AKA – “Another post-it on the wall”

So… This happened.

A few of my friends put on the first SQL Saturday in Southampton which, as it happened, was the last UK SQL Saturday of the year. I thought I’d try something a bit fun and a bit different while surrounded by a bunch of my #SQLfamily friends.

I’m not gonna lie – it could have gone two ways. You’ll have to be the judge of that.

(Note: it’s much better with decent headphones to pick up the baseline.)

Thanks Grant!

I’d like to say a massive thank you to my colleague, Grant Fritchey (b|t). He got the biggest laugh with his cameo and in my opinion it was the best bit of the song. Grant has given me some really great guidance as I’ve started out with blogging, training and speaking. He is a very knowledgeable SQL Server MVP and community member who shares his knowledge with precisely that unique personality you saw in the video. If you don’t already follow him and you enjoyed our song please read one or two of his blog posts.

If you are interested in the content of the rest of the session, after the song I basically skipped to 9m20s into my Elephant in the room session from NDC. If you’d like to watch the full session or read the blog post you can do so here.

My “Don’t be a brick in the wall” series

This blog post is called “Part 3” for a reason. Last year I wrote a post that used the 2014 World Cup (which was about to start) and Pink Floyds classic album “The Wall” as metaphors in a discussion about working efficiently across teams. You can read it here.

In that post I made a prediction about events in the world cup final. These events were totally made up and seemed completely insane but in the semi-final something similar happened. I couldn’t resist writing a follow up blog post. You can read part 2 here.

The song “Another brick in the wall” has three parts – so this series has been waiting for a finale for quite some time.

Why would I commit social suicide like this?

At least that’s what my wife thinks I’m doing. She is not a software geek and doesn’t find the song funny. I think I’ve driven her mad with my practice. Imagine hearing that repetitive baseline over and over for the last two weeks! I’m not even any good. I’m sure some would class that as psychological abuse. Case for divorce I’m sure. Possibly criminal proceedings? I’m a very lucky man.

Anyway… I’ve been thinking about doing something like this for a while purely because I think it’s a bit of fun and I’ve come up with various song ideas over the years. I was even more inspired to do it after NDC Oslo 2014 when I went to Roy Osherove’s session and he finished with a song about the #isTDDdead debate. (His song is way better than mine and he’s done a few now so if you liked “Post-it on the wall” go check him out!)

Then a few months ago I came up with the lyrics “We don’t need no automation, We don’t need no source control” and wrote the rest of the song around it. It seemed to work. At least I thought it was funny anyway.

I had my idea. Now I had to work out if I could execute on it.

Be honest with yourself

On average we are average.

As a teenager I thought it would be cool to play guitars. Like many teenagers I had wild dreams about being a rockstar but the truth of the matter is that I neither had the skill, motivation nor the self-discipline to put in the hours of practice necessary to be in the next Pink Floyd. (As I just demonstrated! :-P) I learned to play a few songs badly and had a bit of fun and that was about as far as it went. The vast majority of us won’t be the next Steve Jobs, Beyonce or Usein Bolt.

On average we are average.

I can’t sing, I’m a bad guitarist and I struggled to play that baseline in time reliably. I can also only do one thing at a time. In order to be successful the vast majority of us mere mortals need to work out what success looks like for us and how to be more than the sum of our parts.

Pick achievable goals

I set what I believed was an achievable goal for myself: The song had to have a message about software development and it had to be funny. I also wanted to inject some energy into my session which was the last session of the day. I had to achieve this within my constraints:

• Time (to practice)
• Ability

These constraints effected my song choice. Brick in the Wall is relatively easy to sing and play.

Automation can help you do more stuff more reliably (mostly)

I automated the hell out of it. I can’t reliably play the baseline in time so I got it right once and automated it. I can’t play two guitars at once, but I can automate my baselines to produce a more interesting piece of music. Getting Grant to do his cameo live either in person or via video conference would have been expensive, impractical, fiddly and/or unreliable so it was pre-recorded.

By focussing my attention, getting something right once and then automating it I achieve much more reliability. I cut out human error (mostly). I free up time to do other stuff. I also reduce the requirement to spend so much of my time practicing as I won’t need to be able to play the baseline consistently for a long period of time – I just need to get it right once. If I make a mistake I can try again.

Of course, the thing I didn’t quite appreciate is that just because you can automate something and can trust it to work flawlessly, it doesn’t mean you can work flawlessly alongside it. You need to become skilled at working with your automation process.

The part I messed up so badly was getting the timing right during the main chorus. In the heat of the moment I completely lost my rhythm and as a result went into the guitar solo all flustered so made mistakes there too. If I do this again I know I need to focus on practicing that bit some more.

Be clear about what you are not going to do

I neither had the time nor the ability to become a talented guitarist and singer in time for the event. (Having watched the video back I can’t help but cringe during the chorus and guitar solo.)

However, I could deliver something that met my requirements (funny, message, energy) much sooner and get feedback much more quickly if I accepted that it was OK if the strumming wasn’t perfect and that my singing voice needed work.

Deliver quickly and learn

I was making an assumption: It is OK to do something like this even if you aren’t a very talented musician.

I aimed to find out quickly and cheaply if this assumption was accurate. Perhaps I do need to spend many more hours rehearsing and invest in some tuition? But then, maybe I don’t. The quickest way to find out for sure is to try it and see.

What’s next?

I’ve delivered my Minimal Viable Pink Floyd cover and I’m interested to see how it goes down.

The song might have been a mistake. I might get trolled. Or, if people like it, I might do something similar again. I can make decisions about further investment of my time and effort based on feedback. I already know one area where it is definitely worth investing in some practice time if I do decide to do this (or something similar) in the future.

If you liked the song please let me know. If you’d like to see me do something like this again I want to hear about it – otherwise I probably won’t.

If you didn’t enjoy it I’d really appreciate (constructive) criticism. Believe me – I am already fully aware of the mistakes I’ve drawn attention to in this post, but do you have any other advice.

If you know someone else that you think will like it please share it with them. That way I might get more feedback. 🙂