With these major projects done, I had some time to think about what to finish next. For some time, I thought I would put together “companion exercises” videos, which frankly won’t help my students that much—they already have access to homework problems and homework help videos—but would add content on the YouTube channel.
Instead, I decided to work on the transcripts. So far, all my videos only used auto-transcribed transcriptions. It’s technically an ADA violation, but the amount of time that would be necessary for the hand-corrected transcriptions was too much, and I was mainly worried about doing all that work to fix the transcripts, only then for me not to use the video anymore. Well, now I have a set of 90 videos which I’m hoping I will use at least for 5 years or so, so I think this is the perfect time to put in the necessary effort to fix this ADA violation.
I started working on Physics 4A lecture video transcripts, and the progress was really slow—I think it took the whole day and a little more for a single relatively shorter video. But I think I found a good workflow that utilizes ChatGPT. It turns out, as it could work on the lecture questions from the auto-transcribed transcripts, it can also fix what the transcript is likely to be from the auto-transcription alone, without directly accessing or re-transcribing from the audio.
And ChatGPT 4o is especially good at this work. See below example. Now ChatGPT did have some access to the context which might have helped, but, you know, I, as the lecturer, know all the context, and it still would have taken me quite a bit of time to do the level of correction that ChatGPT did automatically.
In this workflow, it takes me maybe about double the video length to fully correct and review—to the extent needed—the corrected transcript, and I think I’m actually going to go back to re-do some of the transcripts that were done earlier … with some faults. But even with re-doing two … two and a half lecture videos, I think I’ll have enough time to have some lead time built in, before the semester starts. I’ll probably be able to be about 2 weeks ahead of students, as I fix the transcripts. And I have some plans for the transcripts once it’s done; it’s not just for deaf students; there are other materials that I’m thinking about making, and I would need relatively accurate transcripts of the lecture videos to start working on that.
P.S. I actually started this work using ChatGPT o3, because I’m used to using that model for coming up with the lecture questions, but actually the o3 model was doing “too much thinking”, mainly changing the sentence structures around too much, despite numerous instructions to be “verbatim”.
The idea for the re-organized lectures came to me at the end of Fall 2024 semester, when I thought about feedback from a few students who successfully completed that semester (holding multiple 1-on-1 meetings until they eventually met the standard for an A), that they did find the lectures helpful. From several years’ experience, I knew simply linking the lecture videos on an organized Canvas page wasn’t getting students to watch them (who knows what the reasons are; I have heard pretextual reasons before (like a student who claimed one particular early video was confusing and used that as a pretext to justify skipping all the rest); what I do know statistically is it’s a small minority of a given class that watches a majority of the lecture videos). Some students would skip to the homework help videos, which don’t help them learn as much as they think it might (… because they are reversing my advice, to watch the lecture videos and skip the homework help videos, at least until after trying the questions on their own). Anyways; I had a longer post for the motivation here: Winter 2024 Update – Lecture Videos
The projects for re-organizing lecture videos were an unqualified success—in fact, when I started putting together the course materials for Physics 4A (Spring 2025) using the re-organized lecture videos is when I saw the immense simplification it brought about, and that got me to change my mind about taking on a similar project for Physics 4C (Spring 2025, also), even though that meant a pretty substantial time commitment, non-negligible possibility of falling behind in posting course material, all for a course that I probably won’t teach again in a long time (if all goes well, we’ll be teaching Physics 3A in Spring 2026 and not Physics 4C). I think the simplification is best illustrated in below screenshots. The first screenshot is how a typical weekly sub-modules look in a Physics 4A class before the re-organization; the second screenshot is after re-organization.
Before re-organization, I had three assignments (conceptual questions, lecture reflection, and peer review) that were trying to be the “lecture assignment” (i.e. assignment tasks for the online asynchronous class that replicates/replaces the 4 hours per week of in-person lecture). And I had grown weary of these assignments (too many students were using ChatGPT on conceptual questions; very few students were doing peer reviews earnestly; most lecture reflection posts were basically “no comment” posts). Having that video-cued assessment meant I felt comfortable getting rid of the complex structure of those assignments, replacing them with the much simpler setup of lecture assessment and weekly problem set.
And I feel good about how the re-organized lecture videos came out. In the re-editing process, I taught myself a few things to improve board readability and make the lectures more watchable (“board aid” recording in worst cases; 3D transform to re-shape the wide-shot video; simultaneously displaying shots from left and right cameras, in those circumstances when I was writing in the transition area for the two cameras). And the self-imposed length restrictions (between 1 to 2 hours long, whenever possible, 2 videos per weekly module; sometimes one (1) when students had other tasks to complete that week, and on rare occasions three (3) when there are topics that I want to present, even if “optional”) gave me the right reason to cut out material that we could do without, while re-reviewing some of the older footage to see if they could be edited into usable material. Overall, the videos I have in the re-organized lecture videos playlists (Physics 4A in 30 Lectures and Physics 4C in 30 Lectures; BTW, I’m currently working on Physics 4B in 30 Lectures) are better quality than the videos I started out with. So, I just now unlisted most of the video clips that are in the re-organized lectures (exceptions were made for videos with high view counts), and the original “Physics 4A Lectures” and “Physics 4C Lectures” now only contain problem-solving videos and other lecture videos that didn’t make the cut, in these re-named playlists: Physics 4A Extra Lectures and Problem-Solving Videos and Physics 4C Extra Lectures and Problem-Solving Videos.
Finally, there is an added benefit that I didn’t anticipate and only started seeing once I was holding the required 1-on-1 meetings, particularly as I was completing the “due diligence check”: I now have metadata on which students are likely watching the lecture videos and who are not watching the lecture videos. Before, outside of information I get from the required 1-on-1 meeting (where I can easily tell who knows material covered in my lectures and who does not), I could only make rough guesses on whether students watch lecture videos, for example, by looking at the amount of time students spend on the course Canvas site (as reported in People tab). But the information source here was not very detailed nor very reliable (because it’s easy for a student to leave a web browser page open long enough on Canvas site to just accumulate the hours). The video-cued assessments being MyOpenMath assessment meant that these assessments come with the same detailed timestamp metadata that all MyOpenMath assessments come with. I can get actual time measurements for how long a student spent on a video-cued assessment and accurately categorize each assessment into one of three categories: (1) student likely watched the lecture video (amount of time spent is between 50% of video length and slightly longer than video length; 50% is easily explained by a student watching the video at 2x speed); (2) student likely did not watch the lecture video (amount of time spent is a hilariously low amount of time, like single-digit number of minutes); (3) unsure (mostly when thousands of minutes are recorded, usually resulting from the assessment being accessed over multiple days; for individual lecture assignments this probably isn’t indicative of anything bad, but I’ve come to believe that when all the lecture assignment timestamps look like this, it’s more likely than not that a student is not watching the lecture videos in good faith and are simply attempting to create normal-looking timestamps). On more than one occasion during the 1-on-1 meeting, this information came in useful, because when a student is confronted with (1) lack of their own knowledge of what is covered in lecture and (2) MyOpenMath timestamps showing how little time they spent on lecture assignments, even the most deceitful student has difficult time maintaining the lie (also, for what reason? I emphasize grade is determined on whether or not they can solve physics problems, not on whether they watch the lecture videos).
I’m currently mulling over a first-day presentation (to be used for Physics 4A in Fall 2025 and possibly for Physics 4B) in which to utilize some of this capability, where the main purpose of the presentation is to convince students (1) that it’s worth putting in the effort to learn how to solve physics problems and (2) that there is no other way to get a B or an A, because all the ways of cutting corners that may have worked in other online classes won’t work in this class.
P.S. Oh, and one last thing: YouTube watch statistics are better for these re-organized videos. CoA Physics channel’s watch hours are up by more than 35% in comparison with previous 365 days (I believe a majority of these hours are coming from people—members of the public—who just find these videos). While I’m primarily making the lecture videos for my students, but, hey, when the public watch hours exceed 3,000 hours (over 365 days), I can turn on monetization for the channel and see how that does.
I will write a longer post that relate to these “Physics … in 30 Lectures” series, but in the meantime, I did just finish Physics 4C in 30 Lectures, which is available at the link. For a while I thought I might tack on a couple additional lectures (especially one covering parity violation), but I think this is a good ending point for the semester, without additional “optional” lectures.
I’ll have more updates that relate to Physics 4C (which is on track to being completed, work in progress available here, Physics 4C in 30 Lectures) and my reflections on how well this worked.
This past fall semester, I tried something different with the virtual class session videos, processing them during the semester week-by-week. It did work out pretty well, as a task to be done over the weekend (and the video uploaded early next week). They are in the usual playlists, for now at the end of the list (to be moved into the right place when I update the Canvas shells in a few weeks): Physics 4A playlist and Physics 4B playlist.
Now, for something completely new: towards the end of the last semester, I thought to take on a project that addresses a couple chronic issues when it comes to lecture videos. The first is a feedback I have gotten a few times regarding the length and the organization of the videos, boiling down to some students wishing that they were put into one long video instead of broken into clips the way they are now. The 10 to 20-minute video clip made sense before YouTube videos had chapters, but, well, now they have chapters, so an hour to two-hour long video can be pretty navigable (and allow students to watch a whole set of lectures through without interruption).
The second is the more serious issue: very few students watch the lecture videos. Which is challenging and regrettable, because the feedback I get from the few students who do put in the time to watch the lecture videos (and I don’t mean people who just skip to the homework help videos) is that they do learn from the videos. For vast majority of students who end the class with C (because they didn’t learn to apply any physics problem-solving techniques), what I see has been that they barely put in any time into watching the lecture videos. Some never started watching; others started watching but found one or two minor issue with one early video as an excuse to not watch the rest of the videos.
The homework system I use, MyOpenMath, does have an assignment setup that might address this issue, called “video-cued assessment,” and the project I took on is to collect the lecture videos into sets of about 1 hour in length each (some of them maybe up to 1.5 hours; never longer than 2 hours), both to standardize their presentation (and re-edit some clips that need better color balancing, etc.) and to use as basis for the video-cued assessment.
In terms of effectiveness of this approach (will more students watch lecture videos?) I’ll need to see in Spring 2025. In the meantime, the videos I am putting together (I’ve done 10 of 30 so far and have uploaded 4 of them; the rest are on daily—workdaily—release schedule) are in this playlist: Physics 4A in 30 Lectures.
Edited lecture videos for the most recently completed classes (Physics 4A and 4C) have been uploaded to our YouTube channel, in the usual playlists: Physics 4A playlist and Physics 4C playlist. Also, the backlog of Physics 4B videos from Fall 2023 semester has been processed (and in Physics 4B playlist).
During the last semester, it was feasible to process these virtual class session videos during the semester (I kinda had to in Spring 2024, as I had two sections of Physics 4A and I couldn’t hold a separate set of virtual class sessions for the later section of Physics 4A), so I think I’ll try doing that this semester again. It’ll help lighten the prep work between semesters.
Finally, I recorded a lecture I’ve been meaning to for a while: Physics 4B – Intro to Gauss’s Law and Electric Flux. It replaces the one set of in-person lectures that I’ve never been all that happy with. I have even tried re-editing it before (Physics 4B – Introduction to Gauss’s Law, ver 2), but at some point, the only way to fix it properly was to re-record it from scratch. It took a long time writing out the full script (I usually just wing it live) and then trying to actually follow that script for the recording (but, you know, not actually learn the lines, because I’m not an actor), but, well, I think it’s definitely better than the other lecture. I’ll have to see what the reception is like.
Edited lecture videos for Physics 4A (Fall 2023) have been uploaded to our YouTube channel, and as usual, they are available in the playlist: Physics 4A playlist.
P.S. I have the Physics 4B videos from Fall 2023 semester, but I won’t be able to get to editing and processing them until Summer.
Edited lecture videos for Physics 4A (Spring 2023), Physics 4B (Fall 2022), and Physics 4C (Spring 2023) have been uploaded to our YouTube channel.
As usual, the most recent orientation session video has replaced the older orientation session video, at the beginning of the playlist. The newly edited videos are at the end of the playlist, and as I work through updating next semester’s course sites on Canvas LMS (Physics 4A and 4B for Fall 2023), I will place these videos into proper place.
Oh, and one new playlist I’m creating (mostly for material that I imagine I need to do on a semesterly basis, because things change between semesters): Physics 4 Conceptual Questions.
And as usual, the regular playlists are organized by courses:
This is something to fill in the gap in OpenStax coverage I found last semester, motion graphs. Other textbooks I have used before have a section or a subsection dedicated to this topic, which seems to be missing on OpenStax University Physics Volume 1.
So, this is the note written to cover the gap (may need to revise a little in the future, after lecturing on additional examples).
I finished processing recorded virtual class sessions from Spring 2022 (for Physics 4A and Physics 4C) and posted them on our YouTube channel (https://youtube.com/c/CoAPhysics). As usual, they are added at the end of the existing playlists (Physics 4A playlist and Physics 4C playlist) and moved around as I prepare for the new semester (the videos occur roughly in the order they appear in the course LMS site).
Oh, I guess this means the Physics 4C playlist will remain in that order for some time, as I am working on Physics 4A and 4B material for the upcoming semester.
