Instructional Video in the Academy: In the Can or On-the-Fly?
Five years ago, the claim that an online course can be just as effective and engaging as its equivalent in-person course may have raised eyebrows and possibly even ignited controversy. Since then, the bar has risen for online courses to the point at which you can claim that online instruction should be even better than the same course taught in a classroom without raising heated objections. Online courses free instruction from the arbitrary constraints imposed by allocation decisions of physical classrooms and limitations of what can be done in them during fixed class times.
One of the core purposes of formal education is to develop in students the foundation necessary to be lifelong learners. In our hyper-connected world, much of the learning and skills training done by adults is self-directed and online, and that trend is sure to increase in the coming years. Teaching students how to learn effectively online is a necessary function of high-quality schooling. While streaming media is not a necessary component of a high-quality online course, as I argued in a Class Act column two years ago, a strong student expectation is that there will be video in the curriculum.
In educational technology nomenclature, the main distinction between video services is asynchronous versus synchronous. Asynchronous video is traditional streaming media, based on what we would now call an OTT, on-demand video platform, in which video is streamed from a server to a client for viewing. Synchronous video is akin to multiparty video conferencing, in which the class meets in real time, with video streaming to and from all teachers and students in the online class session. With improved network quality in many areas and the reduced cost of computation, synchronous video platforms are quickly coming of age.
Asynchronous may be a confusing term to apply to a modern streaming media service: The idea is that the students don’t need to be using the platform at a specific time. Students can instead watch course video whenever their availability allows it—the learning is asynchronous with the teaching. The trade-off for that freedom is that they don’t typically interact with anyone through the platform while using it; asynchronous video is a one-way street, with only QoS and viewer behavior metadata going back to the server. In ed tech parlance, “asynchronous” typically means on demand rather than live, although the platforms for asynchronous video delivery are also often used by schools for broadcasting live streams of events. Live streaming is very rarely useful in support of the instructional missions of schools, so the distinction in an educational context is not as confusing as it would be for media professionals.
The Asynchronous Microlecture
The massive open online course (MOOC) era brought about an explosion of experimentation in lecture video production techniques and more quietly a massive trove of data on student behaviors within the MOOCs. One major finding from that dataset identified the effectiveness of the asynchronous microlecture. Multiple studies show that students prefer and learn well from short videos covering a very specific topic rather than from one covering multiple topics over a runtime that’s comparable to a traditional 50-minute class session. In fact, the effectiveness of the microlecture has influenced face-to-face teaching both in dramatic ways, with flipped classrooms, and more subtle ways, with teachers consciously structuring their in-class teaching periods around short 5- to 7-minute lectures alternating with activities or discussions to flesh out the students’ understanding of the material. Flipping the classroom is a way of running a course in which students are expected to watch the lecture video prior to attending class so that the classroom time can be spent more interactively doing deliberate group activities that reinforce the lecture materials. This is an attractive option: It seems like a good use of class time and an opportunity for students and faculty members to interact with one another rather than having students passively listen to the teacher lecture. The other scenario described might be appropriate for schools in which students don’t have reliable access to streaming video at home or in which the teacher is more interested in following the data-driven best-practices than producing lecture videos.
The asynchronous microlecture is here to stay and has major advantages for both teachers and students. The teachers benefit most obviously in that they can nail the lecture performance once and use it term after term, freeing their time to develop engaging activities and monitor student learning. The quality of the lecture is also not susceptible to degradation by students interrupting, either with a misunderstanding that takes the discussion off on a tangent or with an insight that prematurely tips dominoes being carefully set up by the thoughtfully prepared lecture. A great lecture doesn’t cover all of the material in the topic, but instead leads students around the rabbit holes they can explore on their own time.
The students benefit from the lack of interruptions, but additionally gain the ability to rewind and hear a portion of the lecture again, to tailor the playback rate to suit their available attention, and to take advantage of high-quality captions to ensure familiarity with all the terms used. And of course, the students can watch the lecture on demand to fit into their schedule instead of needing to be in class, well-rested, and ready to go at a specific time.
The Synchronous Video Platform
If asynchronous video is something like on-demand broadcast television, synchronous video is souped-up videoconferencing done in software rather than H.323 hardware installations. The simplest possible synchronous system is one without a platform, using peer-to-peer streaming over WebRTC. This can be done through WebRTC demonstration websites like Talky.io, but this solution won’t scale. The problem is that a peer-to-peer multiparty videoconference requires everyone in the call to stream video back and forth individually to everyone else in the call. Since most internet connections are asynchronous in the sense that upload speeds are slower than download speeds, often by one-fifth the bandwidth, the upload link will quickly saturate when having to stream out a copy of the video to all parties on the call. With a respectable class size, the download links would also saturate if every class participant were streaming out HD video. Synchronous video platforms solve this problem by providing an endpoint where all participants stream their video; that server then either streams out, to all callers, lower-bitrate and lower-resolution versions of non-focused participants plus, optionally, a high-quality version of the person currently speaking or even composites the videos into one outbound video stream. Synchronous video platforms, then, require substantial resource provisioning both in bandwidth and CPU.
To learn more, I strongly endorse the open source Jitsi project—formerly owned by Atlassian and now by 8x8—and recommend trying out its demonstration platform called Jitsi Meet, which supports free WebRTC-based calls with up to 50 participants. An interesting feature it included is the ability to pull a YouTube video into focus; in an online synchronous class session, that might be akin to the breaking up discussion with asynchronous lecture videos. Another great resource for learning about this technology is Tsahi Levent-Levi’s BlogGeek.me, where he very helpfully evangelizes it.
A four-way call with Jitsi Meet, with our jolly author at the bottom right
Finally, it should be noted that with adequate CPU provisioning, venerable media servers like Adobe Media Server, Wowza, and Red5 are all perfectly capable of and originally designed for serving as back ends in both synchronous and asynchronous video platforms, although any of them are typically exclusively used by schools for asynchronous video at scale.
Educational Use Cases for Synchronous Video
In contrast to the advantages of asynchronous microlectures previously described, the synchronous online lecture is the worst of all worlds. Teachers need to perform the lecture each time the course is offered. Depending on the platform, students can interrupt the lecture as easily as in a traditional physical classroom (and may feel a strong need to, since they cannot rewind). Students and teachers must be available at specific times, which is not only inconvenient but also risky, since synchronous video sessions are a case in which education trades paint with show business. If there are technical problems with any component of the synchronous lecture system, such as the teacher’s camera and microphone or the platform to the student’s computer setup, the scheduled class time is frustratingly wasted for many people, and critical lecture content is missed. It is thus necessary for teachers to be well-trained on how to use their synchronous platform and A/V gear to avoid tech problems that at the time feel cataclysmic.
In higher education, 79% of institutions use a video lecture capture solution, letting students return to lectures as easily as they can return to text-based material.
The live news network makes its second major expansion of the year, this time with a college-focused news streaming to 1,600 public screens.
College professors don't need to be video experts with Kaltura Lecture Capture which offers one-button recording. Live video support is in the works.
Video advocates must continually work with (sometimes reluctant) professors to understand the power of video and to find a simple way to deliver it to students who are using a variety of equipment, software, and internet connection speeds.