I run the MIT+K12 Videos Program in MIT's Office of Digital Learning. I mostly write and think about work (educational media, digital learning, how to make them better, etc.), though an occasional picture of my brother might slip in every now and then. @echoetweets
'Believing in evolution is just a description about you. It’s not an account of how you reason.'
Why Do So Many Reasonable People Doubt Science? National Geographic 2015
Several months ago, I heard Dan Kahan of Yale's law school give a talk at a science media symposium on why, despite established scientific consensus about climate change and lots of media like Years of Living Dangerously and An Inconvenient Truth, public opinion about climate change basically hasn't changed since the 80s. Probably one of my favorite conference talks ever (science, media, or education). The gist is that people's scientific beliefs are not a reflection of how they reason (or of their scientific literacy) but of who they are and to change those beliefs means doing something besides appealing to reason. That when you throw facts at them you're not attacking intelligence, but that you're attacking their community and who they are as a person. Effective science communication means meeting people in their worldview. Anti-vaxxers are going to respond much better to someone who is in a New Age community who says, look, I understand what it's like to not want to vaccinate your kids, and I am one of you, and I love who you are and I can reconcile that with actually vaccinating my kids. In general, it requires showing more empathy (even though the John Stewart method of public humiliation and "anti"-science-shaming is much more fun to witness) and recognizing the legitimacy of where people's beliefs come from, regardless of how ludicrous they might sound.
In other news, Pope Francis is supposed to publish an encyclical on climate change very soon. It always baffled me that many practitioners of western religion haven't embraced "environmentalism" more - not only does it seem reconcilable, but actually fundamental to the faith.
Love this profile of the Jack and Kate Spade from several years ago after they'd sold the Kate Spade brand. (Surprise! Kate Spade doesn't actually own Kate Spade anymore, nor does it have creative ties to the company.) This sentiment, expressed by Jack, is resonating a lot with me these days regarding digital education. Despite [allegedly] arising out of a desire to escape the factory model of education, so many digital education products all seem to be made by the same factory. They all sort of share the same corporate sheen. I think part of it is that there are so many content distributors out there (video aggregators, educational "social networks," etc.) but so few content creators (and they all pretty much exist on YouTube) - that's certainly a big thing that frustrates me about ed-tech in general. Perhaps related is the lack of a push for this idea of "acting smaller" or personal.
I once heard one of the designers of the Olympic velodrome in London talk about the idea of craftsmanship and the handmade being undervalued in design. The context of his discussion was that people would remember a hand-drawn figure in a presentation much better than some stock image or shiny graph. It's probably one of the things I think about most in branding MIT+K12 Videos - from the animations I do for Science Out Loud, to the new website re-design that's coming, to this blog (and the background and logo and the very idea of it). Granted, I hate things hand-drawn for the sake of the style (please, for the love of all that is creative, let the whiteboard animation shtick die already unless you have a legitimate reason to do it). What hand-drawn and "homemade" really mean is creating or conveying a novel identity that's authentically yours (you can still have influences, of course), and maintaining the mom-and-pop mentality even if you're trying to scale up. Of course, it's way easier said than done. But I wonder how many folks in ed-tech even see this as an issue?
You have content experts (or at least content enthusiasts). You have a medium that reaches a high volume of audiences and also has a very low barrier to entry. Add a little competition and enticing prizes, and your outcome should be some awesome, crowdsourced videos that showcase science, technology, engineering, and math in a compelling, authentic way. Right?
(Winner of the National Academy of Engineering's Global Grand Challenges Video Contest, 2013)
(via) The problem with this video, and so many winners of similar video contests, is that it doesn't actually accomplish what the ultimate goal of contests like this is: to inspire an interest in science/tech/engineering/math among an audience who might not be interested. (I presume.) The video creator is already an engineering major in college, and it's hard to imagine kids really being converted one way or another about STEM after watching this video. Though it does have many good qualities, I'd argue that its failure to reach potential was largely due to how the contest itself was set up.
A recipe for less-than-optimal contest results
Incredibly broad visions "1-2 minute video showing engineering’s impact on society in the last 50 years and to project its contributions in the next 50 years" (similar traps: "Make science cool")
Pulling off expected product requires participants to adopt an unreasonably large and new skill set ("A 13-year-old can go viral on YouTube - should be pretty easy for someone who understands rocket science."). 2b) Little support or guidance on product expectations/how to achieve results ("We'll let them come up with something... and that something is undefined!").
No one judging who is familiar with or good at making the product trying to be achieved.
Viewing video as a free ticket to engagement and outreach (and being "hip and cool").
Incredibly broad vision is something a lot of folks (including myself and K12 Videos, very much so) are suckers for. MIT's SciEx Competition, a video contest that called for "extreme science videos" that made science cool, suffered from this during its first iteration last year, which led to a very mixed bag of entries that were, for the most part, not... uh (well, I'll say it)... very good (though some were very, very close). Now, to their credit, the organizers are very aware of this and were extremely keen on revamping the vision to get entries that better matched their overall goal, which is to inspire more kids to consider learning about STEM fields. More on that at the end.
Another issue with broad vision and expectations is that it often leads to vague, grandiose, sentimental entries that don't really have much substance but also don't have the production value to pull of inspiration. That's the issue with this $25,000 winning video (and pretty much every winning video) for the National Academy of Engineering:
Here is inspirational pulled off well (IMHO):
This GE ad (and pretty much all of GE's videos) is awesome - not necessarily more substantive content-wise, but certainly aesthetic-wise. The only catch? It's made by professionals. (Because that's the job of a professional. To make professional videos.)
And this points to a greater problem of a lack of digital media literacy - understanding what goes into making good digital media products and the self-awareness to know what you can and can't pull off - among technical communities. I say this as someone who was trained as an engineer and then learned how to produce media - production is hard and requires a different mindset to do well. That doesn't mean that people should be discouraged from doing inspirational videos - what it's really about is identifying what is feasible to best leverage content knowledge and excitement.
You don't have to make science cool by trying to copy "extreme" or super high production-value videos - you just need to... not get in the way of it already being cool. (Unless you are good at making those kinds of videos, then by all means - do an extreme science video. The point is, there's lower-hanging fruit that can be just as compelling.)
If your intended participant pool has to pick up some intense skills (pre-production, scripting, shooting, post-production) in order to marginally achieve the goal, rethink the goal, unless you can really, really equip people with the skills necessary to pull that off.
Two examples of what I deem "successes:"
Harvard Medical School's Organ Challenge
Why it works:
Tight, simple, clear call-to-action - make a music video/song about the human body. As a result, these videos also succeeded in making physiology fun.
Call-to-action is totally attainable! Doesn't matter if you don't really know how to film or edit - all of the winners achieve the call-to-action completely.
Regardless of the product outcomes, the participants probably had a lot of fun and got to learn something new in the process of making these videos. In general, that's why I don't have as much beef with video contests aimed for younger participants.
The organizers themselves are "experts" in the product - this was organized by the HMS students who made the "What Does The Spleen Say" video. They knew what to look for and had clear expectations for participants. (This is actually a super salient point - you see a lot of engineers judging these contests, or documentary-makers, but no one who's made entertaining, educational video. It's partially because there just aren't a lot of people in the edu-tain-spiration field in the first place. )
GE's 6 Second Science Fair
It's not really a contest, and it's arguably not really video (?), and the videos are demos which isn't exactly science... BUT:
Very tight format and expectation.
Product goal is very easily attainable (although even within Vine, there's a spectrum of quality).
It got people experimenting and being curious about science, and gained some media traction (and subsequently more engagement).
As far as the whole issue of judging goes (the idea of making things a social media contest of 'Likes' could constitute its own post), I'll leave you with this co-winner of Change the Equation's S.T.E.M. is Cool Video Contest from 2010. This was a contest to get STEM corporations to make videos to get kids interested in STEM careers.
I'm trying to be as civil and understanding as I can be here, but this. THIS. THIS WAS MADE IN 2010. BY A COMPANY. TO GET KIDS TO THINK STEM IS COOL. COOL. AND THEN OTHER ADULTS THOUGHT IT WAS COOL.
(via) (As a side note, I find the way it handles/simplifies the gender disparity and lack of diversity in STEM incredibly problematic. This video evokes a lot of mixed feelings from me, ranging from deep sadness to incapacitation from laughing to utter denial that such a heinous and out-of-touch thing could exist.)
Video is not a free ticket to engagement. Video itself is not inherently cool, unless we're living in 1890s Germany where moving pictures are being made for the first time. Then, I guess the novelty of a moving picture is inherently cool. But there are lots of ways to get people fascinated about science (look at the popularity of IFL Science on Facebook, science gifs, infographics). As Ben Wyatt from Parks and Rec said,
(via)
Back to the SciEx Competition. The whole point of this hubbub is that K12 Videos is sponsoring SciEx this year. The reasons why are:
The organizers had clear and genuine motivation. They're two grad students at MIT who study computer science, and they love what they do and want to share that with kids. No PR gimmick, no engagement quota. No one's making them do this. They're pretty awesome.
They came fully aware of the issues from last year and wanted to figure out how to improve things and were willing to completely rethink the contest. So we had several discussions about it, and came to a conclusion. Granted, this conclusion came pretty late (after they'd done their first round of advertisement and workshop sessions), but I'm leading a workshop tonight and hoping that the revised message gets through to at least some people. The key conclusion involved making much clearer entry expectations (make the videos much shorter, stop advertising it as an "extreme science video" competition, but one to make science go viral, having clearer examples, etc.).
I found the whole meetup idea (where they bring together film students, engineering students, musicians, to form teams) interesting and a step in the right direction.
Last year's videos got pretty close to achieving their last year's goal (second place winner was the best one, IMHO, but needed to be way, way, WAY shorter).
It's not going to be perfect, but I do think it'll be better than what it was before, and will be a useful case study itself.
Ever since reading The Viral Video Manifesto (I was very, very skeptical given the title and cover design. Yes, I literally judged the books by its cover. It ended up being a worthwhile read.), I've been very mindful of the habit of projecting the TV aesthetic onto web videos. B-roll, or the filler-footage that you sometimes see in flashbacks and interviews (shot of the subject at work while their voice is in the background), is a "very TV" thing that folks often bring over into web video, thinking it's evidence of high-quality production. I've been much more hesitant about using b-roll (or at least try to be much more intentional about it) in Science Out Loud now. The tl;dr version of VVM is that virality in web videos (which is often the proxy for value/success/what is good, for better or for worse) hinges on authenticity, and that things that are "very TV" (like b-roll) and considered good in traditional production may end up getting in the way of a "good" web production.
One aspect of slick production that, I think, still carries a great deal of relevancy in web video is camera movement. Specifically, movement that advances the illusion that the viewer is in the world of the video, even if that video was made with really nice equipment. Movement that immerses viewers in the video. The type of movement that I favor in SOL are Go-Pro or iPhone-like shots - making the camera act like the eyes of someone who might be standing there and watching the video happen in real-life.
Admittedly, the main reason why we tried them out in the first place was to add visual interest to a scene that was starting to get a little boring because she was just standing there delivering lines. So, it's a little gratuitous.
(My favorite shot of SOL so far. Also would be slightly creepy if this was actually someone's eye movement looking at A.*)
Now, to achieve eye-level dolly shots that mimic the eye-trace of a bystander of the video (or just cool moves in general), some *actual* production outfits use big, expensive equipment.
(Not in our budget.)
There are also lots of DIY dolly tutorials out there, but you have to pay up in cold, hard free time. We don't have much of that around here, either.
Option 1: Human Dolly
I like how this followed the beam of light, which isn't exactly how a bystander would see it (we tried a wide-angle at first, but decided against it for the reasons below). However, there is still something immersive about the eye-level POV, and the pacing of the reveal of the green cloak matched up nicely with the narration to make it clearer. Setup:
(We eventually abandoned the ladder idea because a) it was somewhat unsafe and I thought he was going to fall off at any point and b) because we needed to get more on eye-level. Same movement, though. I've been wanting to try this DIY stabilizer for a while now. I think it would've made this scene a lot easier to shoot.)
Again, this movement matched up well with the narration (talking about shining a laser through diffraction slits, eventually revealing how light diffracts). Setup:
(Very, very tight setup. Had to physically guide camera op in the dark so he wouldn't run into anything.)
Pros: Cheap, can work in tight spaces.
Cons: Physically very, very tiring (and hard to get consistent shots once your arm gets tired) (also hard to get consistent shots using your arm) (also hard to get decent shots, in general).
Option 2: Use what is available at that moment in time (i.e. handcart + human + tripod. Lots of tape, optional.)
(Kind of reminds me of those POV shots of kids walking up to a huge door and looking up in the movies.)
Using a fisheye lens here would've also been cool. An instance where high production value (i.e. looks less n00b-like than if you shot it with your webcam) doesn't necessarily make it look less authentic (and as a consequence, less good). You still feel immersed in the world of D-Lab.
I don't have documentation of the setup here, since I was sitting in a handcart holding a tripod (whose neck had been flipped upside-down to make it shorter) steady as the director pushed me around to film this.
We couldn't really film in this clean-room (bad sound, having to sterilize all the equipment, limited space to move, etc.), so we ended up shooting in the adjacent hallway that looks into it. Setup:
(Similar to that of the D-Lab sequences).
Setup:
Sometimes we change things up and tape the tripod to the FRONT of the hand cart. Are we the Massachusetts Institute of Technology, or are we the Massachusetts Institute of Technology??
Point is, even though you can achieve authenticity in many different ways, there are variations in execution (and I suppose we could be thinking about changing ours up). I suppose we were lucky enough to even have a handcart at all. The slickness of whatever moves/production elements we include is definitely something I've been thinking about a lot lately, though.
*By the way, the way we did the slow pan up in that plants gif at the beginning was by creating a lever system out of the tripod. Took off the head and used the base of the tripod as the fulcrum, then used a long boom pole as a lever. Fixed the camera on one end - with a lot of clamps - and hung the camera shoulder rig on the other. Adjusted r1 and r2 so that the torques from the camera and shoulder rig juuuust equaled each other (creating zero angular acceleration) and gently touched the system into movement so that the camera would look like it was panning up at constant speed. Full credit to George for devising the setup.
The easiest way to make a [good] video on a technical topic (or why making videos on technical topics is hard)
A few weeks ago, the crew and I shot a video about semiconductors. It was, by far, the smoothest and easiest video to shoot (maybe out of all the videos I have ever made, Science Out Loud or otherwise).
Ironically, we thought it'd be one of the trickiest to shoot, given the logistics of the demo...
...And also the fact that we had about three locations with very different native lighting/audio situations.
(Server rooms, while visually appropriate for a segment on how semiconductors are used in modern electronics, are loud and have fluorescent lighting that gives the following effect...)
But all-in-all, start-to-finish, we wrapped in a little under 12 hours (which is on the faster end for a 5-min. episode of Science Out Loud). Crew call was 7:30 AM, but we didn't even start rolling camera until close to 11 AM (Jamie met us at 8 AM, but we spent some time hunting down shorter alligator clips to tidy up the circuit he'd built/make it look clearer on camera.) We were also a bit at the mercy of the availabilities of the managers of the server room and clean room spaces (and MIT was closing early that day due to a snow storm). So, despite the logistical hurdles, why did this one end up seeming so [relatively] effortless?
Long story short (or shorter than it could be), this video worked for the following reasons (in increasing order of importance):
This was Jamie's second time making a Science Out Loud video - he was mentally prepared for how physically-demanding the day would be, and I think he rationed his energy accordingly. (I'm making this sound like a much more trivial thing than it really is. Everyone, including myself, underestimates how hard hosting is, and how much the skill of hosting lies in what you do both when the camera is rolling and when it isn't.) In general, he was prepared - he'd tested out the demo beforehand, was comfortable with all his lines. These things may sound obvious, but I'd venture to say that an overwhelming majority (~80%?) of folks hosting things like MITx course videos don't do this (mostly because it takes up an enormous amount of time that they don't have - but that is a post for another day).
Jamie knew the content inside-and-out (he's graduating with a Ph.D. this summer and his research directly relates to semiconductors). Whenever we'd realize that certain lines weren't working in the context of the props or movement (things we couldn't predict just reading scripts), we could easily ask clarifying questions to help us restructure or edit on-the-fly. Because of his "expert status," he was connected - he had pretty easy access to all the facilities because he works in them, and knew all the space managers (so we had much more freedom in what we could try shooting and where). He also advertised his previous video on comp sci blogs, which landed a Reddit embed and made his previous episode the most-watched SOL one to-date. And his videos are partially footed by a portion of his lab's NSF grant set aside for outreach projects.
He had a great, committed attitude the entire process. From the beginning, he was totally on-board with the outreach aspect of it (his wife is an elementary school teacher... I think that helped him during the scripting process, actually), was up for approaching material in a new way. He was one of the most eager for feedback and was able to efficiently incorporate our feedback on scripts (mostly because I think he was excited to do it). Actually, the first thing he said to us on shoot day was something along the lines of, "I noticed that in my last video, there were moments where I didn't sound as happy to be there as I wanted to be. So I want to make sure I sound more enthusiastic or more accurately reflect how I feel about his material for this video."
Preparation: Preparation takes time, the most in-demand commodity at MIT (and in a lot of other places, academic and otherwise). It also takes much more work for some people to be good on camera than others. (I know I taught a class titled "Becoming the Next Bill Nye," but I am convinced that not everyone can become the next Bill Nye. Post for another time.) I'm not entirely sure how to tackle this one, besides hiring semi-informed "talent," the way hosts are hired in commercial productions. We actually played around with this idea - bringing on a separate host unrelated to the script developer - for this season (which led us to cast this individual in an entirely new series). But MIT's DIY culture tends to not be as receptive to this idea, the idea that it's not always possible to hack your way through everything - including being a great, live science communicator. Occasionally, folks like Jamie (and several of our other students) come by, jumping head-first into this little experiment of ours. But can we build a pipeline that doesn't fundamentally rely on this (because I'm not sure these people are as common and many might think)?
Expertise: In general, this is the thing that MIT has going for it that shows like SciShow, Veritasium, Vsauce, etc. don't. It's also not leveraged as much as I would like for it to be. At this point, as far as content programming goes, I try to focus on making videos that pretty much only MIT can make, cognizant that preparation and attitude are the qualities that MIT struggles with regarding science/educational videos (and that those channels don't as much). That means we don't make Khan Academy-style videos anymore. That means we don't make tutorial videos (because we don't have the time, we're actually not super great at it). That means we don't make explainer videos on basic physics concepts unless we involve something unique and contextual like the MIT Skydiving Club. Expertise is the self-awareness of knowing what you know and how to leverage it well. Harder than it sounds. And content expertise is actually something our students (especially undergrads, understandably) struggle with a bit (and has been a fascinating learning experience this season - more on that later). You also run into the struggle of balancing getting undergrads and younger students involved, being mindful of the near-peer aspect of what we do, with grad students (who may "know" more, but who the audience may see more as a typical instructor figure and less of a relatable peer).
Attitude: It is always difficult to critique and discuss anything that involves assuming people's intentions. I think most of the folks I've worked with on these videos come at it from an outreach-oriented attitude (and maybe a little fascination with fame, but who wouldn't?). I also know that there's a moment - maybe when we're in the third or fourth draft of the script, or maybe during the tenth take of a shot, or maybe during the 15th hour of the day - where people hit a wall. A what-did-I-get-myself-into wall. And there's a fork that diverges from that point - people start to realize and appreciate the difficulties and nuances associated with communicating science well, or people start to wonder why you're wasting their time. I don't think the latter is an intentionally malicious attitude, or even a mean one. It's very natural. But I do wonder why communication has been extrapolated out of science and been given a less-legit status as an endeavor at places like MIT. People are able to hold communication in lower regard (or maybe lower priority) because it's been designated as a separate thing in the first place. George, the director of our SOL videos, was telling me that he had a conversation with a grad student Bioengineering Communication Fellow who said he was participating in that program because he didn't "see a delineation between science and science communication." Would being more intentional about integrating communication back into science training more seamlessly address this (something beyond "you will have to write 20-pages of material for this technical class as your communication requirement")? I've toyed with the idea of turning the class I taught into one that fulfills the communication requirement at MIT... but there are more confounding variables associated with that issue than I could expound upon here due limitations of space and your attention span. What it seems to boil down to is that everyone is on-board with the idea of communicating science well... to the extent that efforts to do so don't infringe on the "real work" of science itself. Because science and science communication are two separate things.
Lest I sound like I'm giving a cop-out conclusion, I am aware that I'm totally giving a cop-out conclusion when I say these are all unresolved phenomena whose roots are well-beyond my jurisdiction. The nice thing about K12 Videos is that I feel like it's a microcosm of the MIT and science communities, and that it's a small sandbox to tinker with approaches to these issues on a much smaller scale. Jamie did a great job on his episode. Is that scalable to delivering better and higher volume content?
I'm all for the personalized learning that homeschooling offers - my homeschooled friends growing up had exceptionally well-developed talents because they had the time/learning space to pursue the things they really intellectually thirsted for, be it a music, coding, art... But OOF, some of these commenters are mega-proselytizing the Church of Digital Learning/traditional education-shaming.
The way I see it, things like Khan Academy in a classroom context should do more than just "free up time for one-on-one mentoring" - in all this personalized learning talk, shouldn't it also free up more time? Time to work on projects as a class? More "unconventional" learning experiences? I had a group of 6th graders from a nearby middle school come and visit MIT and I got to show them around the Personal Robotics Group at the Media Lab, the Koch Institute for Cancer Research, and then they got to mentor a bunch of MIT students on science communication. I'm pretty sure it was only possible because they were part of a STEM enrichment class (that not every public school has) and had a principal and teacher who were very enthusiastic about taking on this kind of thing (and being okay with having the students miss some of their other classes to come). </horn toot> Those previously-mentioned homeschooled buddies of mine also had parents who made sure they took part in lots of communities with other students (sports, extracurriculars, etc.). Point is, I cringe thinking about visions of an "alt school" where kids get bused over only to sit at a computer for the majority of the day and where their interactions with each other only occur over recess. 1. Lord knows self-assembling socio-economic stratifications already occur enough on their own in middle school - can't imagine what purely personalized learning would do to exacerbate this. 2. Why does everyone want to hide the classroom landscape of learning diversity? Heaven forbid having a kid who learns faster having to wait for a slower learner. I had a classmate in fifth grade who was a painfully slow reader, and also one of the coolest kids who totally embraced his dyslexia and legitimately made it okay for everyone else to be okay with themselves. I also had an amazing fifth grade teacher who established the classroom environment to facilitate that happening. Public school in the middle of Missouri. (Okay, yeah, maybe that was a rare experience.) Point is, it was somewhat formative in developing empathy and self-confidence at a pretty critical age, an experience that I probably wouldn't have had at Khan School For Kids Who Want To Read Good But Don't Want Anyone Else To Know They Can't Read Good. (Although, to be real, I probably wouldn't have had that experience at any other school on any other planet, because Mrs. Winchester's Rock Bridge Elementary fifth grade class was unreal.)
tl;dr There is more to school than academic learning. Working with others builds character, etc.
Jyri Engestrom, Caterina Fake’s partner, signed up with AltSchool this year. The couple had been homeschooling for a couple of years, an experiment that gradually expanded into a 10-student “microschool” called Sesat School. This year, his students started attending AltSchool part-time, in what he calls a “hybrid” approach. He says it’s just one example of how a new crop of startups could use technology to create new educational models, somewhere between homeschooling and traditional school. He foresees a day when the same forces that have upended everything from the entertainment industry to transportation wreak havoc on our current model of education, when you can hire a teacher by the hour, just as you would hire a TaskRabbit to assemble your Ikea furniture.
