the increased reach of courses offered in either standard or massively online formats improves discussion and student interaction by raising the number of participants. I’ve already experienced this in my own online East Asia course which has thirty students, triple what I’ve ever had in an in-person 3000-level course at BSU. Next semester I won’t have an online course, but thereafter I plan to lean into online with dual-listed upper-level courses in the summer and every semester for History majors, Social Studies Ed. Students, and High School teachers. And I’m looking forward to trying an online survey, beginning with People of the Environment in Fall 2020 and continuing with Modern World.

Massively open online courses (MOOCs) are much more of a dirty word in the US than they are in the rest of the world, possibly due to the way some American for-profit education institutions have misused the format. In Europe and Asia, MOOCs are to courses as OER are to textbooks. For subjects where there is a significant percentage of fact-learning, assessment can be automated fairly easily. Even when qualitative judgments must be made about discussion posts and forum interactions, vendors like Moodle are developing peer evaluation modules with AI agents that prioritize the evaluations of students who have earned high evaluations on their own work (while still allowing the instructor to have the final say), encouraging student collaboration in course management as well as learning. Even if we don’t decide to go all the way to the MOOC environment, there’s a lot we can learn from these courses and the tools they use to improve the learning experience.

Credentials are central to the power exercised by universities over students. There have been online sources for the highest-quality educational content for the last couple of decades, such as MIT OpenCourseWare, Britain’s Open University, and open learning initiatives at Stanford, Harvard, and UMass. Students can watch and listen to some of the best instructors for free, read open textbooks, and educate themselves in a wide variety of topics – but they generally can’t get credit for that learning. Colleges and universities are being challenged as the gatekeepers of credit, however. Micro-credentialing apps such as Badgr are gaining credibility. Badgr is currently used by more than 12,000 credentialing agencies in 100 countries. The NEA recognizes badges and micro-credentials, and several university systems like SUNY have pledged to lead the way in “High-Quality Micro-Credentials”. I attended a workshop on a new blockchain-based app being developed to allow organizations to decentralize credentialing. I suggested that individuals ultimately will want to own their own “personal wallet” of credentials (as well as other personal digital info such as their genome, credit history, medical history, and CV). Martin Dougiamas (CEO of Moodle) picked up that thread and said Moodle was working on a decentralized network for educators and students that he hinted might include some of these features. MinnState has just been through an extensive formulation of transfer pathways, which may be a good first step in a process of thinking about how we want to respond to this challenge.

As usual, I’m going to advocate for trying to be ahead of the changes and meeting them with a plan. I think my department’s shift toward offering more online courses for concurrent enrollment teachers (teachers in Minnesota who want to teach “college at high school courses” have recently been required to have at least 18 credits in the subject they’re teaching, in addition to education credits) is a great first step. It addresses our need to increase enrollment while giving us practice improving our online courses. Using more automated tools may take some of the pressure off instructors and allow us to focus more on quality interactions with students. And making our courses widely-visible models should help insure our relevance as residential universities begin to lose their position at the center of higher ed.

I’m at the University of Minnesota in Minneapolis today, participating in a group for the Open Textbook Network to define specifications for a new OER authoring and publishing tool they’re going to develop. But I’ll be back Friday, and I’ll start talking to folks in my program and on campus about these issues.