This blog contains reflections and thoughts on my work as a software engineer

søndag den 28. december 2008

Buy it - just buy it.

The best thing about Christmas is the days off from work. You get to spend a lot of time with your family and that's great. You also get to see a lot of movies in the TV - which is nice because I like to see a good movie. I don't see a lot of TV basicly because I choose otherwise - my life probably goes on even though I don't know who got voted off the island during the last episode of Survivor / American Idol / Big Brother / (insert any mediocre TV-show here)...

You also get to read a lot so I've finished the book I blogged about three days ago - Michael Feathers Working effectively with legacy code. The short story is: Buy it - I was completely sold on his final appendix where he goes through all the various ways of decoupling dependencies - each example is followed by a codeexample and each example and explanation covers about two to four pages. Extremely nice setup and well thought.

One of his examples is the "Extend and subclass" refactoring. I never thought of that one but it is so obvious and offers lots of advantages with only a few minor disadvanteges so I'll share it with you here:

...(and 15 minutes passed...)

No excuses: I have failed to come up with a decent coding example on my own - the code just turned against me once I looked at the time and found out that one of the three "Lord of the Rings" are beginning in 5 minutes... Not that I haven't seen all of them at least 3-4 times but this is Christmas and about the first vacation this year which doesn't include a lot of work on our house or trips to far-far-away family - so watching a movie I've seen before and NOT get up from the couch and get some work done is really a new sensational feeling..... I could probably learn something by doing a bit of coding on the issue I'll just find a perfectly good example written by Jimmy Bogard for you  :o)

I'm off to see some orc bashing. Happy New Year to all of you - see you in 2009  :o)

Regards K.

torsdag den 25. december 2008

TDD on untested code - why does the first test take so long?

Most software developers have at some point in time been assigned a task with the following description: "We need to extend report X with a new column which sums up all the other columns in each row" - or something like that. Even if you don't know anything about the codebase tasks like this one in most cases don't cause too much trouble because you extend already something which is already working and in production. Even if you like to deliver this new functionality with shiny tests with 100% coverage of the new functionality it isn't a big deal because you don't have to touch the existing armageddon of untested code.

Worse - much worse is the task assigned to you the next day which says "Well - we have this bug in our invoicing system which causes all sales on the 1. of January to never show up in our reports. We need you to fix it". Even if you know that the invoicing system is a pretty decent working piece of software without a lot of maintenance - you also know that it has a mere code coverage of about 2%... When asked "How long will it take" what do you respond?

If you and your team are on the TDD side of things I have always been astonished how long it takes for someone to take the first step towards TDD-enabling untested software. However after having read the first 10-or-some chapters of Michael Feathers Working effectively with legacy code I am beginning to put words to my gutfeeling of "This will probably take a while..." when I see a change request to a system I know doesn't have a decent testsuite attached.

You've been assigned the task and you need to write your first test - why does it take forever to write it? Michael Feathers goes deep with numerous code examples but from a birdseye view it all sums up to the fact that before you refactor the dependencies in your untested code you can't write a suite of good unittests. Unittests must be kept as short as possible - I've heard the "No test should be longer than 15 lines of code" and it is a good best-practise to follow. Of course it doesn't always happen but if you have to write 20 lines of code to set up your test you need to look for design smells - most notably violations of the Single Responsibility Principle on the class or classes you are testing.

What I've recognized after reading these chapters are that code which have been developed without testing in mind have lots of dependencies to other classes. I've whupped up quite a few of myself - no doubt about it. When all you do is write code and never actually use your own code in a test you'll never get to see how your code "looks" from the outside. You basicly just make things work - and whoever is left to use your code after you have signed off on a task are pretty much on their own. This is how i.e constructors and methods with 10+ parameters are being born - I've seen them in my entire professional career wherever I've been - if you are "coding away" and need an extra parameter, you change the signature of your method, change the caller to use your new signature and everything is fine - from your perspective. What happens is that the more parameters you pass into a method the more state you will have to set up in your "Arrange" section of a unittest. Given that the more parameters a method or constructor takes the more state is injected your tests of that method will also be bigger in terms of lines of code - and long "arrange" sections in your tests are as much of a codesmell as any other codesmell you can think of.

The first test you write on untested code will have to take into account the time spent refactoring all the intertangled dependencies you dig up trying to get to those precious three lines of code you want to test on. It's a fact of life - get used to it. You might have to split up large classes, move object creation to constructors, extract interfaces to be able to mock dependant libraries etc. before you can actually make the single test that solves your problem turn green... The next time somebody has to write a test on your refactored code it will be easier - but the first test takes the longest - and the first test is also the one people fear the most because there is no other tests around to back you up when you begin to alter working code.

I agree with him on most of his writings - not everything though. He actually promotes writing "characterization tests" for already existing code "to be able to test for the existing behaviour". I disagree with him. The time you spend writing tests on existing code could be better used focusing on solving the problem you have before you. It's a mere waste of time in my opinion - when do you know enough and have written enough tests to conclude that the core existing behaviour is covered and documented in your tests? No sir - Keep It Simple Stupid, just get the job done. Use some of the many useful techniques he has documented but please - don't write tests for existing code. He means it well but honestly - it's just not valuable work to write code for existing functionality.

Even though I'm not finished with the book I can highly recommend it to anyone who just wants to be able to navigate through untested code without causing too big waves in the pond. It's a good book with quality stuff in it - definately one I will use to plan my attack on codebases that I've never touched before

Regards K.

onsdag den 17. december 2008

Silverlight - the pros and cons

Some time ago the team I work with decided to go to a one-day Silverlight seminar at Microsoft Copenhagen. Today we went and with an openminded heart (read: I didn't know anything about Silverlight except that I know I at some point downloaded a plugin to be able to access a beta version on MSDN (I think it was...).  I admit it: All I knew about Silverlight in advance can be told as "It must be some sort of Flash something-something from Microsoft".

So I went - and here's the sum of my notes and the people I talked to during lunch etc. Let's start with the "What": 

Silverlight is a cross-browser plugin which enables you to write light-weight Windows Presentation Foundation (WPF) applications and embed them into a website. It doesn't require the full .NET framework  on your machine but instead the plugin you download contains a smaller, compressed version of the .NET framework which enables you to run compiled Silverlight IL code. Silverlight uses XAML which enables you to write markup language using an XML syntax to express a UI.

That's the short story. There's much more to it (please leave a comment if the above needs to be corrected) but that's the basics as I know them of tonight. However - I talked a little and reflected a little during the rest of the presentations today and finally home again I have come up with this list of pro's and con's from my notes and my memories of the conversations I had:


  • The Silverlight Toolkit on Codeplex is a good thing and it reminds me that someone at Microsoft is beginning to "get it". The toolkit is released frequently and doesn't require the client to download an updated version of the browser's Silverlight rendering engine. Any additional usercontrols except the standard set of controls (which only covers the most basic textboxes etc)  are embedded in your specific Silverlight application.
  • If you are a WPF developer writing desktop apps the switch to writing Silverlight website apps is easy to overcome because the context (XAML, .NET) are very similar. This isn't exactly the case with any other existing Flash technology where the learning curve is steeper no matter what you have tried before Flash
  • Expression Blend is a working environment aimed at designers. How much you tweak and turn Visual Studio I have never heard anyone say that it is a good working environment if you aren't committed to reading code. Designers have had more luck with products from i.e. Macromedia and I think that Expression Blend is trying to please other roles than the "system developer" - which is good  :o)


  • I was told that if you i.e. copy/paste XAML from one file to another without also copying the codebehind, the target XAML file (with an empty code-behind) would only fail runtime - the compiler wouldn't let you know that you were missing method stubs in your codebehind to match the various events you had set up in the XAML markup. Error messages could be pure nonsense and debugging XAML is impossible. Of course you don't get to succesfully build a broken XAML document missing i.e. and end tag but checking for something as simple as method stubs should be a no-brainer. What am I missing here - we're at version 2.0, not some 0.8 beta release.
  • It appears that there is a Silverlight unittesting framework available - but it sounds a bit disturbing in my ears that the suite of unittests runs in the context of a browser? In order for me to adapt to Silverlight I need to have a good testing framework around when adopting a new technology - definately something I need to check up upon.
  • Styling in XAML means no inheritance - and your styles must be explicitly set on every element... That means that you can't have a general style for i.e. all your buttons which styles every single button on your site the same way. Oh no - you explicitly have to tell each and every button how it should be styled using an attribute on the button....... The guy from Microsoft breaking the news to us tried to be diplomatic about it. I can only imagine why this problem hasn't been solved when releasing version 1.1 - I wouldn't expect such stupidity to survive much beyond a version 1.0 of any product
  • Silverlight applications on the web can't be indexed by Google. Microsoft is working on this issue which should be solved in version 3.0 (due in Spring of 2009). Google is a cornerstone in any marketing strategy so the lack of searchengine compatibility shouldn't be neglected when qualifying Silverlight as a mature product.

Bottom line is: I think it could be fun to work with Silverlight (I'm nothing but a programming scarecrow who likes to bring home new and shiny technology to my computer) and I think that Silverlight has the potential to be "big". I also think that right now I wouldn't use Silverlight for much more than menues, banners and other non-critical stuff which requires the user to have a decent visual experience but I couldn't recommend anyone using Silverlight for building fullscale websites based on what I've seen today. However this technology has come to stay and the web is increasingly getting focused on rich clients and UI so this isn't my last encounter to the Silverlight technology.

Regards Kristian



Microsoft Silverlight: 10 reasons to love it, 10 reasons to hate it