Week 3: Digital-Era Teaching

There’s a massive incompatibility between the realities of how much information & tooling is readily available online right now, versus the teaching approaches and methods commonly used in engineering higher ed. If I wanted to implement a website with a backend, physically construct and implement controls for a robot, integrate multiple smarthome devices to suit my needs, or just about anything related to electrical/computer/software engineering, there is next to no relevant course content for me to draw from across my CS & ECE dual engineering bachelors degrees. What little I would be using would come not from senior year ‘specialized’ courses, but from sophomore year ‘fundamental’ courses.

This discrepancy is made visible via the problems discussed in the readings from this week, but originates in a deeper mismatch in what “engineering” entails.

“Design”: What engineering culture calls actual engineering

Among a lay-person audience, a simple description of what ‘engineering’ entails would probably land somewhere near “solving problems using technology”. Engineering academia would likely claim that’s what one learns to do over the course of an engineering bachelors. However, once you are exposed to engineering education’s discussions of engineering content, especially as an engineer, it’s clear there’s two wholly distinct types of ‘problem’ being referred to: ‘idealized’ math problems completely abstracted out of any and all relevant context- social, geographical, or even physical or economic- and “real”, actual problems experienced by specific human beings that individual engineers, situated from their particular social/political/institutional context, are attempting to address. With few exceptions, engineering instruction at the university level focuses almost entirely on solving idealized, glorified math problems and simply renames “solving real problems” to “engineering design”. We then spend in the ballpark of 120 out of 128 credit hours solving ‘idealized’ math problems, and only bother with ‘real’ problems in a generally-final-year-long “design capstone” involving little actual instruction.

This fundamental misunderstanding, in my opinion, drives so, so much of the tensions described in the readings this week between old and newer teaching styles. Why cultivate curiosity and self-directed engagement if the students simply are there to learn the patterns of specific new math problems? Why should students have access to the internet- along with all its tools for ‘collaborating’ ‘cheating’- during class? How is experimentation even useful when the ultimate endeavor ‘engineering’ consists of is answering math questions?

These newer, alternative teaching practices are only alternative because the underlying need that justifies them has been deemed, a-priori, irrelevant to the course content we’re supposed to provide. When put alongside the misaligned incentives and absent teaching training of faculty/instructors/TAs at major research universities, of course we continue using the arcane, demonstrably inferior teaching practices that establish boring, soulsucking, uninformative classrooms. Why wouldn’t we?

Lectures: Specialized Tools for Specific Content

I LOVED the 4 Things Lecture is good for piece; it gives name to one struggle I’ve faced with paying attention to or absorbing anything from so, so many instructors, talks, or conferences. In naming the 4 things lectures do well, not only does it imply that if you arent doing one of those, there’s a better method than a lecture; it also implies that for any given lecture you’re writing you should be able to articulate which of the four it is at any given point in time. I have looked back at my lectures from last semester’s course, all of which I recorded live, and can see the lack of all 4 styles during those specific, least-engaging lectures.

So what sort of content are these four use cases most compelling for?

When I go to teach how to solve some idealized math problem, like finding the far-field beam pattern from a given antenna structure analytically, the overwhelming bulk of the time is spent walking through algebraic manipulations. There will be some inevitable point where I need to apply a trick, or think about the math in a clever way, where we briefly use modeling thought processes, but often for such a problem it’s 30 seconds at best out of 10 minutes straight. If I’m in my differential equations course and we’re doing an ‘applied’ example problem, perhaps the instructor needs to give a brief hint of context from the field of application; most likely the example problem came from some textbook, however, so the instructor often doesn’t actually have that context. My best instructors only managed to do this by teaching the story behind the development of the content at hand; the stories of Einstein’s correspondences grappling with the instantaneous information transmission implied by naive interpretations of Maxwell’s equations formed the basis for my understanding of special relativity. If you’re an engineer, think back- how many of the ‘idealized’ problems our coursework focused on in homeworks and tests actually would involve 10% or more of the time fitting any of these 4 things:

  • Modeling Thought Processes
  • Sharing Cognitive Structures
  • Giving Context
  • Telling Stories

But when focusing on ‘real’ problems, engineering design problems, you see all 4 of the above in spades. Because we’re actually trying to design things to do stuff to address problems, we need to have developed intuitive understandings of what the different technical tools in our toolbox do. We need to explore the thought processes involved in driving all the way from abstract user needs like “I want a thing that can deliver my water bottle to me from my car mid-hike”, to the broad strokes questions of finding what options for design-types are available and selecting one. Suddenly we’re lecturing on the cognitive structures by which I conceptualize different types of motors, because in order to actually make things you need to lay out a whole plan without restricting yourself to specific exact components. And because these problems are coming from addressing needs, examples used in class naturally lend themselves into telling the stories behind them- they are problems involving individual human beings, rather than purely hypothetical interactions between abstract objects removed from place and time.

Technology in the Classroom

The other piece I found myself really responding to was the discussion of tech access in the classroom. The accessibility aspect of policies blocking students from participating cannot be understated; I was outed unwillingly in several courses at my alma mater by professors with strict no-technology rules running afoul of my accomodations from the disability resources office. I’d end up sitting there in the front two rows with my laptop clearly out, open, and on, blatantly using it unpunished while the instructor scolded student after student for taking out their phones. In the end in each course I ended up fielding questions or accusations of favoritism so much that I stopped using my laptop despite my accomodations, and eventually no longer bothered attending the class times. It’s an uphill battle to even feel able to ask for accomodations, such policies pose additional barriers each of which renders your classroom increasingly inaccessible, filtering out disabled students ‘multiplicably’.

This is a hot-button issue, however, as plenty of instructors feel resentment against percieved threats to their sole dominion over the classroom. There are instead even easier, completely unabiguously positive possible applications of technology in the classroom: Recording, uploading, and transcribing lectures.

Due to the rise in live videogame streaming sites like Twitch.tv, where individual full-time professional streamers play videogames on a usually daily basis to audiences ranging from 5 to 15,000 concurrent viewers, it’s never been easier to record and distribute your lectures for your students. Programs like Open Broadcaster Software are extremely approachable, even to the tech non-savvy instructor. With laptops including a webcam and mic, the barest minimum quality recording is most likely achievable for free with under 5 minutes of initial installation & configuration time and less than 20 seconds of pre-lecture prep time. Significantly better quality is easily achieved via the use of a simple webcam, external microphone, and potentially a tripod, with total cost for a solid kit running $110 in class-relevant expenses and increasing pre-lecture prep time to roughly 6 minutes, like using a Logitech C920, a BlueYeti mic, and any cheap portable tripod, which is my recording setup. CMS programs like Canvas include built-in media hosting functionality, which is what you see in the screenshot above, in case you only want to expose your lecture videos to your current class; Instead hosting them on a public resource like Youtube or a self-hosted place like on IPFS can dramatically increase your instructional reach beyond those students currently in your classroom. Both Canvas and Youtube include automated transcription systems costing nothing to use (though Youtube’s in particular is flooringly inaccurate at any rare or remotely technical term) and are compatible with uploading & auto-synchronizing transcripts of your own should you have them.

The benefits for recording lectures and sharing them this past semester in my/Dr Wicks’/Dr. Asbeck’s ~100person Mechatronics course cannot be overstated. Students could re-watch lectures to better absorb the content, and were more likely to do so because the playback speed can be varied anywhere between absurdly slow and too fast to comprehend. They can actually listen and ask questions in class because they know they don’t need to take explicit notes in order to have a record of what was discussed. Students unable to attend class, whether due to disability, social, or financial reasons, could easily still see the exact same lecture as those who were present. When giving live demos, such as coding the microcontroller to do a certain behavior or demonstrating the debugging thought process, students can rewind and watch my screen as I do it, pausing and unpausing as they follow along. Students for whom english is their second language, or hard of hearing students, could actually see my lecture with captions! As I move toward sharing them publicly, my lectures can reach so many more people, beyond merely the students who happened to get into my course and attend that day, including people who could never have made it into the college in the first place due to finances or race, gender, or class barriers.

I’ll share a detailed tutorial with pictures later, in a future piece on this blog, but a great place to start is simply using OBS, whatever CMS you’re already using, and your laptop’s built in mic and webcam. Here’s an imperfect example recording, my “Sensor Cornucopia” lecture describing the immense array of different sensors available to the mechatronicist and providing the framework by which I organize their fundamentally simple underlying technologies, all captured via my laptop’s webcam and my external $50 Blue Yeti microphone:

Youtube video: Transducers of All Sorts, recorded on Nov. 7, 2018 by Hani Awni for the Mechatronics 1 course, ME 4735, at Virginia Tech. The webcam completely missed my face for the bulk of the lecture; sorry.

In sum:

There’s immense underexplored potential for these new, digital alternative teaching approaches for the modern era, particularly in Engineering education. Fighting to update our teaching practices is the same problem as fighting to focus our engineering instruction on real, human problems instead of purely abstracted math problems, this shift also corresponds to the fight to meaningfully engage with the sociopolitical aspects of our engineering work instead of our current, presumed-apolitical-by-omission-only framing. In taking on this transition, many of our current teaching practices in engineering educational contexts will need to change if we wish to continue to claim to be acting in light of current research.

Deploying Hypothes.is on your WordPress Blog

I was very excited by the potential of Hypothes.is to enrich or expand the capability of a blog to support discussion in depth. I’ll definitely be using it elsewhere, but want to help others who visit my blog to know it exists and use it.

Turns out it’s easy to add to WordPress, even the plugin is 2 years out of date, ’cause it still works.

Add the Plugin

Inside the dashboard, mouse over ‘Plugins’ and select ‘Add New’ in the main menu.
In the display that opens up of all the plugins you have installed, hit the “Add New” button
Simply select the “Search Plugins” input box, enter “hypothesis”
And it’s the first search result, so hit “add plugin”

Configure the Hypothes.is Plugin

Hypothes.is is now installed on your blog, but actually defaults to being minimally-intrusive; that is, it’s not actually displaying anywhere. I doubt this is what most people want, so here’s how you change where and how visibly it displays:

Back on the dashboard, mouse over Settings, where you’ll now have a “Hypothesis” entry to choose.
The list of settings is relatively straightforward; you can control its default level of visibility, and can control what portions of your blog it’s active on.

I’m gonna just go ahead and open the sidebar by default, and enable it on blog posts, blog pages, and pdfs, in order to encourage joining the conversation!

Troubleshooting

If it’s not showing up even after being installed and configured, I’m guessing your browser simply cached (and is re-using) the old version of the page, without Hypothes.is. You can simply force-refresh the page (Control-Shift-R on Windows, Command-Shift-R on Mac OSx) and your browser will actually ask the web server for the newest version.

If that’s not working, feel free to scream obscenities into the abyss comment here!

Week 2: Networked Learning, Open Working, and Tension

Hi y’all!

I cannot speak for others, but throughout my development as an engineer (2.5 decades) and now as an instructor (4 semesters), I’ve experienced significant tension between the externally-imposed expectations and goals I’m supposed to work toward, versus the underlying reasons I am an engineer or instructor at all.

Tension

I’m lucky to have done plenty of reflection & exploration, on my own, into my “why”s. I’ve an essay from kindergarten explaining that I want to become an “inventor-doctor” and “give people new arms or legs or stomachs”, pieces from my first undergrad about facilitating “the direct transmission of abstractions and experience” in order to “cultivate empathy”, from my second about “engineering as a liberatory endeavor”, and from grad school about “design for community autonomy”. I openly discuss the content I teach, and have accumulated a double-digit waiting list of friends, acquaintances, or friends-of-friends who I want to provide access to my lecture recordings.

But when I engineer, the focus imposed on me prioritizes markets, profit, hunting ‘whales’, intellectual property and preventing others from being able to use the fruits of my labor unless they can pay. These are not and never were *my* goals; I engineer because I want people’s lives to suck less, want to expand our capacity for interaction or self-expression, or to cultivate wonderment and empathy.

When I teach, I am told between the lines that the course is to function as a rigidly pre-prescribed, standardized content delivery vehicle of a specific list of mathematical abstractions and analyses, with an ultimate output of future laborers, labeled neatly like grades of ground beef, for their future employers’ benefit at the salary negotiation table. For engineering-education-as-a-buisness reasons, I am specifically to only teach these individuals who are able to pay (not those on the waiting list), only teach them the content of my course (not others; that’s another 3 credit hours of tuition), only discuss content deemed ‘nonpolitical’ (not the frameworks and practices they need in order for them to feel good about the role they play in the world at the end of the day). In reality, I put myself in a classroom because I want to cultivate a sense of agency and responsibility in my students as agents of change. I want to empower my students with the requisite skills, knowledge, and practices to let them look at the ways the world sucks for those around them and dive in to help. My teaching strives to be a rapturous, awe-filled celebration of the absurd complexity of the world we sentient great apes live in, and hopes to cultivate a similar sense of wonder and curiosity. Where is the SPOT question, GPA scaling factor, or undergrad resume entry for that?

My goals, as an engineer and an instructor, are best served by openness– of designs, processes, concepts, plans, abstractions, and aspirations. I’ve done enough by now that the incentive systems and expectations around me will simply have to adapt. I’ve plenty yet to learn on practices I can adopt which match and richly enact my theoretical inclinations and intent.

Open, Networked Academia

I found the readings this week on open working and networked learning, along with the in-class discussion last week around open-access education, grading, and surreptitious data gathering by for-profit web services in academic contexts all exceedingly resonant. In the class discussion, I even found myself missing established relatively non-obtrusive “that’s a good point” group discussion signals the way snapping or golf clapping are used in certain contexts.

The LSE piece, “Twitter and Blogs Are Not Just Add-Ons to Research” strongly echoes the tension I expressed, which is quite curious to me as it very much speaks from a humanities scholar’s perspective. I can easily see how the work of engineering academics would be rapidly steered by the incentives of military or industry funding sources toward closed, insular, and obfuscated endeavors; and I can see how the role of the engineering instructor would be re-oriented away from content like ethics, human-centered design, and social justice and toward producing easily-categorized replaceable-part engineers for companies. I am still early in my growing understanding that similar processes have been co-opting potentially revolutionary or hegemony-challenging work and terms and de-fanging them; I absolutely had not considered how such incentive systems can be steering education & research within humanities departments.

One particular example of a novel form of public academic via open, networked writing and engagement is Rachel Garner, whose blog “Why Animals Do The Thing” is home to both short-form explanations of animal behavior and caretaking guidelines on social media, and long-form, cited, traditional academic articles such as her recent work on the estimated big cat populations in the wild and captivity, “Fact or Fiction: Big Cat Crisis” . In a similar vein, several ‘public media creators’ routinely do work that, while in the form of a Youtube video and definitely not in the wonderfully-incoherent Formal Academic Voice of an academic publication, engage with the relevant concepts at the depth of their corresponding academic work. For some examples, feel free to look into Veritasium (physics), Smarter Every Day (physics), PBS Spacetime (astronomy and physics), Styropyro (Optics), The Brain Scoop (taxidermy and animal museum preparation), Contrapoints (anarchist and feminist thought), HBomberguy (media criticism), Healthcare Triage (public health and medical practice), Kat Blaque (feminist thought), One Yard Revolution (gardening), or former creators like ViHart (math), and PBS Idea Channel (?cultural anthropology?).

A notable upside to approaches like this is that their funding sources are often wholly divorced from existing institutions, permitting much greater freedom in research & outreach efforts as compared to the extremely limited availability of grants. To me, these individuals are living out a very new form of academic, that takes the roles of researcher and educator out of the sole domain of the university and back into direct engagement with the public.

Reinventing the Open Web

The piece Working Openly: A Manifesto, struck upon a particular struggle I’ve felt in my engineering work more recently: the very software systems and tools in which we conduct our work are structured to facilitate the locking-down of knowledge and content. A CRM platform like Canvas structures everything from their data storage to their web interface around the assumption that for each user, most courses will be inaccessible. I have that double-digit waiting list of people who want to learn my course content exclusively because for me to share the lecture videos and notes, they need access to VT’s Canvas instance, then need access to my course within that. The primary advertising point of Github for Education, for example, is that they offer unlimited private git repositories. My engineering courses were pushed to use Matlab and Mathematica by the companies selling them, who have then successfully set themselves up for market dominance outcompeting literally free, equivalent tools despite ludicrous cost and abhorrent business models. Changing our practices and what tools we use is the first part of escaping that.

Unfortunately, in its pursuit of brevity, Working Openly left undone the task of finding practical specific tools to use instead. I’d like to list a bunch of technologies that I’m aware of and building on for personal control of your data and systems; this particular piece is already PLENTY long, so that can come later.

For now, I’ll simply share that I was extremely pleasantly surprised at Reclaim Hosting’s approachability and their library of existing, integrated applications available, and especially enthused at the unparalleled value for a $2.5/month cost load. For comparison, my vanity domain costs approximately $13/yr on Namecheap on a discount, and a webserver capable of running WordPress (you’re on your own for management/support of it) on DigitalOcean’s cloud server space will run you $10/mo. I even emailed Reclaim Hosting’s support at 2am asking about self-hosting Red5 (a live-streaming service like Twitch.tv), a personal Mastodon.social instance (a twitter-like federated social network), and an IPFS node (a decentralized, content-addressed file storage system),

and I received a detailed, technical reply from the co-founder within an hour:

Finally:

I very much look forward to continuing this course. I strongly suspect it will finally address the questions I’ve been struggling with around “How do I do, in practice, the sort of open, accessible, agency-cultivating teaching I have been struggling to even conceptualize?”