Where do we go from here?

I will soon be made redundant from a job I’ve held for 6 years.  Thats the longest I’ve ever been anywhere.

The last time I found myself without a job, I was terrified.  How will I keep a roof over my family’s head?  What if I couldn’t ever find another job?  The fear of the unknown was enormous, mostly because the financial pressure felt crushing and imposter syndrome loomed large.  No-one will hire me because I have no idea what I’m doing.  Perhaps I should go talk to McDonalds.

This time round however, I find myself in the enviable position of having enough of a redundancy package to keep me afloat for 6-8 months.  If I find a job right away, thats a nice lump sum that I could do something useful with, either personally or socially, but if not, 6 months should be plenty of time to find something.  Anything.  So thats the financial fear taken care of.

And so we come, as we inevitably do, to imposter syndrome.  What am I worth?  The last time I did anything other than iOS development was about 10 years ago.  My Ruby/Rails skills are out of date, my Java is ancient history and my work consulting and training with XP/Agile process is almost certainly too old school.  I have, in 10 years buried in Apple tools and technologies, failed to keep up with pretty much anything else. The changes in web technologies, in process, in buzzwords and frameworks seem all too massive.

When I did C++ for money, I did Java for fun.  When I did Java for money, I did Ruby for fun.  When I did Ruby for money I did iOS for fun, and then I kind of stopped.  Family life meant there was little time for programming outside of work, so I stopped having an eye on the new things, and which of them might be interesting as a future direction.  Back then, each transition seemed obvious.  I knew what I was interested in before I had to look for a job doing it, and each time, because I was interested and made friends in that space, I didn’t have to look hard to find a place to do it for money.

This time though I find myself a little lost.

How can I find a job with skills so out of date and irrelevant ?  I don’t know any of this stuff?  I need to buy books on Elixir and Go and a 10 volume set of Javascript frameworks.

But wait, I know how to build good software.

I know how to build good software and I’ve done it in FORTRAN and C and C++ and Java and Ruby and Lotus Notes for goodness sake and C# and ObjectiveC and Swift.

I know how to build good software and I’ve done it with UML and XP and DDD and Scrum and using Jira and Pivotal Tracker and Rational Rose and index cards.

I know how to build good software and I’ve done it for banks and startups and research labs and electrical retailers and power stations and Coca Cola and telecoms.

I know how to build good software in anything.

So.  The imposter syndrome is kicked into touch this time too.

I feel strong.  I feel ready.  I have no idea where I’m going or what I’m going to do next, but thats a blessing, not a curse.

It might, as I’ve said, be lovely to spend the next 6 months as a junior Mac developer or Project Manager or devops engineer and be the worst guy in the band.  To focus on the small and the specific.  It might also be lovely to finally take a step up into technical leadership across a small organisation.  To think bigger and work harder out of my comfort zone.

Its a magical world, Hobbes ol’ buddy.  Let’s go exploring.



The Worst Guy in the Band

In Chad Fowler’s wonderful Passionate Programmer book (which you should buy), theres a chapter called “Be the Worst”.

Legendary jazz guitarist Pat Metheny has a stock piece of advice for young musicians, which is “Always be the worst guy in every band you’re in”

Chad spins this musical advice into advice for programmers.

Being the worst person on the team has the same effect as being the worst person  in the band.  You find that you’re unexplainably smarter.  You even speak and write more intelligent.  Your code and designs get more elegant, and you find that you’re able to solve hard problems with increasingly creative solutions.

I’ve tried to make a habit of this, and so far I think I’ve succeeded.  You’d need to speak to my co-workers to confirm.  I’m looking for a change in work right now, and its got me thinking.

The advice above is essentially saying be around people who are better than you, so you’re always learning.  But there are a couple of issues with this.

Firstly, you’re good enough to be in the band at all.  Chad talks a little about this in context.  You want to be the worst guy in the first division, rather than the best guy in the second, but that assumes you have the requisite skills to be in the first division.  Secondly, it assumes some relevance or connection between you and the others.

You can become a better saxophonist by playing regular jazz with a really good pianist, drummer and bassist, assuming they let you in to their band, but what if you really want to play drums.  You can’t be the worst drummer in that band.  They already have a drummer who’s good.

Maybe you go join a drumming group.

You’re surrounded by drummers who are better than you.  You are the worst drummer and every day you work with better drummers and get better at drumming.  But now you’re a junior drummer, and you’re getting paid as a junior drummer, and you’re playing basic stuff all the time, which is improving you as s drummer, but not really stretching you as a musician.  Your expert saxophone status does not buy you recognition in that drumming group. You’re a beginner.

What you want is a bigger band.  A band that already has a great drummer, but can hire you for your saxophone skills 6 days a week and indulge your desire to be a junior drummer 1 day a week.

A couple of times in my career I’ve wanted to change direction, and both times I’ve been lucky.  My first proper Ruby job, my Java experience and Agile consulting work was enough to pay my way while I was a junior Ruby engineer working with a team of experts.  My first proper iOS job, that Ruby (and the Agile work again) was enough to justify a senior engineers position and salary while allowing me to essentially be a junior iOS guy and learn the ropes.

I’m feeling like its time again.  I’d like to be a junior something.  Maybe Mac development, maybe project management, maybe something else entirely, but I have to find the right band.  Someone that’ll happily hire my saxophone while I learn my crash from my splash.


My pencils are sharp enough

Its time I started blogging again.

I’ve had several over the years and usually end up deleting them after a while.  I used MovableType, and Ghost and Blosxom and Jekyll and all sorts of other tools.  Twitter really killed it for me.  It made it easier to throw out a thought than to sit down a craft a post. This is a wordpress site because it was quick to get working and typing.

I pulled across 3 or 4 older things I still liked, though they’re ten years old and maybe not relevant any more, it meant I wasn’t starting with the pressure of a blank slate.  Back then I was doing Java and Ruby.  These days its ObjectiveC and Swift.

So I guess I’ll see where this goes.

Smalltalk Best Practice Patterns

unknown-1In the 1990s the “patterns” movement in software was in full flight. The poster child, and many peoples introduction to patterns was the so-called Gang of Four book Design Patterns, published in 1994. Design Patterns catalogued a series of blueprints, strategies for combining classes to solve a problem, and the idea spread through the industry quickly, to point where Design Patterns become a buzzword and to some extent lost a lot of its original meaning.

But software patterns come in all different shapes and sizes and one of the people instrumental in the movement towards utilising the notion of describing software with patterns was Kent Beck. Kent is involved with many of the most important (IMO) ideas in software over the last 20 years, including object-orientation through Smalltalk, Test-Driven Development, Refactoring and our subject here – Patterns,.

unknownSmalltalk Best Practice Patterns is a collection of tiny, very specific patterns describing how Kent wrote Smalltalk. I’ve heard Kent describe how the book was written once, and I’ve tried and failed to find a reference.

In essence, when he start to write a line of code, he looked to see if he had a pattern describing what he was about to do. If yes, he used it, if no, he amended an existing one or wrote a new one. The result is a reasonably complete specification describing the choices available to him in writing each and every line of code in a program, from variable naming to inheritance hierarchies.

The power of this method was obvious to me, despite never having really seen Smalltalk before, so in order to properly read the book, I wrote to Wilf LaLonde’s JOOP column asking him for a beginners guide to Smalltalk.  He wrote one for me, and doubtless other Java programmers keen to understand how a “pure” Object-Oriented language worked

Smalltalk Best Practice Patterns is, without doubt, the best programming book I’ve ever read. Despite never having written a line of production Smalltalk in my career, it fundamentally altered how I write code in any language, forever.

Acceptance Testing Classes

I have seen a lot of unit tests in my 10 years as an XP practitioner and coach. I have also had a lot of debates about styles of unit testing. The longest running of these debates is about the role of “mock objects” in TDD.

Martin Fowler has characterised the use of mocks as “interaction testing” in that you are verifying that certain methods get called as a result of stimulating the object under test, rather than that the state of the object changed.

I don’t like this style of testing for a couple of reasons.

The first reason, which will really have to be an article on it’s own, is that it tends to make refactoring a lot harder than it might otherwise be. This is due to a lot of tests that verify specific methods are called in a specific order. The names and order of these methods can change often early in development, as well as later if a large refactoring is performed.

The second is what I wanted to talk about here.

When every collaborator of a class is mocked, it becomes divorced from reality. It is truly being tested in isolation. We verify that when this fake object pokes it in a certain way it responds by poking these other fake objects in a different way. We have no evidence (at this point) to suggest that the object will be poked in the same ways in production.

What we end up with, I think, is a set of acceptance tests for an individual class.

When I mind map this I end up with “waterfall”. We are designing and building and acceptance testing a class in isolation, and increasing the cost of change of that class. We lose a holistic view of the system and are discouraged from refactoring, especially across classes. The software becomes less soft.