Observer: John O’Reilly
Observee: Jasper Shuoyang Zheng
Part One: Context
What is the context of this session/artefact within the curriculum?
- Methods 2: Digital Systems is a Year 1 unit in BSc Computer Science and BSc Data Science and AI, at UAL Creative Computing Institute. The unit is led by my colleague Dr Kayalvizhi Jayavel, who commissioned me to give two guest lectures on digital image processing and linear algebra. I will be delivering the sessions on the 10th and 17th of March. Each session will last about 3 hours, with 1.5 hours of lecture and followed by 1.5 hour of class activities.
How long have you been working with this group and in what capacity?
- This is my first point of contact with most of the students. About 10 students from the BSc Data Science and AI course already met me when they were in one of the units that I taught last term. I’m new to other students from BSc Computer Science.
What are the intended or expected learning outcomes?
- Taken from the unit brief:
- [These two sessions] investigates the process of digitising and treating reality as sets of numbers. It introduces techniques of linear algebra as ways of dealing with large datasets, such as digital images, and how they can be manipulated using software packages. Inherent in all this abstraction are codified power relations that need to be unpacked to begin understanding the effects of computing on all groups in our larger, interconnected world.
- I’ll be covering the LO2 and LO3 of the unit:
- LO2: Identify and apply basic concepts of linear algebra such as vectors and matrixes (Knowledge)
- LO3: Experiment with different methods of representing, storing, and manipulating datasets in digital systems (Enquiry)
What are the anticipated outputs (anything students will make/do)?
- The planned activity session is for students to study the Python/JavaScript codes for the case studies in groups, and verbally describe and explain the codes to the class.
Are there potential difficulties or specific areas of concern?
- Since it’s a joint unit, students from BSc Computer Science might not be familiar with some of the concepts in linear algebra, because they didn’t have the Math and Statistics unit last term. I’ll try to explain these concepts in easier terms.
How will students be informed of the observation/review?
- I’ll send a Slack message to the students before the class to inform them that the session will be observed by a colleague in the room, and clarify that the instructor is the subject/observee. This will be verbally reiterated at the beginning of the session.
What would you particularly like feedback on?
- One of the feedback I got from Ignacia last week was that the lecture part was overly long, it would be good to break it down into smaller sessions, and let students play with the hands-on practical activities right after every concept was introduced. I’ll try this strategy this week, so it would be great if I could get some feedback on the overall flow of the lecture, and whether students are actively experimenting with the practical activities.
How will feedback be exchanged?
- Ideally by email.
Part Two: Feedback from John
#teachingspace
The classroom space is bright and large with quite intense strip lights. There are two large screens at the top of the room and two screens in the middle of the room attached to the ceiling. There is a table at the back of the room by the entrance where students can engage, it looks like a way in which community is being fostered. In relation to this, I liked the image of the forest as the screensaver, I wondered if this image of nature was intended as a contrast to the bright, lab-like intensity of the room? A good idea! I also wondered whether an image from the lecture might have been displayed, just to connect people with the content of the class. It can be a challenge to bring learning spaces to life – perhaps I was aware of this also due to the size of the room and numbers of students.
#stagingtheintroduction
As the time gets closer to the start you have the lecture title on the screen which is a useful and helpful way of focusing people as they come into the room. You make a lovely introduction inviting questions about last week’s class – establishing the narrative and continuity of the learning design is great practice (need to do more of this myself!) and you announce that the lecture is going to start in two minutes, which is a nice, effective way of slowly gathering people’s attention away from their chat.
#learningexpectations
What’s interesting is that you have the text running along the bottom of the screen – this makes the lecture more accessible, the scrolling text is another place for the students’ attention to land, complementing your voice and the slide. Having the slide with the agenda and the schedule is great and again, connecting to last week’s lecture is really excellent in establishing the continuity and design of the learning experience for the students. As you show the introductory slides you have an effective narrative voiceover that enthuses the students on why this topic is so interesting. Your pacing, to this listener, is really excellent. You also explain that the case studies are going to be the focus of the activities later on which is a great way of helping the students organize their expectations and energies for the session. It was also really useful to situate this particular class in terms of the learning outcomes – it’s such a helpful landing point for students, a place to situate what they are learning in this particular session.
#messyworkingout
When you begin to work through the slides you write on the screen in red-ink, working out and showing the mathematics of the activity. The ‘messiness’ of this is really helpful in mirroring how we figure things out when we learn, it is a really strong ‘aesthetic’ of learning, of representing how we begin to know things.
#learningwithchallenges
I thought how you foregrounded activities and slides in terms of ‘challenges’ was great too, that you identified the tricky stuff that this class was going to work with – it is really helpful to present specific problems and then do the storytelling of working out. There was a nice image example explaining image recognition that the machine needs to recognize the pattern – throughout the session you used the image examples rally well, especially in the contrasts.
#checkingin
As you worked through the slides it would be worth checking in if people are following your thinking, if they have any questions – these questions may of course emerge in anyway in the practical part of the class when the students do the case studies. It is always a difficult process to know when to take questions, some students are happy to keep going others can get lost, when to check-in is a matter of judgment. In a good way checking in also changes the class dynamic, makes it ‘interactive’ when you did check-in there is lots of nodding and confirmation that they understood the concept you were teaching.
#breakout
When you shared the link and asked them to experiment with the kernel and the values it is a useful breakout moment, just to break up the concentration demanded in listening to the lecture. All the students engaged with the activity discussing and sharing how they are changing the image post up. It was great you then checked in with the student who has support from the two signers,
#storytelling
You were effective in connecting the theoretical and technical aspects of the first part, to the different kind of learning happening later in the class. The nod to Eliza was excellent and it may have been interesting to give a little more history and storytelling around this. When we become expert in our research, as teachers we often forget just how much we know and take for granted, and telling the ‘human’ (and industry) histories around examples such as Eliza is a great way of bringing technical or abstract material to life.
#sharingquestions
You are really considered and thoughtful building out the context from each bullet-point on the slides. When a student asked a question (think it was Nicola?) it might be worth repeating the question out to the class as a whole, so every can hear the question and follow your answer – “Nicola just asked…” Sometimes you can then ask others to see if they have any thoughts on the question. When you do talk through the answer the student has an ‘aha moment’. It’s also excellent that you direct the student to the slides from the previous week – students often don’t connect their learning in the ways we design it!
#debatehistories
There was a really interesting example and questioning around the issue of gender bias in vectors that are embedded and wondering whether in material like this it might be worth opening this out for discussion? Perhaps even just a nod to the history of this and how was it discovered. Sharing histories and contexts in such hot topics are invaluable ways of helping students connect the technical elements of their learning.
Summary
I made some suggestions above Jasper, and in general to a non-expert such as myself, the pace of your delivery was really considered and measured – that can be a really hard thing to do, well done. You had a lot of material to work through, but it never felt rushed or dense. And though I didn’t get to see the workshopping part of the class, the way you signposted different parts of the lecture to later work on the case studies was really well designed. It helps the student imagine and locate themselves in a future task. I hope the rest of the class went well.
Part Three: Response from Jasper
John gave thorough and thoughtful suggestions for the lecture session. Many ideas pointed out by John were very inspiring and constructive.
John pointed out the idea of using a forest screensaver to contrast the lab-like intensity of the room. There hasn’t been any design of this (it’s actually just one of the default wallpaper in Macintosh), but pointing out how it might have created a lively environment is a reminder for me to think about setting the tone/vibe of the classroom. For instance, () mentions how lighting can affect alertness and the overall ambience in a learning space. And in fact colleagues at CCI have recently bought plants for the third-floor kitchen, which is such a nice idea for brightening the room.
I learnt the on-screen red-ink approach from typical math pedagogy – the instructor works out equations on a writing board. One of the advantages I found is that addressing while writing down key points/formulas is an effective way to slow the pace down, and therefore giving learners time to think.
Oratory and fluency in hosting conversations between the instructor and students (or between students and students) is a skill that I hope to improve at. John suggested repeating questions from students for the class (e.g., using sentences like “[name] just asked…”), which is indeed something I feel that I should have done. In addition, the suggestion of directing questions to other students is inspiring! I’ll definitely put that into my “strategy list”, as well as the suggestion of turning the bias issues of language models into a topic for an open discussion.
The second half of the class – the hands-on workshop session ended smoothly, lasted for about 60 minutes. We divided the class into three groups, and each group worked on a case study. I instructed all groups to run the source code for their case study, and explore how it implements concepts mentioned in the lecture. I tasked them to present the case study they worked on to the class. I prepared prompt questions for groups to structure their presentation (e.g., What does the project do? What library did they use? How was the data loaded and processed? How were the algorithms/techniques implemented?).
During the activity, I observed an interesting collaboration strategy: Two groups stayed at their table, after each student played with the source code individually on their machine, they eventually moved all content to a shared laptop. Their inspection of the source code and preparation of the presentation were mostly done together. During the activity, one or two students actively contributed to the slides and took control of the source code. Some students were less active but still had their attention on the shared laptop, others were working silently on their own laptops. This mirrors the suggestion from Harris (2022) that “students who are quiet because they are inclined toward silence”, this preference to “lurk in the shadows” for processing, considering and reflecting should also be acknowledged.
Overall, the unit is at its growing stage (this is the second cohort of this unit and I’m handling the unit for the first time), I’m planning to add more content and materials and maybe replace some of the existing ones. It was nice to have John to feedback on the session from a third-person point of view.
Reference
- Harris, K., 2022. Embracing the silence: introverted learning and the online classroom. Spark: UAL Creative Teaching and Learning Journal 5, 101–104.