DataJConf 4.0 & C+J

June 23, 2023

Last week we held the 4th edition of the European Data & Computational Journalism Conference (#datajconf) in Zurich, Switzerland, and this time it was held jointly with the Computation + Journalism Conference (#cplusj), which is usually held in the US.

I wrote about the genesis of DataJConf back when we held the first edition in Dublin in 2017, and I spoke a bit more about the general ethos of the conference with the third edition in Malaga in 2019. We’ve always encouraged a strong multi-disciplinary approach, aiming to attract journalists, developers, industry professionals and academics from across many fields including journalism and media studies, computer science and data science. It was really gratifying to see that this year was no different, with a great mix of all these groups, and more, in attendance. Our teaming up with C+J meant that we had more attendees from the US than we’ve had before, which I think will have been a great opportunity to strengthen links between Europe and the US, and has also helped raise the profile of our European ‘little sister’ conference to the more well established C+J conference.

This was our first conference back since the pandemic, it had been 4 years since we last got together and discussed all things Computational + Data Journalism. It was fascinating to see the progress that has been made in the last few years, and how both academic research and industry practice has evolved.

It’s no surprise that AI dominated much of the agenda, with some very interesting and revealing discussion of the use of GenerativeAI by news and media organisations. What was most revealing though was that widespread adoption of GenerativeAI tools is still some way off, if it will ever happen at all. Most organisations that have experimented with these sorts of technologies have found that the unreliability and ‘hallucinations’ that can be introduced create all sorts of integrity and trust issues when using GenAI to create content, which generally outweigh the benefits. Many newsrooms/organisations are instead for now sticking with rule based/templating automation for content generation, which is much more reliable and controllable. Where GenAI has found use is in transforming/summarising existing content, which again can be more controlled/reliable. On the AI front there was also some discussion of deepfakes, and the potential issues arising there, though little discussion of solutions, perhaps because we don’t have them, or perhaps because existing fact-checking and verification techniques are already sufficient to deal with the problem?

The other big development that was noticeable at this conference was the increase in algorithmic accountability efforts - news organisations and others working to investigate the impacts and biases present in algorithmic decision making processes that have a real impact on people and society. An increasingly important area of concern, that was not really touched upon in previous editions of the conference, but which is now a focus for many teams.

Teams was also an interesting revelation from some of the talks and discussions - where previously the EU has perhaps lagged behind the US in terms of the development and position of data/computational teams in the newsroom and other media organisations, it does feel like there has been a change in the intervening years in some organisations, where these teams are now bigger, more established, and perhaps more central. Certainly, whereas in our first edition much of the talk was about small, niche teams on the fringes of media operations, this time it felt like we had a number of talks where the data/computational teams were key to some of the organisational strategy. A good development to see, though it is also clear from a number of talks that there is still work to do in this area.

For a really nice roundup of some of the issues I’ve touched on above, and more, see this thread from Jim Haddadin:

You can also go back and check through the hashtag to see how the conference unfolded.

Overall it was a really successful conference. Personally I didn’t have to get involved in much of the organisation this time around. The local team did a great job pulling it together, with Bahareh doing most of the steering effort. It was great to catch up with familiar faces from previous editions and conferences, and to meet new people too. I’m looking forward to the next edition …

Students on a scale

April 5, 2022

After our first teaching workshop of our review of undergraduate programmes we had some intial thoughts about what students in 2025 might be like, and what they might learn.

I’d like to unpack the framing that Carl Jones came up with after that workshop, presenting the complimentary but perhaps also conflicting elements that might make up a future Computer Science or Software Engineering graduate, as a series of points along a set of different dimensions:

CS/SE Specific

  • Fabricators v Assemblers (or White-box v Black-box)
  • Low-level v High-level
  • Work with computers v work with people

General/Transferable Skills

  • Work alone v work with others
  • Rigidly structured vs. Flexible learning
  • Research-aligned v industry-aligned

With apologies to Carl for anything I get wrong, and with apologies to Computer Scientists and Software Engineers for the cliches and gross generalisations, here’s how I think this works and where it might fit in developing or redeveloping degree programmes:

Fabricators Vs Assemblers

Fabricators build the bits that the Assemblers put together to build the products.

Fabricators work at the level of components and services, they design and implement algorithms, APIs and interfaces. Their work is often about implementing the theoretical ‘new’.

Assemblers take the components and services and put them together into applications and products. They architect solutions to problems using existing code and technologies, sometimes with little (or no) coding. They are focused on the solutions, and the trade-offs that must be made to achieve them.

Low-level vs. High-level

At the low-level the focus is on efficiency, speed, memory consumption, CPU cycles, power usage. They work close to the algorithms and the data. They are often delivering marginal gains that translate to larger gains at higher levels.

At the high-level the focus is more on wider systems, with concerns for system reliability, data consistency and coherence, uptime and response.

At both high-level and low-level there is an explicit understanding of the trade-offs and impact that the design decisions made at that level have on the project/product/output.

Work with Computers vs. Work with People

At the computer end of the scale, we are familiar with a wide range of languages, tools and environments. We write code, we are on the command line and in the text editor.

At the human end of the scale, we talk to lots of different people - our colleagues, our clients, the stakeholders. We can translate and communicate between these groups.

Work alone vs. Work with others

Working alone is not about the solo coder, at home in the garage delivering a project single-handedly, but is about autonomy, independence and self-direction. It is about being able to align yourself with the goals of the project and team but to determine your own direction along that alignment.

Working with others is about collaboration and communication. It is about sharing of all kinds (ideas, time, respect) and about being an effective colleague, offering and responding to feedback, and keeping the teams goals in sight and mind.

Research aligned vs. Industry aligned

Research aligned work focuses on the cutting edge of CS and SE in both the pure theoretical and applied senses. It may not yet have practical applications in industry but will be pushing at the boundaries of the subject area.

Industry aligned links CS and SE to a more practical applied sphere, delivering necessary skills for the digital economy and powering the wider adoption of CS and SE technologies across business and society.

So what?

If we have these six dimensions (there may be more, or we may need to lose some…) what does this mean for our teaching and our degree programmes?

To me, this is a framework within which our teaching sits. It could even be considered a menu. There are clearly overlaps between some of these areas, and I don’t really think a student would get away with only sitting at one end or the other of any one of these dimensions without at least knowing about the other end. However, I can envision a student coming out of Year 1 of a degree scheme, a solid grounding in core skills under their belt, having experienced both ends of the scales, and now they are making the decisions about where to specialise and using these dimensions as a guide for some of those decisions. Perhaps they are leaning towards being a high-level fabricator type who works with people … so we steer them towards modules that cover or lead towards those areas. Perhaps they are a low-level assembler who is very research-aligned, and so again, we steer them in the direction of the modules that will enable those sorts of outcomes.

I think this is a model we’ll come back to as we continue to explore the space surrounding our degree programmes, and it will be interesting to discover the other dimensions, if there are any, and to see if this remains a useful metaphor to base some of the programme design decisions on.

Computer Science Students in 2025

April 1, 2022

Our first teaching workshop of our review of undergraduate programmes was student-centred, and focused on what our cohort of undergraduates might be like in 2025. Our staff got together online and in-person to try and tease out some of the answers to questions like these:

  • who are they?
  • what skills and knowledge do they have?
  • how do they learn?
  • what will they do and learn on our programmes?
  • where will they go once they leave us?

A wide range of opinions were gathered, and there were some interesting common themes. What follows is a summary that sets the base from which we can explore what the future of our programmes may look like.

Who are the students, what skills and knowledge do they have

A diverse set of entry qualifications is probably one of the key things that will mark out our student intake in 2025. We’re looking at students with and without A-Levels or T-Levels. Some may have Maths qualifications beyond GCSE, other’s won’t. The same is true of formal study of Computer Science, the curriculum for which differs between England and Wales substantially. The larger changes to the Welsh school curriculum will not yet have filtered through to impact incoming students in 2025, but it won’t be far off, and are something that need to be considered for the future.

While school/college leavers are likely to still be the main bulk of the cohort, there is a potential for older students at undergraduate level, particularly those who want more depth than a postgraduate conversion course may provide, or perhaps a more applied programme.

We need to remember that Cardiff as a city and a destination is a big draw to applicants, who will come from across the UK. Our international intake has also grown significantly over the last few years at undergraduate.

How will they learn?

A level of flexibility will be key here. Different students with different support needs, accessing teaching through a range of methods and modes that will enable their learning. While some may appreciate or expect the ‘traditional’ rigidly timetabled lecture, lab and tutorial experience of University, others will be looking for a more dynamic, chunked education accessed on their terms when it is convenient for them. This will most likely be delivered predominantly online, with additional flexible in-person support. Between these two extremes are a cohort of other students who may expect or want something in-between the ‘traditional’ and the ‘new’. To fully support this wide range of preferences across all teaching may be an impossible task. As we move to a more blended model of learning, we’ll need to help our students learn how to become self-paced and self-led learners - something that (as in many industries) important in CS, given the fast pace of change and the need to constantly develop and learn throughout a career. We need to help students realise that we are not the only source of information and knowledge.

Where are they going, What do they learn?

Students who graduate from our programmes in 2028/2029 will be heading to a wide range of destinations. This range will be getting wider all the time as Computer Science continues to become more pervasive throughout society, as it continues to reach across disciplinary barriers and entwine itself in more and more subject areas, and as the Software Engineering principles that underpin the applications of CS become more and more important. The vast majority of our students will come to us seeking a career in industry rather than research, though the pipeline of students from undergraduates to PhD will remain an important part of what we do as a Research-led institution.

One of the main considerations will be the skills that go alongside the subject specific knowledge that we deliver. While a grounding in the core concepts of CS and SE is essential, a lot of the technology specific content of the degrees is less important than developing students as independent reflective lifelong learners, able to adapt, change and reskill as the technologies move on and the subject evolves. An emphasis on skills alongside the technical is crucial: team working and collaboration, writing, reflection, critical thinking and analysis, decision making, entrepeneurship. Consideration must be given throughout the degree programmes to crucial factors such as sustainability, ethics, and employability.

Our excellent Deputy DLT for undergraduate Carl Jones related a lot of this in a nice way to an Agile manifesto type concept; the idea of a range of opposing ideas or views, with students sitting somewhere along each of these scales:

  • Fabricators v Assemblers (or White-box v Black-box)
  • Low-level v High-level
  • Work with computers v work with people
  • Work alone v work with others
  • Research-aligned v industry-aligned

Each scale reveals something about the type of Computer Scientist or Software Engineer that a student will become, and you could see this model being refined into a tool that could guide a student through the decision points and learning in their degree as they choose modules and pathways. They’ll need to see both ends of the scales throughout their time, but will likely end up specialising towards one end or the other on some of them by the time they leave. We’ll dig into this a bit more later…

Hybrid Teaching Workshop

March 16, 2022

as I’ve mentioned previously, we’re running some workshops to review some of our programmes with a view to changes needed in the future. As part of this process we held our first hybrid teaching workshop this week, and I thought I’d write about the general process before I write about the outputs

“Our intention is to run this workshop in a hybrid fashion - with people in a location tbc … and also online. We’ll see how much of a car crash that is …”

That’s from the invite to our latest teaching workshop, which we ran yesterday with several participants physically present in one of our seminar rooms, and more participants joining in from a Teams meeting online. It was our first hybrid teaching workshop with staff (though not our first hybrid teaching session by some distance[1]). I was a bit concerned whether anyone would actually turn up in person, especially as the weather was a typically rainy Cardiff day, but in the end we had a bunch of ‘real’ people in the room, and a bunch of equally ‘real’ but less physically present people online. It worked well, and I think it’s worth getting some thoughts down about the process.

The structure

The first thing that helped was that the session was structured very much as an active session to facilitate reflection in and gather input from the participants, but with a slight cheat in that the amount of ‘whole group’ discussion was artificially limited. The basic structure was:

  1. A short introduction from me to set the context of the workshop, lay out the plan for the session, and how it fits in with the overall review we’re doing. The aim was for this to take 5 minutes. It took 15.
  2. A first activity looking at a potential University applicant in 3-4 years time
  3. A second activity looking at a potential University graduate in 8-9 years time
  4. A third activity asking ‘what would we do if there were no rules?’

Each activity was run as a group discussion session, with groups of 6-8 people talking about the topic in question for 15-20 minutes, with a feedback session after each discussion. In order to capture the discussions the group present in the room had access to flipcharts and pens, but being a bunch of academics naturally gravitated to the whiteboard. Online the groups were in breakout rooms within the Teams meeting, and they all had access to a shared powerpoint file, with an individual slide for each room on which they could capture their discussion points.

The setup

The online meeting was in Teams and my laptop was plugged into the room’s AV system so that the teams meeting itself was displayed on the screens in the room. I kept the directional microphone on the lectern pointed at the laptop so the participants in the room could definitely hear the people online. A few of the ‘physical’ participants had their laptops with them and were also signed into Teams, so whenever they wanted to contribute to the discussion they could unmute and speak towards a device near them rather than relying on my laptop mic in the centre of the room. I had to remember to mute the audio on my laptop during those moments to avoid horrible feedback.

The reflection

Did it work? Yes. Participants both online and in the room were able to participate in the session. Some of the good things that worked well:

  • Shared .pptx files were great. This is a good technique I’ve used before, and that has been written about in a few places online. Having each group working on a different slide makes it very easy to see which groups are getting on with the task without needing to hop in and out of the breakout rooms. I added an extra slide in each deck for the team in the room, then when we were wrapping up the discussion I could take a photo of the ‘physical’ team’s whiteboard, and drop that into the shared powerpoint so the people online could see the full contribution from the room.
  • The feedback for each activity was also made simple by using the shared powerpoint - we just shared that within the team and moved through the slides, asking each room to highlight anything on their slide that they thought was pertinent.
  • By monitoring the online text chat the people in the room were also able to participate in the side-channel discussion that often happens in online meetings alongside the main thread of audio conversation.

Some caveats on this though:

  • It was a smallish workshop. We had 7 people in the room, and another 24 online. I think it’s scalable. Certainly the notion of sharing the file between all groups would work, but with more groups in the room there’d need to be a bit more time for adding the ‘in room’ content to the online presentation.
  • I assume (thought not with any evidence) that there’s going to be an upper limit somewhere on the number of people that can be editing a powerpoint presentation at the same time before something craps out on you.
  • You do need to be comfortable with multi-tasking for this. During ‘whole group’ sessions you’re trying to watch the text chat online for pertinent comments, look out for people online wanting to contribute and also keeping an eye on the room. During the breakout activity you need to keep an eye on each room’s chat, the activity in the shared presentation and the conversation going on in the room.
  • As I said, the amount of ‘whole group’ discussion was limited. This was partially down to time, as there was a lot to do in a short session. It was also partly as there is a further asynchronous feedback/discussion space available online after the workshop. It was also because I think that’s probably one activity that may not work so well in a hybrid session such as this, so minimising the amount of discussion between groups makes sense.

Really though, the only thing I’d change in future is try to have more of a balance of in-room vs. online participants, so perhaps at the next one we’ll try and have cake to entice a few more out of the digital space. Also we’ll not clash it with an event in the main school building that’s booked all the good seminar rooms, so we can hold it in the same building that people actually work in.

The feedback

Other people also seemed to think it worked:

The outro

Some thoughts there on the Hybrid workshop we ran. I thought it would be useful for people to hear how it went, before we dig into what we actually got out of it.

  1. I’m not here to rehash any of the arguments about hybrid teaching, or hyflex, or whatever the hell people are calling it now. My experience of hybrid teaching has been good to date, and I’ve had numerous students thank me for running hybrid sessions when they’ve not been able to make it in in person. However, as my Head of School noted today, it’s probably not for everyone and it likely suits the sort of educator who thrives on a bit of chaos in their teaching sessions, and it turns out that’s me ↩︎

Programme Review

March 8, 2022

So you’ve decided to embark upon a review of your teaching programmes

We’ve started looking at what we do in some of our programmes. For now we’re starting the process in-house, which means the Scholarship group within the School get to set the direction, speed and path that we take[1]. I thought it would be helpful to lay down some ideas and principles for this process. There are a few things I’d like this review to be, in no particular order:

  • open - we’ll be sharing outputs from workshops (and the inputs for workshop) widely within the School and inviting contributions at all stages from all key ‘stakeholders’[2]. I’m also going to be posting here whenever key outputs are created, so they might get sanity checked and have pressure applied by public scrutiny, which I think will help keep us on the right track.
  • inclusive - this isn’t just going to be a cabal of five or six individuals deciding how we should do things. All staff who want to be involved will be able to be involved. Students both past and present will be included in the design process, as will other key parties such as our external examiners, our industry partners and our external advisory board.
  • comprehensive - this isn’t tinkering around the edges. Let’s really test ourselves here - prod and poke at all the things we do and question why we do them. We’re only going to get to do this once for a good few years, so lets not squander the opportunity by shying away from the difficult questions.
  • limitless - at least initially, let’s not worry so much about practicalities, those can come later. To begin with lets think about how and what we’d teach in an ideal world unconstrained by academic regulations, workload, and academic systems and processes. We can compromise the dream later.
  • iterative - if we need to go back to an earlier stage based on things we’ve worked out in a later stage, we will. Each stage may need many repetitions to capture contributions from all the involved parties, and that’s fine. We’ll go round as many times as we need to get it right…

Why do these things matter? It’s all based on my past experiences of programme development. We’ve done some of the above in most of our previous developments, and when we have they’ve been a positive part of the process - but we’ve never done all of them at the same time. I’ve also seen quite a few programme developments that have not been up to scratch, and they’ve usually done the exact opposite of these things. So these then are the principles I think that will lead us to a successful conclusion, whether we decide that a full redevelopment is needed, or whether we decide that there’s nothing that actually needs to change.

The first task for us is to determine how much change is needed in our undergraduate programmes, if any at all. The model for this is fairly simple: decide what the ideal Computer Science and Software Engineering degree programmes would look like in 2025, then examine how our degree programmes currently are, and if the two are not the same, some change is needed. So that’s what we’re going to do next…

  1. I’m aware that there’s an education development service coming down the track at Cardiff. I got to see a snippet of it, and thankfully it matched my expectations in many ways. However I have no idea how far along the track we are, so we’re forging ahead a bit and we’ll let them catch up to us later as I’m sure they will. At least for now with high-level programme design I’m confident in our abilities, because not to blow our own trumpets but we are really quite good at this. ↩︎

  2. god this management speak is depressing, but I’m struggling for a better and less nauseau inducing phrase. ↩︎

Teaching CS

March 7, 2022

in which we start to think about redeveloping our programmes

Back in early 2020 we had plans. Big plans. We were going to look ahead, far ahead, and think about what Computer Science education really was, what it meant, and how it might change over the next decade. I was planning to scope out a project that enabled us to go and visit people, people wiser than us (like we did at TuE), and see what they thought about it, and how they did it, and then we were going to synthesise all that into a plan for the next decade of how we were going to teach Computer Science.

Then there was the small matter of a global pandemic, and instead of spending a couple of years looking at what might change in teaching over the next decade we instead implemented a decades worth of change in teaching in six months. The last two years have seen an unprecedented shift in how we deliver CS education, compressing the changes we could barely see on the horizon and only had the beginnings of thought within our plans into six months, and slow, planned cautious rollout became a fast, reactive and dynamic launch. We went from ‘the old way’ to entirely online to blended learning in the space of two academic years, and now it is time to reflect on those changes and decide what they mean for the future.

At the same time, this is cs that we’re talking about, that ever-shifting, always-developing, forever-moving subject with its tentacles firmly embedded in all of modern society and its boundaries constantly blurring as it impacts on disciplines that previously had nothing to do with computers and computing. Our student intakes become more familiar with computational thinking, with algorithms, with data structures, with programming, with technology, and the places they will go after us become ever more many and varied.

So the review of how and what we teach, delayed for 2 years, is back on.

There’s a lot of drivers and inputs in this review. Not just the changes to teaching in the last two years, and the pace of change within the subject. The School itself has changed, growing in size, shifting in research focus. We’re doing a wider strategy review in the School at the same time, which is also looking at the future of CS teaching. The University itself has changed, with new support for teaching development, and further changes on the horizon. The school is due for academic revalidation of everything we do in a couple of years, and there’s no harm getting a head start on that. QAA are about to publish new benchmark statements for the subject, accreditation criteria have been revised, and there’s a significant amount of learning from our involvement in projects like the IoC to put into practice.

I’m planning to do this in as open a way as possible. I’ll be sharing plans, outputs and thinking as we go. As I said in my invite to the first staff teaching workshop as part of this review, this could be a car crash, but lets see what happens anyway…


October 8, 2021

Much of this week at work felt like settling back into the usual rhythm.

Monday: meetings and catching up with things and people, putting out minor fires and preparing for the week ahead.

Tuesday: I spent the day online with students, doing the first ‘programming’ session practicing problem solving with a fake programming language. Last year’s students had christened this fake language ‘MartinScript’; this year’s went with ‘BennyScript’. I have not taken this personally. The session itself seemed to go well, there was good engagement, and some of the students even carried on afterwards to attempt the problem solving with a real programming language, which was awesome to see.

Wednesday: CompJ Lab + the usual Wednesday meetings.

Thursday: another day on site, this time introducing the assessment for the Computational Thinking module.

Friday: possibly the most interesting day of the week. 4 open forums with different groups of students, where we were able to talk about some of the things we’re interested in as a School, but also hear some of the issues students are facing, and start to solve them. The turnout varied a bit between the years, but everyone who turned up had some interesting and useful things to say, and by the end of Friday afternoon I’d already managed to put out a few minor fires around the place. A tiring day though, spent almost solidly in Teams meetings (especially when you throw in a DLT forum with the PVC halfway through the day) - by the end of the day I was horizontal on the office sofa with the sleeping pup, attempting to send emails without falling asleep myself.

Overall then, a strangely ‘standard’ week. A couple of work highlights (apart from the teaching) were around some input I managed to have into a couple of new things on the horizon: I managed to make a positive impact on some potential new job roles, and on some curriculum development guidance. Small wins, but I’ll take them where I can.

Outside of work though, a really positive week. Having been accustomed to being a solo runner for a very long time, I realised when parkrun returned that actually I do enjoy the social side of running, and having people around while running is good for me. So, I took the plunge and joined a local running club. I originally signed up for two trial runs: a training session on Monday and then a conversational run on Wednesday, thinking I’d make a decision about joining once I’d done them both. Monday night we did hill sprints on a hill in Penarth and I had such a good time I paid my subs the minute I got home and signed up. Wednesday was a really pleasant gentle run around strangetown where I managed to talk to quite a few other people and had a very nice time. I’m looking forward to having a little bit of a more regular structure to my running (which I’ve missed since stopping the running commuting) and also to having more social things going on.

riding or running or that

Did Parkrun at Grangemoor on Saturday, another 30 seconds off the time, so we’re getting closer…plus there’s that whole ‘running club’ thing.

Did the really really ‘old’ commute a couple of times to get to the Queen’s buildings for in-person teaching. Turns out I can cycle between the two sites in about three minutes, which is very handy when you get caught talking in your office in Abacws five minutes before you’re due in class in Queens…


October 8, 2021

Oh, we’re back baby.

What a first week back. Monday was a blur of teaching prep, the usual start of week meetings and catch-ups, and then in the evening the first virtual governors meeting of the year for one of the Schools I’m involved with.

Tuesday was when the semester really kicked off with the first online teaching session of the year. I had a lot of pre-game nerves for some reason, which is unusual. It got so bad that I had to stop work on Tuesday morning and go out for a run across the barrage and back just to get rid of some of the energy. The teaching session itself was fantastic though; the students were engaged and interactive, the activities went well, and everyone loved Bennie (of course).

On Wednesday it was my first small in-person session; we had our first ‘lab’ with the CompJ students, and we really got to use the new building for the first time. We all met up in one of the seminar rooms, set them a task, then sent them off to work somewhere else in the building. We could then wander between the different collaboration spaces meeting with the teams and making sure all was well. It worked beautifully - exactly the kind of dynamic, flexible and collaborative teaching the building was designed for.

Thursday was the real highlight - in person teaching with a much larger group of students, back in what has been my teaching ‘home’ for the last few years: The Turing Suite in the Data Science Academy space in Queens. It was lovely to be in a big flat active space with students again, and they rose to my challenge (‘recreate the Cardiff University homepage in an hour and a half after only a few days teaching in HTML & CSS’) with great enthusiasm. It was also nice to see a lot of my TA colleagues in person again for the first time in 18 months, and to meet one of our TAs who I’ve previously only met through Zoom/Teams.

Interestingly one of the students approached me after the session to ask what the point of the session was. They explained that they were a ‘beginner at web things’, and they’d expected to come to the session to be taught HTML and CSS, and hadn’t expected to spend a couple of hours working with others to actually create a website. I think it partially means I’d not done a good job at explaining the structure of the week and the session, so there’s a ‘note for next time’, but also partially speaks to a perception some students have that unless there is someone stood at the front ‘imparting wisdom’ then whatever is happening in the classroom isn’t ‘teaching’. I talked them through everything that had happened this week:

  • Monday I’d given them several pages of notes to read, some links to tutorials to look at, and some videos explaining the theory for them to look at during the week
  • Tuesday we’d met up online, done a lot of Q&A, I’d done some live coding demos, then re-introduced the notes and self-study exercises.
  • Wednesday having asked students to send me their outputs from the self-study exercises once they completed them, I spent a couple of hours responding and providing feedback where there was something I could comment on
  • Thursday we’d got together in person and I’d gone through some group feedback on the exercises, then set them a group challenge to work together to put everything they’d seen this week into practice. This was purposefully a hard task to get them to discuss the ideas, seek out further information, and start learning how to expand their core knowledge themselves, with their peers (and with myself and the TAs as backup for when they got really stuck)

Each one of these things is an important part of the ‘teaching’, and I would estimate it all adds up to at least 15 hours of ‘learning’ for the students over the week (and that’s only for my part of the module). Somehow though, because I spent only ten minutes actually stood in front of them talking, it’s not ‘teaching’?

More work to do then on the expectation setting and explaining the process…

riding or running or that

Did Parkrun at Grangemoor on Saturday, took a minute off my last time there. Still some way off a PB, but the fitness is improving

Tried a new version of the ‘old’ commute. As suspected getting through the city centre is tiresome, but it was still ten minutes quicker


Oh my heart. LOVE THIS


October 1, 2021

Right, back to this then is it? Well, it’s been quite the week, so why not celebrate with a weeknotes…

This week was induction week, when all our new students officially enrol with us for the year, and we get to meet them for the first time. For years induction week was my own personal hell, back when I was running operations for the Postgraduate courses in the School, at a time of unprecedented growth. The logistics of trying to get a few hundred students to meet everyone they need to meet (and each other), get the information they need and prepare them for the teaching that will start next week is a real challenge. These days I get to sit back and the organisation is handled by others. Having been so involved in the process previously its one of those difficult times where having moved to another level of management you have to work very hard not to start butting in to your ‘old’ job; treading that fine line between offering some advice from your own experience and micro-managing tasks. Fortunately the team running induction are incredibly capable, so my involvement was limited to signing on to a few Zoom calls[1] and saying hello to all the new and returning students, which was a pleasure.

Outside of the School induction, we also got to meet our new batch of CompJ students. Once again we’ve got a really fantastic mix: some working journalists, some people just out of university, some techy/CS types. They seem like a good bunch and I think they’ll keep Aidan and I on our toes. One of the highlights of their induction (for me anyway) was the Data Walk on Monday. We took them out of the building and around some of the city, stopping off in a few places to consider different elements of data collection. Of course, this being Cardiff the weather fluctuated wildly from gloriously sunny to absolutely chucking it down, and we went from happily strolling along the river in the sunshine to sheltering under a fir tree in the park, desperately trying to stay dry in a downpour. A brilliant introduction to the city, and probably a moment the students will remember for a long time!

This week was also the first week that we were able to really start using our shiny new building. I spent my first full day ‘in the office’ since early 2020. I saw a lot of people, some of whom I’ve not seen for ages. I was also reminded of the dangers of working in the office when I was interrupted at least fifteen times by people dropping by to ask seemingly random questions. My office setup is also not quite right; my new standing desk is lovely, but the webcam and microphone I’ve got at work are nowhere near as good as the kit at home, so I may need to upgrade. Also my work laptop is still rubbish, so thats going to need sorting soon, I’ve spoiled myself by using my personal macbook for work for too long.

One of the more interesting things about the new building isn’t the lovely teaching rooms, or the socialising, or all the collaboration spaces, but the commute. I cycled all the way from Penarth to the new building and there were only two small sections where I was mixed in with road traffic. The rest of the route is entirely segregated cycle lanes. It makes for a much less stressful journey, even if it is a couple of km further than the old commute. I need to try out a route that’s a bit more like the ‘old’ commute across the barrage, but as ever it’s that tricky ‘getting across the city centre from south to north’ that makes it difficult. Oh for a cycleway straight up the Hayes…

Overall it was a positive week. Outside of induction events and prepping for teaching next week I had a lot of meetings where we talked about the future and things that we’re going to do this year and they made me feel like we’re moving in the right direction. Onwards.


New PSB album hasn’t grabbed me the same way as Every Valley did, but this is great

  1. We’re doing a lot of induction online because our shiny new building wasn’t quite all ready at the time we were planning, and we didn’t want to risk more chaos! ↩︎

Embedding an external website in Blackboard

October 21, 2020

Imagine that you’ve written a set of course notes using your favourite static site generator, and that you’re using a modern development process to automatically check, compile, and publish that site to the web. Wouldn’t it be great if that automatic build process could also automatically update the course notes in your University’s VLE?

It sure would!

But it can’t. At least, not if you’re using Blackboard, like us. Or at least, not if you’re using Blackboard in the way we’re using it.

But wait! All is not lost. If we’re using our super duper modern dev process to deploy to a securely hosted website, we can simply embed that site inside our VLE! That way any updates we push to the course notes will automatically be reflected in the VLE pages for our module. Our students can access the notes either through the VLE or by visiting the external site itself, and everyone wins!

Embedding in Blackboard

Embedding an external website in Blackboard is actually surprisingly easy. First, we need to find the place in our module where we want to put the notes, click ‘Build Content’ and then ‘Item’ under the ‘Create’ section:

Then we want to click the ‘HTML’ button on the content editor:

create a content item

In the HTML editor that opens we’ll want to add an <iframe> that embeds our notes pages. We can set the width to 100% to make the most of the space, and set the height to whatever you think makes sense. Remember that people need to be able to see content to read it, so I’d advise something bigger than 0px.

create a content item

Update your content, submit the new content item (yes, you’re not imagining things, that really was two clicks to accept your changes where one would have done …) and voila! One website embedded in your VLE. Students can easily find your notes, and you can easily keep them up to date using a super smooth and simple build process, rather than whatever editor and process your VLE demands of you:

create a content item