Skip to content

Pegasus Librarian Posts

Video Production: One Novice’s Workflow

Like lots of us, my work pretty quickly shifted this year to emphasize remote instruction. And remote instruction means (among other things) instructional videos. And I have never made an instructional video or any other kind of video that wasn’t just pointing my phone at something cute my nephews or pets are doing and then sharing that with friends and family. So… I don’t know what I’m doing. Like, not at all.

Last spring I made a few videos using our institution’s lecture capture system, Panopto. (Insert shudder here about the panopticon…) Pros: I was able to get up and running with no-frills videos quickly, and I really appreciate any help I can get with accessibility features like captions. Cons: editing is extremely limited, and I just couldn’t get it to do some of the things I needed.

Me in full video-projection mode

So over the summer I watched some YouTube videos about making videos (very meta), and then I faced down almost a week of script-writer’s block, and then I spent a week writing a whole bunch of scripts. And I made myself a slide template so that my videos would have a consistent look to them, which I’m hoping will help me mix and match them for the various courses I’ll be supporting. And now I’m deep in the weeds of video production.

Here’s the process I’ve developed so far:

  1. Write a script (trying to get things down to 5 minutes or less means I can’t risk too many tangents, and making videos that people may need to watch more than once means I can’t risk too many stumbles, so scripts are where it’s at for me right now)
  2. Create slides in PowerPoint using my template
  3. Export the slides as large-ish JPEG images
  4. In QuickTime, record a “movie” of me going through the script. (I don’t use my face through the whole finished video, but if there’s any part of this where I want the video and audio synced up, it’s when my mouth is moving, so it’s easiest for me to just record this all and then overlay it with other stuff later where all I need is my voice.)
  5. In QuickTime again, record any screen captures I’ll need of me navigating through things or whatever.
  6. Sometimes I need screen captures of me drawing or annotating PDFs or whatever, and I do those on my iPad.
  7. In iMovie, edit the places where I stumbled or whatever, and then drop in the Slide images and screen captures (usually sped up to x2 or x4 speeds) where appropriate.
  8. Sometimes I need to do more voice-over work in iMovie.
  9. Export my movie to my computer
  10. Import my movie to Panopto
  11. Use Panopto to generate auto-captions and then go through and edit the captions as needed.

If the video isn’t super specific to a single course, I’ve added two more steps:

  1. Download the caption file from Panopto
  2. Upload the movie and the caption file to YouTube

Now I have two places where students can find my videos:

  • Panopto: easy to feed into their Moodle courses, etc, and familiar on campus for course-related viewing
  • YouTube: easier to stumble across or use for less formal work

And through all of this, one of the big things I’ve learned is that it is absolutely possible to be super corny and super boring all at the same time! Weeeee!

My main other take-away is that need to figure out a teleprompter situation. Right now I’m not very happy with the fact that my eyes are always just slightly down from camera even though I’ve pushed my script up as high as I can on my computer screen. Recommendations for good set-ups are welcome!


Preparing for Fall Term during Pandemic Times

Like everyone, I’m deep in the weeds of pandemic-era librarianship. Unlike lots of folks, my institution’s classes haven’t started yet. But because I’ll be teaching 100% remotely this Fall (even though some classes will have on-campus components) I’m basically doing a whole ton of my Fall instruction right now, at my dining table, while trying to keep my cat and my bird off camera and off mic.

I’ve been teaching for 15 years — longer if you count the years I spent as a dance teacher. There are parts of this gig that I can do in my sleep. There are moves I’ve learned to make in the classroom as naturally as breathing. Sure, I’m always learning and tweaking and generally feel like I’m not actually teaching as well as I’d like to be, but I’ve developed a style, a pace, a repertoire.

And right now it feels like everything I know how to do in the classroom is varying degrees of useless. I’m back to square 1. Or maybe square 2.

Here’s some of what I’m learning and thinking about right now.

  • Accessibility is hard
    The vast majority of my video production time goes to captioning, and for other online things I’m working really hard to make them fully accessible. In the face-to-face classroom these issues exist too, of course, but up till now most of my time was spent learning how to make accessible face-to-face encounters work out well for everyone. Now that’s all out the window and I’m spending hours upon hours editing and syncing up captions and click-through tables of contents and alt-text. Super important work, but extremely time-consuming.
  • Panopto (my campus’ main lecture capture tool) is both easy and hard
    I like that I can relatively quickly capture video of me talking while demonstrating or using a slide deck. I also really like that it produces videos in an interface that our students are becoming pretty used to, so I can more easily assume that they’ll know what to do when they land on a Panopto video page. And I like that as the faculty and I are all re-learning our jobs, it produces analytics that can help me figure out which approaches worked in which situations. (And yes, I always set it up for anonymous access – I don’t want or need individual student information.) So I’m planning to have all of my course-integrated videos served up via Panopto into Moodle. But I don’t like that I can’t do some of the video-clip combing that I want to do for a few modules, and editing the captions is an absolute bear (not only does it take forever for each change to save, but if your caption is timed to start within about a second of a cut the caption won’t show on the public side! Ack!! So much fiddly editing even beyond fixing “in utero” back to “in Zotero”). I also don’t like that I can’t figure out how to do good revisions in Panopto — if I have to change a small thing I have to re-record the video. And of course, then there’s the ominous name “Panopto”…
  • So then there’s more video production to learn
    Today I’ve been learning iMovie. And how to record my screen on my iPad in a way that allows me to draw things and then import those videos of me drawing things into my screencast. And how to record new snippets of audio to replace audio that I messed up in the original recording. And how to write timed closed caption files to upload to Panopto and YouTube whenever I upload a video.
  • Plus I now have a YouTube channel
    I was thinking about how to live in a world where help-seeking will look really different compared to the extremely in-person-based methods we generally use at my institution. I can’t rely on people happening to see me when they come to print their papers. I can’t rely on people wanting to email me rather than talking to me on the sidewalk. So in the spirit of “be where they are” I’m trying to be more in their Google results… hence my brand new YouTube channel. I’m planning to put all the videos I make (that aren’t super specific to a particular assignment or class) up on YouTube so that if students Google something research related they’re more likely to find me and then be reminded of the things I taught them in our library session. Right now, it’s home to a playlist I made for a French course on getting started with Zotero. (If you watch it, please look past my stumbles…)
  • Moodle Integration
    I worked with our Moodle person on campus to get the LibGuides LTI working, so now we can get appropriate guides sucked right into Moodle. I still wish that there was something easier than custom metadata that would allow this kind of interaction, but right now the fiddly custom metadata route is the price we’re paying for more seamless integration into what has become students’ primary classroom. Such is life.
  • More chat reference widgets
    I made a chat widget to integrate into several of our core database platforms, and our eResources person is working on getting those loaded.
  • Custom vs Generic
    One big thing I’m wrestling with right now is when to do highly customized instruction (which is our norm) and when to provide generic videos paired with assignments, guides, and/or Q&A sessions. Students here really respond to the personalized, custom, course-integrated work we generally do, and this funnels them into our liaison appointments quite nicely. But I simply can’t do that in the current environment 100% of the time. 6 short Zotero videos took me most of a week’s work… So which things need to be exactly how customized? For example, I have an upper level course in one department where I know students get very few opportunities for library instruction before their senior thesis, so even though they need a pretty generic thing from me, I decided it was important to have that thing delivered all in my own voice rather than in equally good (or better) videos made by my colleague. I want them to know that I’m their librarian. But then for a first year seminar I kind of want to find videos from my colleagues as much as possible so that they come away having learned that there’s a whole team of librarians at this library, all of whom are awesome and available to help students throughout their college careers.

So yeah, nothing earth shattering here, but that’s where my brain’s been for the last while, ricocheting wildly between big things and little things – solvable things and unsolvable things. And now I’m going to go back to story-boarding a couple of videos for core concepts that I teach ALL THE TIME, and that take 5 minutes or less in a classroom, but will probably take me several hours to put into video form… Wish me luck!

1 Comment

Teaching Primary Source Discovery

I was recently asked if I could walk someone through how I typically approach teaching primary sources. I’m still thinking through how my approach will change this fall in light of my increasing sense that our primary source discovery options, at least in the world of digital primary sources, are part of the structural racism that permeates academia. But here are three of the main kinds of instruction I have done around primary source discovery up to now.

Non-Text-Based Source Discovery

  1. Start with an interesting primary source related to the assignment at hand.
    1. Working in small groups, discuss what kinds of things you can learn from this primary source. What themes would it help you explore. What do you notice that you’d want to investigate further… generally, do a bit of primary source analysis on the spot.
    2. Come together and collect these themes and “topics” and questions, and this source might be one of your sources for that paper — if you were doing a research paper this might be the topic of your paper or your research question.
  2. Now look at the record for that primary source (maybe in an archives catalog, or a digital collection, or whatever). 
    1. What are the words that are available to you in that record? If we use Control-F on the page, what could we find? 
    2. These are the words that we’ll probably have to use in a search box to find items like this one.
  3. I explain that a general strategy for finding primary sources is to find a “bucket of stuff” to comb through. You, the human, will be able to see what the primary sources are “about” but the computer only knows what the primary sources “are.” So the goal is to use words that are very concrete, like the ones we saw in that one record, to gather together a subset of the sources in a collection. And then we as humans look at what we’ve gathered and select 1 or 2 or whatever that we can see are “on our topics.”
  4. Students practice
    1. Sometimes I’ll have a worksheet where they can take notes about the search terms they come across as they open up item records so that later they can remember what search terms to experiment with.
    2. Keep track of what worked well and what was a challenge for group discussion
  5. The group discusses what they found, what was hard, etc.
  6. I impress upon them that this is not a task that can be rushed — it will take them a lot longer than they probably think it should. Plan accordingly!

Text-Based Primary Source Discovery

This is the one that will probably need the most change, but here’s what I’ve done so far.

  1. Sources are not categorized by what they’re “about.” That kind of meaning making is the work that you will do in your analysis. They’re also not categorized by the kind of person who created them. In most search systems, we can’t search for poems about colonial oppression or novels by Black women. (quick demo — zero results)
  2. Instead, you have three main strategies. In order:
    1. Use secondary/tertiary sources to identify primary sources. Find articles etc about your topic and see what they used for primary texts. Then find those texts or texts by the same authors.
    2. Use Instrumental Reading (more on this below) to find words and names that are associated by your topic and used by the people who are creating the kinds of sources you’re interested in. Then use those words/names in your searching. (Pro tip: this also works with scholarly source searching — not just primary source searching.)
    3. Use your own creativity and empathic powers to put yourselves in the shoes of the authors who might have created the works you’re interested in. What words would they have used?

Instrumental Reading

Instrumental Reading is a way to gain access to the language of the authors I care about for my topic, and then I can use that language to come up with fancy boolean searches.

  1. Hand out a page or two of a reading (include the bibliography if there is one in addition to the body pages). For best results, use a reading that they used in class already
    1. Explain that now we’re going to learn a different method of reading the text — not to understand its argument but to understand how to find other things like this thing or in conversation with this thing.
    2. This is all predicated on the idea that social groups develop shared language, and our job if we want to find what they’re saying is to know their language.
  2. Students work in groups of 2 or 3 to mark up the pages they’ve been given. Circle/note
    1. words associated with the various topics in the paper
    2. names of people/institutions
    3. disciplines mentioned (especially important in interdisciplinary classes)
    4. citations referenced or alluded to
  3. As a group, we develop a shared list of the words they’ve come up with, often on the blackboard or a shared doc or something.
  4. For the rest of class, I use words from this shared list for sample searches or to talk about why we might choose one database over another, or how to follow up on citations.

Here’s an example of a reading that I marked up for an intro course to Linguistics.

1 Comment

Reimagining Primary Source Searching to Help Dismantle Institutional Racism

Primary source searching is hard. It has always been hard. First there was the problem of extremely limited access (unless you had travel funding and archives access). Then, after the digitization boom, there’s the new problem of helping students understand that they can’t search for topics or ideas; they have to search for concrete things from the source description or from the text already in the source. “Postcard” will likely be in the metadata about a postcard, but “depicting domesticity in the 18th century” is just not part of the metadata as a general rule. I tell students that they have to search for people, places, or things, not topics. And even then it won’t be comprehensive. And there’s literally no way to search for “paintings by women” or “novels by Black people.” That’s just not how the systems are set up, I say, over and over and over. You have to literally type in the letters-in-a-row that the original authors typed, or you have to know the name of the creator, I explain, over and over and over. If you want to find out how x group is referenced in newspapers, you have to OR together all the names and words that might have been associated with that group, I instruct, over and over and over.

And therein lies the rub. I am no longer willing to inflict on my students the trauma – the violence – of ORing together all the epithets that have been used in newspapers and legislation and editorial cartoons and broadsides to refer to minority groups. It’s one thing to be presented with these terms once you’ve gained access to a historical document. It’s quite another to have to use your imagination, creativity, and research skills to come up with these terms. And then after all that you have to actively recreate these epithets by typing these terms into a search box?? All neatly strung together with your fancy boolean operators?? No. Doing that myself is painful. Requiring students to do that in order to gain access to the historical record is horrible.

Going through and improving the metadata in our digital collections is going to be hard, expensive, and time consuming. The historic record is quite large, after all. But we and our vendors must do this work. It’s our ethical, moral, and social responsibility, and the technology exists to make it possible. We’ve been applying subject metadata to secondary and tertiary sources for years — for decades. And especially now that curricula have shifted toward teaching from primary sources more and more, we can’t hide behind the convenient excuse that “this is just the price you pay for studying history.” No. This is not a price we should have to pay. This is certainly not the price that my Black, Indigenous, LGBTQ, Latinx, and other historically marginalized students should have to pay in order to study history and culture.

1 Comment