Weeknotes 06×02: Love will tear us apart. And by LOVE i mean UNSTRUCTURED ATTEMPTS AT PRIORITISATION.

Structure versus creativity and a quest for micro-strategy

Last week I wrote about priorities and scheduling being the hard things in management. Steph replied, sympathising and noting that planning and headspace are useful:

stricter management of my work-in-progress list, WFH more and starting to use OKRs to prioritise important, measurable things seem to be helping. I reckon a lot is about state of mind: what can only you do?

This led me to think about the relationship between the “structure” that prioritisation and scheduling seem to somehow inherently imply, and the act of management as a creative process:

another (meta) thing I struggle with is striking a balance between structured mode and “creative” mode. Do you find you get bogged down in being strictly structured?

There is a definite yin-yang, back-and-forth here between setting restrictions (“the hardest part of setting a strategy is deciding what not to do.”) and boundaries (time, people, etc.), versus letting thoughts flow – which, in reality, means being empathic and using what you’ve observed to solve problems, or to set a path to solve those problems.

At OCSI, there are certain structures we’ve “solved”, and certain ones we haven’t and which still fall into ad-hoc processes. Our sprints are pretty smooth now (and I’m always happy to talk about this to people if they’re interested in how we do it), but our prioritisation between products still needs a lot of work.

Similarly, as my own work jumps between so many different strands, and I’ve never been able to come up with that micro strategisation – all the past year, with mini post-its, and Trello boards, and notebooks, and to=do lists – these are all just mini strategies, disposable like spring breezes, but capable of establishing entire shipping routes.

Having come to this realisation, I think it’s doable. I think I can draw on the other prioritisation frameworks I’ve used to get some kind of micro-process in place – something small that can deal with whatever life throws up but that, most importantly, lets me be creative within its bounds. Something that keeps me in check in order to set me free.

Freedom is slavery

The many worlds I’m involved in really hit home last week – although oddly not for the usual reasons. Generally, it’s easy to get stressed when there are lots of plates spinning and all are shouting to be looked after.

Plate face

However, last week was this weird mix of NOTHING URGENT plus HALF TERM. I had plenty to be getting on with, but no real sense of direction, and aware of normal homelife routine being thrown as well. On top of that , it was my second week of working 2 half days, instead of having 1 whole day off, and my brain kept telling me that THIS IS NOT THE DAY YOU THOUGHT IT WAS.

So by Wednesday (oh yeah, and hours of pancake cooking on Tuesday)

cat, pancakes,bwaaaaahahahaa

…by Wednesday I was pretty scattered, drifting a bit. Some lunch plans didn’t come through, and I hadn’t really worked out my week. I wouldn’t say I was a mess, but I felt very … unstructured. Which had a bit of a knock-on effect, I think. Not guilt, per se, but some sort of weird dissatisafction that must, MUST come from learning how to be “productive” over the last few years. All this GTD shit – it’s not healthy I tell you. Feels weird to relax.

So this week (whoa, spoiler!)

this post is as dark as the night on the far side of the moon

…this week, I’ve made sure I’ve done two things:

  1. work out the smallest possible amount of work each plate needs to move it forwards
  2. assigned a day to each micro action

Hey look, more micro stuff! I should turn this into a theme or something.

Mmmmmmicro Machines, GET OUT OF MY BODY

Oh right yeah weeknotes nearly forgot sorry

So despite all the moaning, I actually got some good stuff done. Liiiiiike:

  • Cemented the roadmap for Hive Pixie in place, by writing it out in lots of bullet points, and turning it into proper Epics and Stories in Jira. Chatted to Emma (senior researcher) about updating reports, and I may have to dig out some code I wrote ABOUT 8 YEARS AGO. That’s from before #son1 was born.
  • Had a nice 1-1 catchup with Kim (marketing lead)
  • Had a nice 1-1 catchup with Gregor (sysadmin), which was also good because we now have a regularly-scheduled meeting in the calendar, which can be difficult when people are part time but this feels like a goo result coming out of a recent Annual Review. Calendars are amazin.
  • Caught up with Kim and Joel (user support) following on from conversations with some users about some upcoming mapping features – this was good because we don’t do a lot of ‘formal’ user conversations up front, and so we didn’t learn a huge amount of new stuff, but it fees like it’s been really good practice to generate questions, get in touch with people, and then bring the results together. I do like this kind of process practice.

Yassss, process

  • Sprint, despite wearing at least 3 different Product Owner hats
  • Being in a fruitful developer-based retro for our use of VMs, which are coming along nicely
  • Ohhhhhh, probably other stuff too, you know how it is weeknoters. I dunno. I could be making all this up, or it could all have been some whaced-out dream for all I know.

Anyway, next week I’m going to write about empathy and cognitive load, if I remember. STAY TUNED.

Other stuff what I’ve wrote recently

Also been

  • Reading M Train by Patti Smith, and therefore also getting into the Patti Smith back catalogue.

Are Data Skills Equally Distributed? Building a Society with Data at its Heart

I wanted to pitch a session at #ukgovcamp18 recently. Previously I’ve chatted a bit about the usability and accessibility of data – that is, how can we make government data more user-friendly, and open it up to the more people? This year’s session carried on this theme, but was more focused on the skills we need to collect, process, and understand data – not just consume it.   At OCSI, we collect a wide range of data about inequality, but nothing about the skills we actually employ ourselves. Isn’t it time we stepped outside of our ivory tower?

Somehow I managed not to fluff my pitch, and proposed a chat entitled ‘Are Data Skills Distributed Equally?’ It got a few raised eyebrows during the intro, which I take as a sign that people are interested in the topic. And sure enough, the (post-lunch) session went really well – discussion was lively, diverse and productive – you can see the ‘official’ session notes here.

My original aim was to talk a little about how to measure and track data skills as a nation, because that’s the kind of thing we like to do at OCSI. But the conversation rapidly turned into a more fundamental one – What do we mean by “Data Skills”? What are the blockers to learning them? And how can we unblock these?

I came away from the session with three main observations:

1. ‘Data’ is about a broad mix of skills

Successful data projects rely on a potent mix of people, as well as skills – but this is still pretty murky. We can’t think of Data as something that geeks do – from identifying use cases to designing and testing stories, I’d say that there is something that anyone and everyone can bring to a data-related project.

To be fair, I don’t think there are many people who would say that ‘doing data’ boils down to a single skill. But too often, we simmer it down to a particular job, role, or technology, often depending on the audience we’re talking to. Remember that whole thing (wow, 9 years ago) about ‘Statistician’ being the sexy job in the next ten years? The rest of the original quote was actually more important: “statisticians are part of it, but it’s just a part.”

Similarly, fascination with trends and novelty means we’ll often hone in on the latest technical movements, such as ‘Big Data’, or even ‘common APIs’. Each, in its own circle, is great. But you can’t expect to follow a trend, get one skill or API in place, and sit back to see magic happen. Efforts need to be more expansive and coherent than that. And to do it *clearly*.

So we need to keep pushing for a better, more wholesome picture of what ’Data’ actually involves. Let’s broaden this out.

Wouldn’t it be great to…

  • Get a conversation going on what combination of skills is important when working with data
  • Start giving equal value to each of these skills
  • Identify where there are gaps in skills which could be filled to make things better?

2. ‘Data’ is dependent on a bunch of other stuff

This ‘network of skills’ is actually more complex than we would like to think. For many people ‘in the know’, it’s too easy to *assume* that data skills are easy. Someone in the session highlighted that there are severe obstacles to entry in the field, such as basic access to computing, or decent broadband, and if we don’t address *these* blockers, then we’re not going to get anywhere very quickly.

I’d also argue that there can be very real *cultural* blockers, which can prevent skills from being established and supported. It’s really important that we break down stigmas of who ‘should’ be able to do what – that we start establishing data skills as something to be proud of, no matter who does them. The public perception of the *value* of data needs re-examining – it needs to be seen as something that “we can do”, not something that “happens to us”. We need to get punk about data.

Again, this should be something we can start to elucidate, and make more transparent. If we can identify the fundamental obstacles, then we can go back and address them first, before expecting people to just jump in and get on with it.

Wouldn’t it be great to…:

  • Start highlighting the things which are blocking broader takeup of the skills above
  • Have conversations with people who are starting out with data and evidence, to see what their needs and pains are
  • Work with people to actually find ways to remove these blockages?

3. The network is everything

The data movement can no longer rely on well-meaning individuals and skilled, lone coders. Again, I think this is something we all kind of know, really. But it’s easy to be an isolated individual when it comes to making an effort – I’ll reluctantly put my hand up and admit I’m guilty of this, because busy and fragmented and all that. We work as a team at OCSI, but outside of that, it feels like I’m an individual, vaguely thinking I can build tools to empower the world, all by myself. It needs more than that.

By deliberately placing a focus on teamwork, and by tying this into a clearer vision of the skills-mix which really push data forwards, we start to shift the debate away from smart individuals, and more towards communities that need mentoring, and groups that need support above and beyond tech skills. We start to see the *connections* between these individuals as more important than the individuals themselves.

Wouldn’t it be great to…:

  • Encourage a more diverse approach to data skills networks, bringing together a wider range of backgrounds
  • Reach out to other groups which need similar support
  • Help groups to learn from other groups


I’ll admit I’m a little bit stuck here, which is one of the reasons this post has been delayed. I have some initial ideas, but maybe others are already doing this? Maybe I’m just reinventing a wheel that all the other open data groups out there are already inventing, and successfully?

Maybe the initial starting point is these two questions:

  1. What can *I* realistically do, and how can I use my position – where I’m already within a network of data professionals – to push data skills outside of their existing “comfort” networks?
  2. Who should I be talking to as part of that effort, and how? Is that a series of personal conversations, or is there scope for some larger, ongoing “group” to move this on?

More to come soon, but if you’re interested, please do get in touch! I have a list of people from ukgovcamp, and figure we can start to build the network to build the network…

Weeknotes 06×01: What are the two hard things in management?

Revamped opening credits. A dischordant mix of lacrimose strings and upbeat techno. The warning message about scenes of flashing lights jumps onto screen with a flicker of brightly pulsating whiteouts. It’s clear the scriptwriters have been replaced with something cheap at the last minute, but nobody is really sure if it’s by a real human or by a bot. Ratings plummet.

Episode 31. Series 6 is here. I’ll try not to jump the shark.

Last series ended at a busy time. January was the month to catch up on all the stuff that didn’t make it into December because December was so busy. Then UKGovCamp happened and my head was full of ideas which I’m still sitting on, but with good reason and intent.

The other day I was reminded of a quote from Phil Karlton: “There are only two hard things in Computer Science: cache invalidation and naming things.” I’m not going to launch into a computer science lecture here (phew!) but my own attention is simultaneously so scattered and so trying-to-get some structure that I wonder if There are only two hard things in Management: priorities and scheduling.

I started out last week by trying to grapple with having a lot of Bulky Plates spinning at the moment. Sadly these are all long term plates, which means I’ve probably taken too much stuff on over the last few years, foolishly. Looking back over the weeknotes, FOCUS and CLARITY have been the twins to haunt me. Last week wasn’t so different, but their triple sibling, PROGRESS, leaped out and demanded to have its photo taken as well.

I spent some time Monday morning working out all the strands I needed to make progress on. Then I blocked in time to look at the them. This was to be my new organisational regime (Spoiler: Haven’t done it this week yet. I need to make time to look ahead and make time…)

As it’s the start of a new series, here are my current big strands that are pulling me in different directions like a chlidren’s party at Legoland:

  • Team stuff: 1-1s and other people’s careers, which I love but needs some dedicated thought time
  • Tech stuff: Planning next steps and making sure it gets prioritised/done, which is great but needs some writing time and hand-holding
  • Product Ownership on Hive Pixie: Planning out a 9-month roadmap and turning it into something I can follow and communicate easily, which is a brilliant challenge but always more work than I expect and takes a lot of writing and re-writing and drawing strands together which is not very easy in Jira
  • Our own ethical impact: One of our year’s company strategies, which I didn’t originally put forwards or vote for, but it made sense for me to pick up and actually I quite enjoy it, but does seem to mean starting an entire framework from scratch like a mini research project or something, and needs lots of writing time so that I can understand it so that everyone else can understand it
  • GDPR: Which is a whole legal thing apparently

The common threads here are:

  1. I am organising priorities and setting aims, paths and plans for most of this work
  2. A lot of that organisation seems to involve writing up documents and turning things into lists
  3. (Mostly) everything is pretty fragmented with not-a-lot of overlap, and context-switching could potentially kill me

Hard Problem 1: Priorities

What do I/we do next? This is the question that keeps stabbing me in the brain. Everything is continuous, a lot of it is reactive, and even the most ardent of intent isn’t enough to convert the wishes of unicorns into a minimum viable spec.

I think a lot of my cycle looks like this:

  1. Work out what other people’s priorities are
  2. Suggest and discuss a common set of priorities
  3. Do this across multiple things, and so have to do #1 and #2, but for my own work
  4. Write everything up in a list style
  5. Work out all the priorities alongside each other, plus with everyone else’s priorities thrown in.

Basically I may have turned into a Priority Machine. This concerns me.

I would dearly, dearly love to automate this as much as possible – on a personal basis, and on an organisational basis. Please do get in touch with any thoughts on how to come up with a system that takes a few judgements and turns it into a good-enough todo list. I’ve seen the Eisenhower Matrix, which is great for getting rid of stuff that’s not important, but doesn’t help when you’re filling up the top two boxes with the great stuff.

I’ve tried spreadsheets for specific projects, which is great, but I need something much more lightweight for personal decisions. I need tips, suggestions, resources for better ways to pluck priorities out of the air and to shove things into more bullet point lists of their own doing.

Many people try to attack one person all at the same time, from the hit comedy kung fu movie, 'Kung Fu Hustle'. Probably.


Hard Problem 2: Scheduling

So I started last week by blocking in some time in my calendar for the things that I wanted to “MAKE PROGRESS” on (this was my weekly aim, after being caught up in various large tech efforts over the last few months. I’m playing catchup.)

THIS ACTUALLY WORKED QUITE WELL. It meant I had everything laid out for me in advance. Not all of it worked – Thursday and Friday, I didn’t get round to the GDPR stuff because the earlier blocks have time had led to some things needing further work. But the lesson here is that the method works, but just not to that scale. I tried out having 3 focuses for a week some time back, and I still think that’s a good number.

Plus one of the feedback comments from my Annual review recently was that I’m often jumping between things, and it might be useful to be clearer about my availability.

So perhaps that’s something I can address this series. Know my limits. Plan out my own time more. Stick to it.

One of the difficulter things I find about that is balancing the work I’m doing alone, and making time for others. Both are important, but the sheer fact that there are more other people than me in the world means that there’s inherently less weighting for the alone stuff.

Coincidentally, I’ve also started a different working pattern this week – doing half-days on Tuesday and Friday, instead of taking all of Tuesday off. I’m planning to work from home every other Friday, so this may be a good time to book in the “working alone” stuff specifically.

In fact, a lot of what I do comes down to office rhythms – certain people are in on specific days, and certain meetings happen at regular intervals. Maybe I should embrace this more, and plan my efforts around it more consciously. Go with the flow. It’s what a sage would do. (Or quit work and live in a mountain. But we’ve only got the Downs to escape to round here.)

Kung fu on the mountain

The product roadmap is COMPLETE, I tell you

Hard Problem 3: Knowing when to stop

I’ve noticed this is something on my radar. A lot of planning can be quite creative, if you let it – knowing when to put the brush aside and say “enough is enough, this will be OK” is hard.

So I’ll stop there.

Scrolls crash from shelves from the hit kung fu movie 'Hero'. Like a boss.