Weeknotes s01e09


A short week — I had hoped to take it as leave but agreed to cover some stuff for Charlie. Also the last episode of season one of my #weeknotes. I’m still deliberating over whether to commission a second season. I’ve enjoyed doing it and think it does me good to both take the notes and then write them up but as I’ve said before — numbers and pictures are my thing, words most certainly aren’t — so I do struggle with parts of this process.

I’m looking forward to catching up with the other weeknoters on my first week back after leave and I’m pretty sure that will seal me continuing with it even if I do alter the format. Last week was all about the tv/film gifs this week, music.

I certainly can Craig David… but its only three and a bit days

I sent a formal offer on Monday — to a new G7 data scientist who I hope will be joining my team — I’m excited to get another piece of the jigsaw.

I got a formal offer on Tuesday — for my job — although still provisional. This was a relief as much as anything. Still a few things to sort out but I can finally start to plan properly. The last few months has been full of personal uncertainty. Hopefully I’ve sheltered the team from this. I’m pleased it was sorted before my leave.

I forgot to write notes on Wednesday — I’m very much wishing I had and also that I’d written this up on Wednesday — so this may be a little brief in places.

and on Thursday — I did a little work — but I did chill after that.

Week in brief


I got through a few emails before my first meeting of the day. I still have too many emails to process. This is definitely a consequence of not working from home for the past few Wednesdays. I need a way to manage this better.

My first meeting was a high level planning meeting for Manhattan with Becky, Daniel, Gayll and Alaine. Actually this was about separating activity for Manhattan as a product for Employer Engagement from activity more closely related to the strategy work I talked about last week. Becky is doing a very good job in keeping things separate.

The team wanted to check there was no stray information stored in my head before I went on leave — plus for me to outline my thoughts on our approach to the strategy work, of which Manhattan type stuff will form the backbone — which i did. I do love a white board. We talked briefly about applying agile practices to R&D/analysis/data science. Its a topic thats rears its head every now and again — largely when we have to balance a few things in the team and split people’s time between products/projects — we tend to operate Kanban for this stuff. We’re still learning, there is always room for improvement but when in doubt pragmatism wins. Becky has Manhattan work to manage, Daniel is also working on Mercury but the discovery activity for the strategy work is important to scope out and lay foundations for.

I will arrange more formal whole hub planning and prioritisation sessions when I return from leave — we will need two or three days. I want to define a broad work plan for the medium to long term based on the three parts of the vision I mentioned last week and then to break that plan down into well defined short term tasks. Plus its important to make all the other activity that people are involved in visible. I’m looking forward to starting this process.

hold your horses boys, I am ready — but I need a holiday first…

I called Eddie — Kit had suggested we talk. Eddie is a data scientist at Ministry of Justice Digital and broadly speaking was interested in data visualisation and embedding data science.

We had a good chat. I ended up covering some of the conversation I’d had with Philippa the previous week about digital and data and we spoke in a little bit of detail about the work taking place in our respective teams. I said I would share some resources and information on data visualisation. We’ll speak again and meet soon.

Charlie called me — he is/was supposed to be on leave but now that all the data scientist interviews are complete — there were a few things he wanted me to sort out with regard to some of the posts and offers.

Neil T (Data Services) popped to my desk. Neil wanted my perspective/views on how a few pieces of work may or may not tie together. It was a really useful chat — effectively I talked about how a particular conceptually big and complicated piece of work could be broken down, what the components might look like, which bits were actually big and complicated and at a very high level what an MVP might look like for it.

I then went into another meeting with Neil, this time with with Becky, Alaine and Kevin H (who is responsible for security and operational control within Data and Analytics). This was about Manhattan and also the strategy work. Neil walked us through the work he’d been doing on updating the data for Manhattan and we talked about some specific elements of the data plus also covered two factor authentication and audit capabilities for the application amongst a few other things relating to data for the strategy work. I expected the meeting to overrun — meetings with Kev usually do as they necessarily involve addressing lots of interesting and often very nuanced detail. It did overrun and resulted in some quite exciting outcomes and some wait and sees.


London. Presenting at the Government Data Science Community of Interest which had a data visualisation theme.

I’d decided to travel to London early to do a little preparation for the presentation. I bumped into Michael and later on Sree (who was heading to London for her first day in her new team) at the station. Michael is a front end and user experience designer (and a new data visualisation convert). He used to work with my team and now works with Kit. We spent a lot of the journey talking about design patterns and components and how we can best work together. We spoke in detail about some specific elements of the products we are working on in this respect — more to come from this hopefully.

I parked myself in a coffee shop and had high hopes of prepping the material and then clearing some of email mountain. The presentation was mainly demo-ing Churchill and Greyhound however given the audience I wanted to put a couple of slides in for background and some gifs for effect.

I mentioned in episode 3 that some of the content I’d prepared for Civil Service Live rendered better in Safari than Chrome or Firefox. I was reusing some of this so decided to make HTML slides and load the whole thing up in Safari to allow me to move seamlessly between slides, demos and gifs. A foolproof plan, nothing could go wrong.

When I’m presenting, especially externally, I sometimes try to open with something about the particular date I’m presenting on, linked— usually tenuously — to the theme of the presentation. I find this makes people sit up and listen a bit more than hi I’m Ryan and I etc etc. So I prepared something for that.

Occasionally — although more rarely and saved for longer talks— I extend this concept to amuse myself and keep peoples’ focus. For example I gave a talk at the Royal Statistical Society North Eastern Group at Newcastle University on Gary Barlow’s 40th birthday where I incorporated 40 song titles from his back catalogue into the talk.

oh yes I did

During the prep Charlie — still supposed to be on leave — phoned me with another couple of recruitment things which needed a phone call and a few emails. By the time I’d sorted that out I didn’t have time to get my teeth into my emails. Although one I did check notified me that we now had access to some data we wanted to look at so I made a quick call in to the office.

I headed to the new Government Digital HQ in Whitechapel — fancy. I had a meeting first with Philippa and Louis H (Home Office) — it was their meeting but Philippa invited me to gatecrash during our conversation last week — it was good to hear Louis’ perspective on embedding data science invisibly into products and his experiences as a product manager.

Dawn arrived to check the set up for presenters and informed us it would be via Chromecast which doesn’t always play nicely with Apple products and obviously works best with Chrome. It became immediately apparent that my Safari from a MacBook set up wasn’t going to be straightforward.

Philippa kicked things off and Dawn talked about the Government Data Science Partnership which covered some of the content from her recent blogpost.

I was next up. I’d been chatting to the tech support folks during Dawn’s session but despite all the fiddling and emailing, my laptop wouldn’t connect. Then it did — but wasn’t quite right. The we tried an HDMI cable — same problem. While they were contemplating an agenda shuffle I moved everything to Chrome and got it working. All of this took quite a bit of time.

Given the delay I blasted through the presentation to try and recover the agenda timings — I had a train to catch and wanted to see the other presenters — I suspect it seemed rushed.

It was the anniversary of the launch of MTV. I had faffed around and edited an image to get what I wanted to be shown on the screen at the start which was this...

we can’t rewind we’ve gone to far

My intro linked the concerns and mixed attitudes of culture change — my tenuous take on the theme of Video Killed the Radio Star, the first music video shown on the station — with the work going on in my team. I’m not sure how it went down.

I did the demos etc — I didn’t feel it went very well as I was going quickly and probably forgot to mention a few things — but the feedback during chats after was positive and I got some good questions —even if I felt I waffled a bit with some of the answers.

As you can probably tell, overall I’m not sure I did the myself or the work justice.

During the break I had to get photographed for the GDS Digital people blog which Dan H had interviewed me for during episode 3. I felt very self conscious.

strike a pose…

The second set of sessions were from

Robin from Ministry of Justice discussing interactive geographic modelling using Javascript and D3. Robin and his team are deploying browser based maps to policy customers. Check out one of the demos here.

Tomas from the Alan Turing Institute talking about his work developing tools for open data-driven storytelling. This lets users easily create analyses and visualisations that are linked to the original open data source — making the analyses more transparent, reproducible, but also easy to adapt. Some neat ideas building on the what do you think type concepts developed by the New York Times and Office for National Statistics. Check out some of his early work here.

It was my first visit to one of these meet ups — I enjoyed the sessions and hope to attend more. I didn’t get the chance to speak to many people but Emma, who has recently joined DWP as a Data Scientist did introduce herself and we had a quick chat before I dashed off back north. I’ve been on a few calls with Emma so it was good to put a face to the name/voice.


Both the kids were poorly so I was a slightly later into the office than I’d hoped.

I caught up with Becky and Daniel about the data we’d got access to — they’d had a look — seems some really valuable possibilities — more to come on this.

Next I was covering for Charlie at a Data and Analytics planning meeting — one of a series. I read through all of the emails and documents to re-familiarise myself with the details. Paul Lodge, our Director was chairing. He talked through the state of play so far and I picked up an action.

I’d rearranged my regular catch up with Alaine and extended it to allow us to refine our broad communications plan for our private beta-users. Churchill will be available externally — to selected users — imminently and whilst Stuart (data product owner) and I are on leave Alaine is in the driving seat. Ideally we would have been at this stage before the end of July — and before the periods of leave — but matters outside of our control delayed this. Regardless, the team should be very proud when they reach this milestone, its been a brilliant effort.

Alaine and I laid out the path with a bit of flexibility to react and some decision points etc. I do love a flip chart. After that I got through the most pressing emails but didn’t get them all processed before heading off to meet my wife and kids for a meal.


I spent part of the day completing the action I’d picked up from the planning meeting and then slammed my out of office on.

That was my week.