DevOpsLive - what I learnt

Last Wednesday, I travelled down to London with my work bros Neil and Tiago for Computing's DevOps Live event and the DevOps Excellence Awards. Panintelligence were nominated for two ENTIRE awards!

Whilst we didn't win anything, it was awesome seeing Panintelligence on the nomination list amongst such behemoths as Lloyds, Vodafone, Yodel and BT.

What did I learn?

All the speakers shared some fab insights into their DevOps journey such as the difficulties they've faced and the rewards they reaped from implementing DevOps methodologies.

A few things I noted:

• Everyone needs to get into the DevOps mindset - this is always so important. If your Development and Operations team are always paddying at each other, you need to do something about that first!
• Monolithic management - the problems related to getting higher-ups to invest the time needed to transform into a DevOps organisation. Luckily there's an increasing amount of data which shows how beneficial it is, such as DORA's State of DevOps report which shows organisations who do full-on DevOps deploy 46 times more code than pre-DevOps, and they recover from incidents 2,604 times faster
• The importance of measuring DevOps - how can you say you do DevOps and are improving if you don't measure it in the first place?!
• Automation is good - the media seems to be framing automation as some evil force that's coming to steal our jobs, but it really isn't! The more you automate the boring stuff, the more time you can work on the interesting stuff
• It's a gradual journey - start small and don't try to make sweeping changes then wonder why it didn't go well

Ooo, we need to do that!

What was my "ooo, we need to do that!" moment?

It's measuring DevOps! We don't do it at all. So how can we know we're improving? We don't know unless we measure and track it.

I'm going to research the ways we can measure how we're doing with all things DevOps. One measurement we will NOT be tracking is 'number of deploys', because quantity doesn't mean quality. Is a team that does 50 million deploys per 1 Planck time doing better than a team that does 1 deploy per day?

Something Kris Saxton suggested in his presentation was regular surveys to Developers, asking them to rate things such as "Deployments happen in the same way, every time" from Strongly Disagree to Strongly Agree. I thought that was a great idea!

If anyone would like to share how they measure DevOps within their company, that would be super :)

For anyone who was at the DevOps Live event, what was your "ooo, we need to do that!" moment?

The Question of the Event

There was one burning question that kept coming up throughout the day, causing great division. The most important DevOps question:

What's your favourite dinosaur?

Mine is the Tyrannosaurus Rex, because it's clearly the best dinosaur and this is one of the best scenes in a movie of all time:

#DevOpsAwards #workfam #DevOps #CTGdevOpsLive