Continuous Delivery: Professionals vs. Amateurs
Join the DZone community and get the full member experience.Join For Free
‘Amateurs [musicians] practice until they can get it right;
professionals practice until they can’t get it wrong’
This witty and motivational aphorism was said to have popularized by Harold Craxton, a professor at the Royal Academy of Music in London. Where a performance becomes automatic for the performer through practice, and where repeated performance has value, then there could well be a transition from amateur to professional for the performer. It fits music very well of course, given the nature of the sequence of notes, that are a song, and the need to not miss even one. Practice, specifically, was the take away if you missed it.
It seems to me that the saying fits releasing software to production too. If you’re dropping code into production, as FaceBook do, multiple times a week, and you’re not scared by it, then your release group is professional. Your team has pushed through amateur to get to that professional place through practice. It’s not just practice though; One huge key to your success towards that goal was automation. Your individuals, were they to follow a checklist to get the software into production would never qualify for the professional status of the aphorism. By automating the deployment via scripts held under Source-Control (whether a human click was needed for the actual deployment or not), the team is able to repeatedly deploy successfully. Sure there will be mistakes made during deployments. Indeed, there will be occasionally when a release has to be abandoned, or rolled back. Here’s the catch though, the release team is professional (through practice) with that too.
Facilitating such professionalism
The release engineers:
- get to practice deployments outside of the actual go-live cycle
- as #1, but for rollbacks
- have at least one non-live environment that’s configured the same as live (for practicing in)
- have release scripts under source control too
- have contrived scenarios where a release is nobbled deliberately to test the team
- are comfortable with flipping toggles in a running stack
Actually, the last two also apply to support engineers. It is where someone outside the team does something like secretly yank an ethernet cable, or something more subtle in software, and monitors the reactions and problem solving of the team.
The exact opposite of that professional practice?
Things will go wrong in production pushes and how the organization reacts is key. It was common practice before Agile, and is still makes sense to a certain type of manager, but the ‘safety’ from adding more managers, or more checklists after a botched go-live is alluring. The correct reaction to a failed go-live is to ask for root cause, and what can be done to automate through the same issue for the next time. It is counter intuitive for that type of manager or stakeholder, but they have to push towards the automation and practice. That will result in the outcomes they want – successful go-lives every day (or more).
Tracking the origin of the quote
For forty years Craxton (1885 – 1971) was a professor at the Royal Academy of Music. I like to imagine him guiding many students that he oversaw with this aphorism. It could be that he was the first to say it, but did not claim it when he used it. That certainly makes sense if he used it to motivate students. That said, A A Owen, who studied with Craxton for three years, does not remember him saying it once. You’d imagine a great professor would say it to a prodigy or would have said it to a third-party at least once in his presence.
Nigel Rees is a British broadcaster and author who’s issued a “QUOTE … UNQUOTE”:http://en.wikipedia.org/wiki/Quote…_Unquote newsletter since 1992. There is also book spinoff That is where I first encountered his work – various entertaining titles found their way into Christmas stockings and consequentially became well thumbed light reading matter in family homes, if you know what I mean. Before the newsletter, there was (and still is) a radio programme with the panel-game format. It started in 1976 and is in it’s 48th season now. Nigel has been collecting quotes for a long time. In April 1994 (Vol. 3, No 2), QUOTE … UNQUOTE contained the first internet-visible Craxton citation for the quote. What Nigel Rees said when I queried the Craxton entry:
I think I must have been sent this by a listener to my radio show but I don’t think I have used it on the show. It was very definitely ‘quoted by’ rather than ‘originated by’ Harold Craxton.
There’s also a single reference for US composer Milton Babbitt for the same quote, but I think that’s a mistake. Babbitt wrote dozens of books and had many quotes, but that was not one. “Unknown” is what it seems is penciled in for author of the quote, for now.
One last thought
Jez Humble is the ThoughtWorker who wrote the "Continuous Delivery book": http://continuousdelivery.com with former ThoughtWorker Dave Farley. Stuff I say here, shouldn't stop you from reading that definitive tome.
Published at DZone with permission of Paul Hammant, DZone MVB. See the original article here.
Opinions expressed by DZone contributors are their own.