Monki Gras 2014 – Part 2: wherein I learn about confidence, design and Italian cheese

Kent Beck

Working as a coach for new(ish) programmers at facebook, and doing so remotely, it of course made sense for this talk to happen via a Google Hangout. It was surprisingly successful, and this was another topic close to my heart as a new(ish) programmer myself. What did I learn? Coaching will rarely teach you technical skills, it will instead teach you how to use your time effectively, avoiding “misplaced efficiency” and how to build up confidence. I’ll be putting in an order of one miniature Kent to sit on my desk.

Greg Brockman - Building better software through Mathematics 

Cor. Just look at that title. Greg made my maths degree hit me in the face like nothing in pure mathematics actually can because it doens’t exist. Suddenly I remembered the definition of the determinant and pined for a library to lock myself in over night with nothing but a book on group theory and my own tears (and a history fresher complaining they have too much work). It was good to hear the two disciplines of maths and software engineering compared. The overall outcome was that computers can do brute force well and humans can do conceptual thinking well, leading to the soundbite “design software down but build software up”. #maths4lyfe.


#monkigras sketching. And a log with a moustache.

— Ed Moffatt (@EPJMoffatt) January 31, 2014

Phil Gilbert – IBM Design a year on (I’m not sure what the actual title was) The last talk nicely leads onto design overlord Phil Gilbert’s talk, updating us on IBM Design Thinking since his appearance at last year’s conference. Things are going well it seems, a shiny design lab in Austin, executives attending design camp and they even have their own card game (the “I know Steve Mills – lose a turn” card being my favourite).  If it weren’t for my complete lack of creativity and design knowledge I would consider a move over into the design team – they are having a hard time in Hursley but hearing the awesome progress being made in America gives me hope.


IBM making products into a game internally, design w/ constraints -very cool ! @philgilbertsr at #monkigras

— Narinder Singh (@singhns) January 31, 2014

Leisa Reichelt - Some Ass-kicking on How to make organisations care about User Experience

While I’m loitering around the design theme I shall give a nod to Leisa who delivered an excellent talk on UX. She needs to come to Hursley and kick some ass here too. She revealed that members of the user researches are embedded into the teams at GDS, and all members of the team are expected to observe some user research (I think it was around once a month but I can’t actually remember). GDS are taking UX seriously and it shows. I took part in some GUI play/user research a few weeks ago and the difference between how that session was led compared to the clips we were shown by Leisa of their research sessions were quite different.

A word about cheese

I had to do it – O Toma O Morte were there again and I was force-fed cheese off a sword again and I loved every minute of it. I also experienced some of the most middle-class FOMO in the form of having to put my tub of edamame beans in my bag because I couldn’t eat them at the same time as carrying my drink and eating chocolate truffles. My point being the evening entertainment was as glorious as ever. As was the doorman’s outfit (Will.I.Am meets coachman?).


Monki Gras 2014 – Part 1: wherein I learn about bugs, data science and ecology

After a few weeks of letting the cheese dreams settle down I’m finally ready to ponder over this year’s Monki Gras. A fire-hose of information, loosely based around the theme of “Sharing craft”, and a great opportunity to meet other developers and tech nerds. So, here are some of the highlights from this year’s celebration of all things craft:

Rafe Colburn - What Bugs Taught Us About People

Currently sitting in my bay, I am writing this illuminated by the warm glow of the “3″ (elevated to “2″ mid-post) on my team’s DEFCON (Defect Condition) indicator, Rafe’s talk was particularly pertinent.


His story started with 1,000 open bugs and an annoyed service team (I wonder what that feels like). Bug rotations were introduced, where developers were asked to spend one day a month going through the backlog and fixing what they can. The benefits of this system seem obvious: the opportunity to see areas of the code you haven’t worked on before, the chance to work with people outside of your usual team, and of course a happier service team (which usually correlates to a happier customer).  But that damned human nature thing kicked in, and Etsy found that when you tell people they have to do something, suddenly all developers have a barely repressed inner fourteen year old that doesn’t want to do anything you tell them to do. More teenage traits developed: laziness meant that all the bugs that could be categorised as “low hanging fruit” were closed first; developers didn’t mix with new people, instead staying in set groups; and the paradox of choice meant even if developers did attend their bug rotation they struggled to choose a bug to work on. Bug rotation 1.0 failed and it was time for a re-think.

Bug rotation 2.0 went to university and moved back in with its parents, but despite their disappointment it was far more mature for the whole process. Firstly – bug rotation day was now optional, and surprisingly developers actually turned up! Service prioritised the bugs so that those super keen developers could make an informed decision on what to spend their time on, as well as a shake up in the management of the entire process. Bug Rotation 2.0 was a success!

Sean Owen - Factory vs Lab in Data Science: Perspectives From the Field 

A great view into the world of data science. Sean’s view is that data science is 80% engineering and 20% maths, and his talk was all about the engineering of data science, attempting to skip over what a data scientist actually is (“a statistician that lives in San Francisco”). The craft is something of a cottage industry and need to be made more operational. Consider moving your data models from an offline model to getting them deployed online.

It was a topic quite close to my heart, especially with the debate on whether a mathematician can become a good programmer, and whether a programmer can become a mathematician starting up during the questions. As a mathematician turned programmer it is a question I ask myself quite regularly!

This talk also allowed one Monki Gras attendee (who will remain unnamed) to go viral for the first time, and allowing him to refer to himself as the new resident social expert. Exciting to be a part of such a transformation.

Steven Citron-Pousty - Biological Metaphors for Developer Craft: How to Identify an Evangelist In The Wild

I just liked this talk – also forwarded it onto a colleague who is entering the world of developer evangelist (although refuses to give himself the title) as it was so well presented.

The Gentle Author

This section was totally unexpected and absolutely wonderful for it. The Gentle Author is a very appropriate title for the author of the Spitafields Life blog, who graced us with a reading of one of his most popular short stories about Maurice Franklin. Once of my biggest gripes  about conferences is the insistence of the attendees to constantly tweet soundbites and take photos of slides (similar sentiments towards music concerts), as I don’t believe anyone is actually paying attention to the content of a presentation. There was nothing to tweet or to photograph during the reading of this story, which meant I finally saw the majority of the audience with their heads up, listening intently to the story, giving the speaker the deserved level of attention. Laura Cowen was luckily enough to receive a free copy of one of The Gentle Author’s books which I had a flick through the other day, and after about 15 minutes had to stop myself as I was about to dedicate the rest of the day to reading it from cover to cover.  Really, I could write an entire post just on his talk.

ThingMonk Day 1

Thanks to Ed Costello, via Andy Stanford-Clark and Laura Cowen, I managed to get my hands on a ticket to the latest event run by Monki Gras organisers Red Monk: this one going by the name of ThingMonk (do you see the theme yet?). This event promised to host talks on all [Internet of] things connected.  From demoing how to fly drones with fruit to sharing experiences of manufacturing tiny printers with beards;  the speakers walked us through the protocols and transports that make up our Internet, and brought us back to the physical world to present the Things themselves.

As far as Monday mornings go, the morning of 2nd December was about as good as you could ever hope for: having to write a note to a traffic warden explaining the finer points of why a parking ticket isn’t the solution (properly testing the ticket machines on the other hand), and arriving on the platform without a ticket as the train is already kicking out commuters. Despite the set backs I met up with my fellow conference goers and we blundered our way into Shoreditch.  The morning was forgotten over an espresso and a pastry, and the first day of ThingMonk was underway.

Monday of the conference was a sort of hackday/IoT workshop. I shamelessly joined the n00b table to listen to the first talk of the day – an introduction to node.js for IoT.  After getting my laptop (which is not archaic, just big-boned thank you very much) set up with the latest and greatest software, I learned me some node (, and with the help of Nuno was shown some of the magic (read: dark art) of JavaScript, specifically that I can pretend I’m not using JavaScript at all and am instead using a functional programming language.  It was quite early on in the day that James (Mr ThingMonk himself) realised that “although we have internet, we don’t have many things”, and an emergency crusade to Maplin was launched, returning with actual sacks full of Arduinos and cables.

Starting the conference

My spirit animal – I have no idea what I’m doing dog

Now equipped, we were able to start playing. The Arduino, with its bare minimum design and all its innards on show, had always seemed like it came with pre-requisite electrical engineering degree – not having one meant I had never gone near the Arduino. Thankfully ThingMonk was the perfect environment to try this kind of stuff out – with the attendees being a mixture of experts and newcomers, asking questions was easy and getting answers was even easier.

Asking questions and getting answers

Asking questions and getting answers

After the usual standard of lunch to be expected at RedMonk conferences (amazing), the afternoon continued morning’s playtime, with a couple of lightening talks thrown in to mix things up. This nicely set the tone for the Tuesday session, giving us a brief introduction to both speakers and topics that would be reappearing during the next day’s talks. Becky from Codasign did a talk about their work with non-technical organisations, and afterwards I joined her workshop on using the Arduino. This was great – she stepped us through the basic exercises and got us set up with an array of components – potentiometers and pressure sensors among them.

Going back on the earlier comment on having internet but no things, the power decided to go down, leaving us with plenty of things and no internet. Undeterred (learning is just too fun), we continued, noticing that now was the perfect time to put the light sensors to use. By candlelight/backlit by our laptops, fellow IBMer James and I hacked together a system that turned on an LED when it was dark, and turned it off when it was light. Simple but extremely satisfying!

Arduino by Candlelight

Laptop batteries and my caffeine levels flagging, we decided to leave the hoard of guys flicking every different combination of switches in the fuse box to their fun, and headed to the pub for a burger and some quiet reflection (mostly how can I make this my day job?).

Thanks to Jon McNamara and Tom Raftery for the photos.