#150 - 17 Dec 2022

Your End-of-year Review; Research on Team Collaboration; Research Application Managers Redux; New Zealand’s eResearch Future

Your End-Of-Year Review

You did great this year.

I wanted to take some time to personally recognize the work you’ve put in during 2022. You have been a terrific support for your team and the research your team enables, and I appreciate all of your efforts.

It hasn’t been easy, but you worked hard over the last twelve months. You’ve done well by your team members, and by the research your team supports. Enjoy that and celebrate that over the holidays. You deserve it.

We both know that our jobs as leaders and managers in this line of work are really challenging. There’s an incredible amount of work we could usefully do for our research communities. We’re torn between wanting to do everything while having very limited resources.

The work of our organizations starts and ends with teams like yours. They are the ones who get the work done, who support the research with software, with data management and analysis support, with functioning computer systems. Our first job is making sure they’re flourishing individually, and working together well collectively as a team, both for their sakes and the sake of the research they enable.

So I’m particularly impressed by the effort you’re putting in to learn how to better support your team. There’s a million things on your plate right now, and your bosses and louder researchers aren’t going to sing your praises for reading long newsletters or digging into blog posts or asking questions on slack groups or watching talks online about better managing team members and teams. But you know it’s important, so you do it, and I’m impressed by the new skills you’ve picked up.

This year you and I talked about how the patient tending to management tasks is like gardening, and I know you can still find that frustrating sometimes, feeling like you have managerial sensory deprivation compared to getting immediate answers back when you used to do more technical work you used to have. But you’re doing great here, continuing to show curiosity and collecting data by asking questions and making notes.

When working with individuals, I know providing all your team members with the direct feedback they need (and want) continues to be hard and feel a bit unnatural, but I really like the strides you’re continuing to make there. Clear feedback is a key piece making sure everyone has the support they want. We’ve talked about your retention worries that you (and all your colleagues) have, and though it may not feel like it, this will really help.

And of course with the end of year upon us, that reminds me that you and I were talking about helping yourself enjoy time off and team members grow professionally by using holidays to practice delegation, and otherwise helping others get ready for management. Your commitment to making sure your team members continue to grow is admirable — I certainly admire it — and the things you’ve been working on along these lines are fantastic (and will help with retention too!)

We both recognize the importance of managing individuals as individuals and teams as teams, and your dedication to making sure the team is working well collectively is really paying off. Things like retrospectives, acknowledging importance of internal knowledge sharing, and using process as a labour-saving (and variance-reducing) device has meant that your team is moving together more and more strongly, building on each other’s hard work. It’s just fantastic to see.

Working in our communities is really challenging, especially given our commitment to research. Accepting that we are vendors, and ideally professional services firms specifically - bundling up expertise and communicating services clearly is crucial to maintaining our focus, and you’re doing well there.

You’re also doing well with saying no, although I know that never comes easily. I’d like to think our talks about focussing on specific research problems rather than technologies, and discussing strategy and strategic plans helped, but I know that you’ve always been able to prioritize, and to see big problems and figure out clear focussed solutions. Your work collaborating with other teams to jointly address those research problems we were talking about is coming along really nicely; it lets the team scale their impact without necessarily scaling their effort. Great job.

And of course we should talk about your own professional growth, too. I’d love to see you continue to excel in your current role - or at least your current organization, taking on new responsibilities. But your training and your experience gives you superpowers and I’d be remiss if I didn’t make sure you were considering the different kinds of jobs that need your skills. Some of those job types didn’t even exist when you started your career.

Let’s talk a bit about the coming year. I’m cautiously optimistic about that new role you were hoping to hire for. We’ve discussed the internship program and how that could play a role. With luck our discussions about hiring - starting with a definition of success, designing onboarding plans accordingly, and actively recruiting - will be something you can put into practice.

We both know 2023 isn’t going to be easy. Once again, we’re being asked to “make do with less.” Once again, we’re charged with somehow supporting huge swathes of research within our institutions while not being given adequate direction about what we should be prioritizing or how we should do it. Even if we can hire, we certainly won’t be allowed to pay our new or experienced staff anything like the competitive salaries they deserve.

But given what I’ve seen this year, I’m confident you can shine next year, too. You have what it takes; you’re making strides in all the right areas; you know what’s important and you’re building your skills. I’m always around if you have a question or need some support - just email me or ping me if you want to have a call.

Congratulations again, and thanks for everything you’re doing for your team and for our community. Please accept my very warmest wishes for you and yours over the holidays, and may you have an amazing 2023.

All my best through this holiday season,


Managing Teams

The problem of management in two graphs - Ethan Mollick

Mollick collects some research about what’s known about people working in teams:

  • As team size grows with fixed task size, people individually put in less effort, but there’s wider collaboration, leading to better ideas
  • Teams that alternate between working together and time spent working individually come up with a broader set of possible solutions than either teams working entirely as groups or entirely solo
  • Collective intelligence of a team is limited by the emotional intelligence of the least sensitive team member
  • Teams that have good processes are more likely to have higher collective intelligence, because process let’s people asynchronously coordinate
  • Teleconference calls without video seem to promote more frequent syncing and have more even turn-taking than with video on.

Product Management and Working with Research Communities

Infrastructure and Related Services to Support New Zealand’s eResearch Future - New Zealand Ministry of Business, Innovation and Employment

This report is apparently from August, but just crossed my desk this week.

Coming from Canada, this document is a breath of fresh air - an admirably bracing look at an ecosystem for digital research infrastructure and skills which is internationally well respected, but is facing the challenges we all are as demands ramp up. In my own country, we tend to endlessly talk about problems key partners in the ecosystem are facing (or are, erm, causing), but I couldn’t imagine them ever being committed to any form of official document.

It’s written as a set of succinct, numbered items for easy reference. The primacy of people and skills over hardware come up repeatedly in the single-digit points, which is terrific to see.

There’s great stuff discussed here, such as the governance issues of current crown investments. The current structure is partner driven, and “the usual suspect” large partners are committed, but that causes engagement issues with smaller institutes, many of which are going their own way (“..prefer to rely on commercial or in-house facilities rather than pay for and actively engage in the centralised services.”) because they don’t feel they belong (“Club models work when the club members have near common interests, needs and ability to pay. This is clearly not the case across our eResearch community”).

There’s concerns about researcher-friendliness, cost transparency, data management, governance, and everything else that bedevils organizations in our line of work everywhere. It’s a short read and well worth reading. My complements to the people who contributed to this document, by writing it or consulting on it.

Point 16, about acknowledging and then moving past issues that we can’t do much about, really stuck with me - this is good advice for us all:

Some of those factors are essentially contextual [….] However, their existence obliges us to be careful to avoid adding further obstacles by way of poor choices and poor prioritisation.

Research Software Development

I mentioned Turing’s Research Application Manager role description last issue, and they have such a role available now, in the area of digital twins:

The Alan Turing Institute and its partners have invested more than £26m in digital twin research and innovation across a range of dynamic projects, including developing foundational theory and applications in the engineering, environmental and social sciences [….] The Research Application Manager (RAM) at the Turing builds and nurtures the connections with users of research outputs and brings back the user perspective to researchers and research engineers. RAMs work in a dedicated team focused on ensuring that research outputs such as open source tools are used successfully and sustainably by external stakeholders. The key goal for the RAMs will be to maximise the interoperability between different TRIC:DT outputs, as well as to ensure that those outputs are easily accessible and practically useful to diverse stakeholders across a broad partner ecosystem.

Again, I think this is a fantastic commitment to making sure that research outputs actually get used. In the social sciences this is usually called something like “knowledge translation” or “knowledge mobilization”; in medicine it’s “translational medicine”; it’s probably significant that we don’t have a widely recognized term for it other than the more specific “commercialization” in STEM fields.

Emerging Technologies and Practices

Just How Bad is CXL Memory Latency - Tobias Mann, The Next Platform

Interesting discussion about what (local) CXL latency is likely to be, although admittedly mostly by CXL vendors:

They claim customers can expect latencies roughly on par with accessing memory on a second CPU, one NUMA hop away. This puts it in the neighborhood of 170 nanoseconds to 250 nanoseconds [….] This is right in line with what other early CXL adopters are seeing as well.

This sounds like a pretty attractive result for plugging in memory expansion units into a PCIe slot in an era when DIMM slots and channels are getting scarce compared to what we’d like them to be.

Where the big wins from CXL would come, of course, are cross-node memory access, and the latencies there will of course be longer. But we’re not there just yet.


Fascinating - simulate breadboarding in the browser, including with arduinos running your own code, using diode.

The power of visual models is easy to overlook - riffusion uses the standard openly-available stable diffusion image generation model, fine-tuned on spectrograms of music, to create a “text to music” generator for sentences like “funky bassline with a jazzy saxophone solo”. Really interesting music-to-music transitions there too.

A crash course intro to GPT-3.

Sorting with vector intrinsics.

Getting better and better performance in searching the database from Have I Been Pwned. At the end they start talking about count-min sketches, which will be familiar to at least one reader…

That’s it…

And that’s it for 2022. Let me know what you thought, or if you have anything you’d like to share about the newsletter or management. Just email me or reply to this newsletter if you get it in your inbox.

Have a great weekend, a wonderful holidays, and good luck in the coming year with your research computing team!


“A diverse group of computer programmers toast the new year 2023”, Midjourney v4.  Off by a couple hundred years, but hey, it’s been a long twelve months.

“A diverse group of computer programmers toast the new year 2023”, Midjourney v4. Off by a couple hundred years, but hey, it’s been a long twelve months.

About This Newsletter

Research computing - the intertwined streams of software development, systems, data management and analysis - is much more than technology. It’s teams, it’s communities, it’s product management - it’s people. It’s also one of the most important ways we can be supporting science, scholarship, and R&D today.

So research computing teams are too important to research to be managed poorly. But no one teaches us how to be effective managers and leaders in academia. We have an advantage, though - working in research collaborations have taught us the advanced management skills, but not the basics.

This newsletter focusses on providing new and experienced research computing and data managers the tools they need to be good managers without the stress, and to help their teams achieve great results and grow their careers.

Jobs Leading Research Computing Teams

This week’s new-listing highlights are below in the email edition; the full listing of 212 jobs is, as ever, available on the job board.