> Let us also remember, she actually built AND shipped.
It's funny how much further this is than the side projects of so many here (myself included)
For me this is one of those 10,000 things. We can either dismiss it, or share in the joy of learning something new and sharing our own lessons. Teachers learning OBS will be as much of a game changer as teachers moving from overheads to powerpoint, I think.
Also, I feel like this is super indicative of the gate-keeping mentality a lot of IT and tech people have around "not-cutting-edge" technology. Even if it is nothing new, let's be the resource
Talking with non-engineers (particularly friends of mine who work in non-profits), I see incomprehensible amounts of potential value in just increasing the uptake of "not-so-special" tech among people who consider themselves "non-technical". That's the point of technology, after all: not being fancy unto itself, but actually doing stuff that people need, whether that's wheels enabling long-distance transportation or lightboards enabling better long-distance education.
There's too many office people who have no interest / refuse to learn any new skills. Many can't cook to save their lives. I doubt 0.1% of them can solder, 0.03% could troubleshoot something electrical, and I'm doubtful of their ability to handle and finish drywall.
It's very sad when people think everything else is beyond their reach or they don't even look to see what could be there.
> Teachers learning OBS will be as much of a game changer as teachers moving from overheads to powerpoint, I think.
Just an anecdote but as a chemistry student I remembered lessons taught from an overhead far better than those from powerpoint. There was a stark contrast between the lower division organic chemistry series whose instructors used overheads, and the upper division organic chemistry mechanisms class whose teacher used slides. Those slides were mostly bullets and diagrams, sometimes with crude animations which were meant to indicate nucleophilic attack or whatever. Just watching the lecturer draw diagrams and arrows and listening to them talk was far more helpful, for me.
The lightbox seems much more overhead like in presentation. I dig it, and I hope today's college students learn well from it.
I actually hope more teachers go for the simple overhead option. I do, using an overhead webcam.
A physical overhead with the instructor's hand visible has been shown to work far better than presenting ready-made diagrams. It also works better than digital drawing, probably due to the natural inclusion of pointing gestures guiding students' attention [1].
There is also evidence (and good theoretical reasons to believe) that a transparent lightboard leads to unnecessary distraction/split attention. [2]
>Also, I feel like this is super indicative of the gate-keeping mentality a lot of IT and tech
That lumps people together. I'm in IT/tech but don't value intricate, difficult, extreme novelty over everything else. That would be absurd.
Remember Stanfords $1 paper microscope for looking at blood samples? Now that was badass.
The cleverness is in the saving of $7940 to solve a problem. Dumpster-diving for parts and materials, or even freeganning Whole Foods are all perfectly valid too.
I remember a guy used his seat aisle's window on StartupBus as a whiteboard with dry-erase markers, but I guess he should've brought an LED light strip to make it pop more. The trick is installing it. ;-)
And it's made all the more important by the lack of software allowing economists to graphically demonstrate economic concepts on the fly. The best digital substitute is probably a graphic design program (like skitch or gimp), and most economists would not be competent graphic designers, at least not without practice.
OBS is really the big enabler in (my) online teaching at the moment. I use an IPad with a drawing app, I draw on a predefined colored canvas and overlay it in OBS with a chromakey filter. If that is available to you, it is even cheaper. But then again, I am looking slightly down while writing..
Just the other day I was following some complicated tutorial for getting blackmagic's atem mini pro hdmi capture/mixer board to work in Wine in Ubuntu 20.04 .. I finally gave up and tried decided to play around with a webcam in obs instead. There to my surprise, OBS was listing the ATEM's capture input out of the box no wine required, with no config as a video source!
Same! I don't do anything as elaborate as the article, but I use a combination of OBS, PowerPoint, GIMP, IDEs, and a cheap Wacom tablet to demonstrate all the concepts for my classes.
The "Khan Academy"-style drawing I think really helps keep learning online casual and conversational in a way just presenting slides doesn't.
Yes! Educational research supports being casual and conversational. Although for drawing diagrams, a physical overhead with the instructor's hand visible seems to work far better than Khan style video.
See: Fiorella, L., & Mayer, R. E. (2016). Effects of observing the instructor draw diagrams on learning from multimedia messages. Journal of Educational Psychology, 108(4), 528.
Were you using Zoom? I had a lot of trouble using either the OBS virtual camera or sharing the composition window---what was transmitted was either potato quality or the video lagged very far behind.
After a LOT of debugging, I'm convinced it was due to Zoom bandwidth throttling, which I was unable to get waived.
Yes. Zoom is fine, you can either up the resolution for the "speaker view" in your account preferences (maybe just in the paid version), or even better: Use the screen share feature and under "advanced" you can share a second camera, or in this case: the virtual OBS camera. This will usually not be compressed as hard as the speaker views.
In some cases it looked fine for me but for the other persons it looked really ugly. Make sure you use the v"2nd camera mode" in screen sharing options of zoom. Do not do it as your own camera or it will look really bad. On Linux, I have to start zoom after Obs and the virtual camera for all that to work properly.
FYI: This is a good article about a DIY hack that can be made at low cost and with minimal effort. If you want extremely detailed information about the state of the art in this space, it's here:
It's a fantastic ingenious hack for, IMO, a model that needs to die.
There is no reason to reteach the same class every semester . The best minds need to get together and put in N (say 10) semesters worth of effort into creating the highest production value lectures that all can use until the material fundamentally changes. Most of these kinds of lectures have little to no interaction which can come at other times and the need to "do it live" leads to subpar teaching production.
Back in about 1988, I took a required history class from a local community college. (To help make up for the fact that it took me four semesters to get through the two semester calculus course.) The course consisted of...
1. A high-production-value TV program transmitted on the local PBS channel.
2. In-person mid-term and final tests.
3. "Teaching assistants" (it was a local community college; I'm not sure what their exact status was) with office hours for questions.
It...worked. I learned the material, such as it was. As such, it was somewhat better than the worst in-person classes I experienced.
On the other hand, there was no engagement with the material. You read the textbook, watched the TV, and went in to take the tests. End of story. There were no asides---having a teacher take part of a class to look at some current event from the viewpoint of the class, having the teacher answer a completely off-the-wall question that leads to a good discussion. There were no discussions, in fact. I'm sure some of the students who knew the material got together to review for the test or something, but honestly, I don't see why. The whole experience was aimed at the lowest common denominator with no way for the teacher to dial it up for some or all students.
It worked, but it was a worse learning experience than many 500 students classes that I had from enthusiastic, engaged teachers.
I suspect that's why the model hasn't caught on, at least at the level of decent schools---people have been saying (and attempting) that video would take over from lectures literally since Edison invented the movie. (I'm sure it's more appropriate for the degree-mills of the world, though.)
Video _should_ take over from lectures, but there needs to be a model to properly replace it. On my part, I think it's an utter waste of the teacher's time to keep repeating the same lecture over and over, when he/she should be doing exactly the conversation and engagement part. It's just we don't have a good, universal model for this. I'm sure lots of places tried and succeeded, but for some reason no particular way of skinning this cat got out to become well known.
Partly, I suspect, is because 80% of the value of education is signaling, so any "revolution in teaching" would be just optimizing the 20% left.
First link is paywalled, and the second is rather unconvincing, and pales in comparison to Caplan's overly careful argumentation.
I suspect we have different lufe experiences as well. For me most education years were either a waste of time, or at least came with very high opportunity costs.
Don't think intro/extroversion plays much of a role here, I'm probably an extroverted autodidact and that's still my ideal model for learning. What I enjoy doing in my free time is completely separate from that. Sure, in an ideal world I'd have in-person lectures that are as high production value as the recorded lecture(/TV) in question, but 99.9% of the time I won't be getting those.
Only those conferences work requires me to go to. I don’t get much out of them, to be honest. I never attend meetups. (I did a few times in the past, they were worse than useless to me.)
I’ve had beers with other devs in my field, and very productive cross-industry meetings. But general agenda-less meetups are not useful, no.
> the need to "do it live" leads to subpar teaching production
That heavily depends on the subject and the students. The goal of teaching is to teach, not to achieve a high production value of a performance in the Hollywood sense.
What really needs to die is lectures themselves. They are a TERRIBLE method of creating learning. However they are very good at creating the perception of learning. That contradiction is not surprising, it speaks to how we actually learn, by engaging with our reflection and metacognition about attempts at things that end in errors.
Eh...peer learning is nearly as effective, especially when the choice is between a peer and a teacher who holds beliefs about education based in a transmission model of education
I highly agree with you. I remember back in my university days (now long ago). We'd be in chemistry class trying to copy from the professor's writing on the board, and he went so fast that he'd be erasing while you were still writing.
Spending that time on trying to actually understand (from the student's point) and on explaining more/better instead of writing (from the teacher's point) would have been immensely more useful.
Huh it looks like she horizontally mirrors the camera so that the writing appears forwards not backwards. At first I thought she might have been writing in reverse, which would be doable on simple graphs but harder with more words. Later shots show her writing on-camera, forwards from her perspective.
In the past century, glass chalkboards and reverse writing were actually used in the military, mainly in air defense. Specially trained writers were putting objects scanned by radars on the board with coordinates and related info, while officers were reading them from the other side.
As a left handed person, it’s pretty easy for me to write backwards. It feels like something in my brain wants the letters and words to go that way, even though I never practice.
A right handed person can write in reverse with their left hand without much practice, by holding a pen in each hand, pretending to write with their right and mirroring the hand movements. It's even easier if you just "write" with your finger, eg on a foggy mirror, rather than holding a pen. The scripts we use are definitely optimized for right handed writing.
For a minute I was wondering how did she teach herself to write so well in reverse (and so fast), until I realised that it's trivial to flip the output.
for example, where they say of the option "Note: You can tap the Enable video mirroring option if you want a mirror effect for your own video display (where the display of your own Zoom Room looks like a mirror)."
Yeah, I agree with you, this feature always does the same thing in everything (every video conferencing app uses it to show you yourself). The reason is that you're used to seeing yourself in the mirror, and your image would look weird to you if it weren't mirrored.
It would also look weird to everyone else if it were, so they see your normal image.
The principle behind why the writing glows (i.e., totally internally reflected light from the LEDs in the plexiglass getting scattered by the ink) is also how some of the earlier multi-touch systems worked (like the one in Jeff Han's 2006 TED talk). So, with the right software, it might be possible to convert one of these DIY lightboards to support input (maybe to change slides or something).
Perhaps use hand tracking[1] to get candidate finger-tip positions, to simplify using the FTIR touch blobs for contact detection and precision position?
The lightboard.info site has instructions for building an $8k professional setup, using tempered glass etc. It does explain the mechanism of the lightboard (frustrated total internal reflection), but otherwise doesn't seem very useful for DIYers.
Not sure why this title has "goes viral", it is nothing new.
Almost 8 years ago, Dr. Jules White explained a simpler setup(at least IMO as less software) in very detail with a Mirror:
I suppose it depends on what you're used to? From my own (and other people's) experiences, you can have a project with a few hundred site visits - get towards the top of HN, and you can have 100k visits over the next few days.
A good enough project on HN tends to get coverage elsewhere, a lot of Tweets etc etc.
What made to top 10 is the article claims the event is going viral, NOT the event itself. The exaggeration on news articles with this type of title has become a pattern.
There were rooms you could rent at my undergrad in the library that had a built in light board specifically for recording these sorts of lecture videos.
The same reason things describe themselves as One Weird Trick, or N Unbelievably Adjective Things, claim that The Internet Cannot Handle them, etc. very quickly ad nauseam.
I can't stand it, won't click them, but clearly I'm abnormal and it works, most people love it (or have a 'FOMO' on it) and click them.
Do people like lightboards? You have to be careful what you wear, how you stand, how you move etc. so that you don't cover what you have written (see the video by the inexperienced USC professor where she wears a white shirt). As a result, most of the time you can only use a small part of the screen, like here: https://youtu.be/6I48J7Lf4_Y?t=173
Bad for people who are watching your videos on a small screen (phone, old tablet,...).
Or you use low lighting and then walk behind the board, like here:
https://youtu.be/6I48J7Lf4_Y?t=266
Which is a bit annoying because then you have things (i.e. you) moving in the background all the time. Not good for visually impaired audience members and people with low-bandwidth connections.
I will keep my $50 drawing tablet, although I have to admit that it's much easier with a lightboard to point to things (on the other hand, with a drawing tablet, you can just circle the word that you want to draw the audience's attention to and then undo it when you continue).
If I may be so bold and share a relevant plug... As teachers (and parents of children affected by various versions of remote education last year), our team came up with https://sharetheboard.com
The idea is simple: our app identifies and digitizes handwritten content in real time, ignoring any obstacles in the way (like people). The result: $0 light board that you can use anywhere (e.g., with a piece of paper).
Digital annotations by viewers are coming in July, so it's as close to shared remote whiteboarding as you can get. And again, no new/dedicated hardware needed: not fancy cameras, not LED strips, not even whiteboards; just a laptop (and soon, a smartphone).
Looks like there are a lot of folks on this thread looking to solve similar problems. Would love some feedback; thanks in advance.
Neither She nor the article claims it is new. The nifty part is the DIY aspect that makes this decade old technology potentially widwly accessible for teachers. Even that isn't new, but spreading the knowledge of the tool and how to make is accessible is still highly valuable.
The “hack” is buried for some reason. She built a light board, which can show the instructor and “whiteboard” lecture in the same video frame. Reversed, so writing is forwards.
Fairly sure it's a piece of plexiglass with a LED strip around the edges. Normally the light would more or less bounce around inside the glass or escape to light up the environment. If you write on it the light hits the writing instead, giving it this cool-looking glow.
Put a piece of Plexiglas between you and the camera, put a LED strip on the edges of the Plexiglas, and mirror the outgoing video stream so writing looks the right way (unfortunately your face won't), done!
I think it’s simply a piece of glass with an LED strip along the bottom edge (and a frame). So the light shines up through the glass and illuminates the markers.
And the title is click-bait. Hint, for it to not be a click-bait, it should state what the hack is (in this case: "lightboard"), not make people have to read the damned article to find out what it is. In this case it should be something like: "USC professor's DIY lightboard for online teaching to engage students goes viral"
The article writer has a a degree in (print) journalism, do they teach this sort of crap in journalism school nowadays?
> Eric Lindberg is the associate editor for features at USC University Communications. He is a USC graduate with a degree in print journalism.
So this isn’t actually any cheaper than what I would think of as the standard solution, which is to buy A drawing tablet (which also conveniently doesn’t take up a ton of space in your living area)
Let us remember, she's an Economic major, not an engineer. Let us also remember, she actually built AND shipped.