I’m on my way back from Agile Cambridge, its the first Agile Conference I’ve been to and I really enjoyed it but the question on my mind is will anything I learn’t from the sessions change the way I work at Biomni?
Dave Snowden Keynote
This was a proper baptism of fire. Dave spoke at quite a pace about complexity and how it applies to software development. It was challenging and I can’t confess to understanding it all, initially I was concerned that while I was enjoying the intellectual roller coaster I was going to struggle to take anything away but my lasting impression, reinforced by others, is if we are to innovate we need to embrace the unpredictability of the complex adaptive systems (people) we are dealing with.
As our product at Biomni grows more sophisticated it appears to attract a more diverse set of users. We can’t hope to tie down exactly what our customers will want to do with it, but if we understand that our customers will adapt to use the system and adapt our system as they discover better ways of working we can co-evolve. Dave suggested that we should find ways of discovering those adaptions and continually optimise for them. There will be a wide range of adaptions and we can’t hope to satisfy them all as the product evolves but by monitoring a large sample of customers we can begin to understand trends.
Retrospective Dialogue Sheets – Alan Kelly
Since our Retrospectives have settled into a fairly predictable format I’m always on the lookout for alternatives. Alan has produced large dialogue sheets with steps around the edge that you follow and you write your findings in the centre. The main difference is that the sheet does the facilitation. We found it a bit restrictive but I think I’ll give it a try. I’m interested in our team creating their own sheet.
UX and Agile: How to do both and not lose your mind – Darci Dutcher
The talk was really aimed at the UX professional dealing with Agile and I think most of the attendees were Agile Developers wondering how to deal with UX professionals. The concept of bringing anyone onto the team that doesn’t work collaboratively alongside others working on the story freaks me out a bit but it seems many teams do design before as opposed to while the story is coded. Darci discussed various ways of doing it, I like the idea of a UX professional on the team but perhaps more as a trainer of developers in their context than someone who works separately.
Questions not stories: Build a business-value orientated team – Adrian Howard
The session started with a discussion on User Stories, the good bits and their limitations. Adrian raised the question, how can we show that a user story will solve a problem that brings business value? Well we turn the story into a question and devise a series of experiments to find out. It’s very Lean Startup. Adrian guided us through an exercise that involved writing a story, turning it into a question and devising experiments. Apart from reducing the cost of writing code that has no value the method shifts the developer mindset to be more business driven. Adrian emphasised the technique was particularly useful when starting a project but didn’t have experience of it in a more mature product, our project/product has been going for 10+ years and we have enough contact with our customers to know whether something will be used or not so perhaps its not for us but I’m going to look out for opportunities to create questions not stories.
Embracing Uncertainty – Dan North
Dan North introduced a concept that was new to me and I hope I’ve got this right as I don’t do notes. He talked about the 3-ages: exploration, stabilisation, commoditisation. To me this is relevant on 2 levels. On a project level Agile gave us some predictability of when we could deliver working software by ensuring we stabilised early and often, however with a focus on stabilisation and stakeholders desire for certainty are we inclined not to put enough effort into exploration that leads to innovation? In our development process at Biomni we’ve recently been spending more time on exploration and we have seen less predictability in return. After years of being predictable we have a high level of trust with our business people and they are allowing this and dare I say enjoying the increased innovation, but we still have to deliver.
On a story level Dan discussed doing an untested spike as part of the exploration and then when the customer is happy with it spending the time stabilising. Whilst this isn’t a new concept I think its something we should consider doing more.
Commoditisation is about making something more efficient, Dan used the example of automating a build process only after you’ve lean’t what you need and stabilised the way you do it. I think this also applies to that tweaking at the end of a story that really makes it slick or the refactoring of code after it passes tests, its a step that easy to forget if we consider satisfying our acceptance criteria the signal for done.
This was a workshop where we got to experience coaching somebody in something you have no idea how to do yourself, in this case we tried to coach each other in Juggling. Our coaching seemed to settle into a rhythm of Coachee trying, coach observing and questioning, stimulating new ideas in the coachee, a conversation then back to trying. It proved interesting as I’ve not really understood the role of a coach as distinct from teacher or mentor and whether its acceptable to wear more than one hat. When I questioned Rachel she said that it’s ok to share your experiences with other teams to stimulate ideas. This seems a more honest approach than trying to lead a team down a path that you know works with leading questions something that I’m sure must happen to some extent no matter how hard you try to avoid it.
Pleasing Customers: Some Considerations – Giovanni Asproni
Giovanni talked about the relationship between Agile teams and their customers and reinforced my belief that you need to work closely with the customer on stories and backlog rather than blaming them when you don’t get what you need. He talked about the importance of using the language of the business when discussing the value and cost of a story and not complicating discussions with technical issues. There were a number of other good observations about developers and customer such as: No matter what they say even if the say ask for quick and dirty customers always expect quality. Dirty is never quick.
All this +
- The usual stories of other companies that make me feel lucky to work at Biomni
- More confirmation that: All that claims to be Agile is not gold
- One set of Planning poker cards from Rally
- 2 sharpies
- Realisation that I learn much more doing than reading/listening
- New friends/acquaintances 2 of whom work in Cheltenham
Thanks to the speakers, organisers and sponsors and Rachel Davies for permission to use her photos in this post