Archive for December, 2011

Buying Software Tools is like Buying New Sneakers for Your Development Team

December 16th, 2011 by clucca

sneaks Buying Software Tools is like Buying New Sneakers for Your Development TeamSCM tools have a profound effect on the day to day life of a developer. These types of systems have either helped or hindered development teams deliver software. SCM systems are like the “hub” of a development team. It’s where teams artifact important work, integrate changes, save important ideas and add features for customers. It’s the center of our development universe!

It’s all about the developers. They need to be free to innovate and get changes out the door quickly. But they can’t if they are stifled by tools that get in the way. Tools need to be able to ENHANCE the software development process. Many people think that source control is just a place to checkin / checkout code. But it’s more than that, it’s where the software development process comes to life. If the SCM system isn’t up to the task of a complex development process, developers can’t innovate.

Sometimes it’s hard to understand that you have a tooling problem, even if it’s staring you in the face. Think of an old pair of trusty sneakers that you have at your house. We all have a pair, they are many years old, beat-up, dirty, torn… but we still wear them. Our feet hurt when we wear them, but for some reason we refuse to get rid of these old sneakers. Until one day (usually after a sprained toe) we decide to buy a brand new pair and after a little breaking in… WOW our feet feel great! Why did I keep the other pair so long?

Software tools are often like this, there is an “if it ain’t broke (too much), don’t fix it” attitude. We often keep tools too long after their expiration date. You’ll hear it from your development team, moaning about the pains of merging code, switching workspaces, checking out … it’s enough to make you cringe. But still we don’t change. Your old SCM is the sneaker, and collectively as a group you and your team have a hard time recognizing when your feet hurt.

Three Surprises in Software Development in 2012

December 8th, 2011 by lorne cooper

‘Tis the season to make unsupportable predictions for the future.  Despite my prior record (and I remain surprised that we don’t yet have personal jet packs) I’d still like to share a long-range weather forecast for the software industry.

You’ve been warned.  From here on, you’re on your own.

Prediction 1: Everyone Will Claim They Are Agile

And 50% of them will be wrong, just based on the Nokia test.  And of the rest, half won’t get any value from it.

There are a lot, and here I really need to underline a lot, of bad development practices out there.  For every organization that is killing it with Agile, there are five (my agilesta friends say ten) organizations that are limping along, delivering buggy code to their customers, late, and missing committed functionality.  And often all three.

This “Going Agile Without Knowing How” problem is probably an inevitable result of the success the early-adopter teams had with Agile methods.  For instance, when I watch The Olympics, figure skaters make skating look effortless.  When I do it, I look like a drunken hippo and hurt my butt.  It’s hard to stop and remember that these athletes, in addition to good genetics, spent years at the rink with their coaches learning, trying, failing, and improving, before they got in front of the TV cameras.

Agile has crossed the chasm, and the great majority of organizations have too few people, with too little coaching, and hardly any tooling.  Sure, your boss doesn’t realize how useless your stand-up meetings are, or that your code isn’t fully tested at the end of a sprint, but she’ll eventually see that your customers are not happy.

Prediction 2: Development for Mobile Devices Will Still be Small

Yes, Mobile is really big, and moving fast.  It’s just that the great majority of the work to support useful mobile apps remains in the back office.  When we’re finished inventing new ways to swipe our coffee-stained fingers across our screens, the value of the great majority of our apps is back in the glass house, running Java and C++ on big ‘ole honking (virtualized) servers.

The development problem in the era of proliferating small platforms, remains the problem of dealing with large, complex, data and it’s interactions.

Sleep well, Larry Ellison.

Prediction 3: The Gap Between Pros And Amateurs Will Grow

Every new software technology “spike” rewards the early adopters with higher productivity, which can level the playing field for the “newbie” or occasional software developer.  But as application complexity grows, as the platforms become more complex and development environments become richer, the professional advantage becomes more significant.

There isn’t much of a disadvantage in time-to-market for the young developer, maybe working on his laptop with open source tools and no identifiable process. The difference between them and a team of experienced professionals, working with industrial strength tools and procedures, and building apps that run businesses on virtualized hardware in a web connected world, is in “value created.”

Created Value is a concept that we’ve all been learning during the past ten years of Agile evangelism.  Created Value is measured at the customer side, and primarily by the classic metrics of software: how does the software help me get my job done better and faster?

There was a time when the lowest cost labour source for a software project was the key criteria.  Over the past two years, software projects have been revisiting their decisions as they’ve seen the crippling effects of buggy, unmaintainable, badly architected products.  We’ve seen the evidence with a hiring boom in the US for developers and QA alike.

In short, in 2012, we’ll see a renewed focus on quality of development, over quantity.  And a better appreciation for the talent, tools, and techniques, that create it.