Thoughts on PRINCE2 seminar...

I attended a seminary today focusing on PRINCE2. A lot of good speakers and subjects. I thought it was interesting and I'm going to look into PRINCE2 a little bit more.

Being from the software development industry I saw some parallels to RUP which I didn't like.
• PRINCE2 claims to be universal - it should work on every project in every culture all over the world.
• PRINCE2 as RUP is very extensive but it seems like the PRINCE2 community really have understood that it is a part of the method to remove unnecessary elements and processes.
• The manual is to thick...

Speaking on the seminar was also Birna Ósk Einarsdóttir from Síminn. She spoke about how they started using PRINCE2. The projects went fine except from the IT projects. They then introduced Scrum in the IT department. Very interesting to hear that suddenly things went a lot better... She also said the thing that appealed the most to me “You have to know the rules in order to break them!”. For me that says use your head and common sense to change things that are wrong on order to make it work better for you.

Scrum Smells: No impediments

Following the syntax of Micael Cohn:

Symptom: None or few impediments at the daily scrums

Discussion: Have you also been experiencing the feeling of that everything is going just fine in you Scrum project. On the daily meetings there seems like everything is going great. There are never any impediments. Look out! The team might be loosing their edge when it comes to seeing the impediments.

While this isn't always bad (it could possibly no implediments) it is very rare that the isn't any impediments for the scrum master to handle. There is just a matter of seeing them. Not recognizing impediments is a major obstacle to overcome in order to acheivie better efficiency and flow. Try to ask the members of the team if there is anything that makes them less efficient in order to get the discussion going.

Why says everyone that they are using Scrum?

I come across many projects that say that they are using Scrum. But when I scratch the surface and try to determine how they do Scrum there always seem to be parts missing;

• No planning meeting
• Assigning specific items to specific persons
• No burndown chart
• No handling of impediments

One thing that always seems to be in place is the daily meeting, but the goal of the meeting is often unclear.

Why is it like this? Scrum that is such an easy process? Is it just because Scrum and Agile are hot? Or is it a true misunderstanding of what Scrum is? I lean to the latter, Even though scrum is easy in theory it is hard to do it "right".

Why are unit tests too complicated?

One thing that often strikes me is that unit test cases often are too complicated. I think there are many different reasons for this and because the reasons vary one need to take different approached in dealing with them.

1. Too smart developers
Developers in general like to do complicated things, this rubbs off on the unit tests. Instead of keeping them small en easy one introduces too many abstractions and other clever solutions.
How do we deal with that? Increase communication. Let the developers explain the test cases for people that aren't generally involved in development.

2. The design sucks
If the design is poor and complex this will affect the unit tests, and make them smelly. How do we know that this is the case? Let someone new start working with the system and if he or she is slow in being up to speed, there is a possiblity that the design is to complex. One other thing it to run some kind of metric software on the code and make sure that it is not to complex.

3. The unit tests are not unit tests
The unit tests are not unit tests. Ok then so what is a unit test? The unit test aims test the smallest part of the application, eg addition of two numbers (nice school book example ;) ). An example of a test that isn't a unit test is "Log in, place an order verify order in order book." Refactor your tests and separate funcional tests (which of course can be implemented in code) from unit tests.

Wanting to write books?

I have never really been a good writer, but a dream has always been to write a book. I found the web site of the company 37Signals (www.37signals.com) and started reading their book "Getting Real". I'm not gonna comment on the book just yet. I like real book and not just reading from the web. And found Lulu (www.lulu.com). What so great about Lulu? Well everyone can publish their own book, and when somebody orders a book it is printet on demand. For me in Sweden it to about 7 days from order to delivery! Really makes you want to write one on your own.

Why focus so much on technology?

One thing that strikes me as a bit odd is the constant focus on technology that we suffer from in the IT Business (at least in Sweden). What do you mean? Should we not stay committed to technical excellence? Of course we should but are really Java and .Net the "best" technologies for everything? Let me take you through an example!

Finally I've been invited to Company X. According to my research the product I am about to present should really help them be more effective and possibly even be the thing that turns their annual loss into profit. When I enter the room the CEO, CIO and some more people are present (don't catch their name, my heartbeat is to high).

I start the presentation with walking them thorough a scenario that I believe is similar to what they do on a daily basis. Everyone likes the product and some people even realise that this might help the company focus on the right thing and not all that administrative crap that today takes up most of their time. Sensing that, the CEO asks for the pricing model. I tell him that he realize that the money spent is returned acceptably fast. Then the CIO steps in and ask what technology this is built on. We build it on Platform Y (Opps not Java or .Net). The CIO says that he never heard of Platform Y and firmly declares that we are committed to .Net (or why not Java) at this company and that you can come back when you have a .Net version. Your heart stopped (or at least it stands still for a while). The silence is total.

Why does this happen? Is the technology so important that it justifies not using a product that in all other aspects are great? Or is the CIO just showing his lack of knowledge when it comes to Platform Y? What would you do if you were the CEO?
I'm still waiting for the moment

J2ME Game development

In my preparation for the Expert Zone conference http://www.cornerstone.se/expertzone/dev/ talk on J2ME, it has struk me how fun it is to create games the old fashioned way. I hope that the talk will be appreciated and that people attending will learn the basics of J2ME MIDP programing.