Dispersion problem – why am I contributing so little to ArgoUML?...

At SISCOG some of the employees – I included – attended a leadership workshop that lasted 3 days and was distributed by February, March and April of 2009. It was very interesting and useful for the development of my soft skills, which I need as part of my role as a team leader. The person giving the workshop was Teresa Roseta of MQI, whom I strongly recommend. SISCOG developed the leadership principles and values in the workshop.

One of the things that was referred there was the need for a person to know himself better. This because it seams that normally good, natural leaders do know themselves very well. One exercise we did for each of us was our professional coat of arms. It is a bit like a personal SWOT analysis, but, the idea was to minimize the number of things that go into each quadrant and to have a middle row with your motto. The things that go into the quadrants are different too. To make a long story short here is my professional coat of arms (the words in [emphasis] are just legends for the contents):

[Strength] easy learning and appreciation for knowledge	[Dream] to fulfill one or more remarkable ideas (hint: MEDEIA)
[Motto] do things right (i.e., well/correctly)
[Vulnerability] dispersion	[Facilitator] persistence

My vulnerability is dispersion. This causes some problems for my average rate of contributions for anything that isn't my primary activity and currently for my contributions to ArgoUML, which are too few and too slow. Issue 5864 illustrates this. I was given a patch, so, I looked into it and thought, OK, lets help this person scratch his hitch and maybe ArgoUML can get a new developer aboard. Then a difficulty came and I feared that it was something that was related to a part of the code that I never delved into and I placed it on hold until I had enough energy for it. But, in the meantime I had enough energy to do a Programming Praxis problem, to start developing a friendly user interface for htmled, to start a crazy attempt to update all the software in my old Windows XP laptop and to investigate the possibility to move some personal Subversion repositories into Git! At least these are the significant things I remember about between 7th and 30th of October. Why didn't I finished the thing I started in the first place? Dispersion!

I disperse because I'm the kind of person that loves to learn and I have lots of ideas. These ideas are all very interesting for me, but, they take some time to look into, develop and document. They are mostly related to software development, but, they don't intersect as much as possible for them to be developed together or in order for some parts to add to the others. So, these things start conflicting with each other and I don't get much done of anything unless at my work, where I managed to finish some significant things.

I like ArgoUML. The developers there aren't the greatest community ever, but, they are nice persons who care most about making ArgoUML incrementally better and some are pretty decent software developers with high commitment level. I think that I contributed some important work for the C++ module and did some nice work in the UML profile support. But, I'm involved in ArgoUML since 2004 and the pace of my contributions has been very snail like or even slower. Even worst, both the C++ module and the profile support seam a bit of an half-finished business. I think I know where the problem lies – dispersion!

So, I do I fight this? I do it by setting up a challenge for myself – spend one year focused at contributing to ArgoUML. Forget about the software blogs, software books and exercises. Blog about ideas I have, take some notes, so that they don't get neurologically garbage collected, but, then go back to work. In the end, take some ArgoUML vacations for some undetermined time, look at one of these ideas and pick it up if I want to give it a try. (Probably I will want to do something related to my work at SISCOG and if so, it may take one year or more to finish, but, I don't know and for now don't care what that will be.)

So, my aim is mainly to move profile support in ArgoUML up into something that looks a lot like finished business and try to do the same thing for the C++ module (this one is harder, though).

But first, as a last bit of devilish concession to my dispersion I'll spend some time writing and publishing some of the significant ideas in either Argonaut's life if they are ArgoUML or software development related, or in Idiota if they aren't.

Oh, and I have an idea for a trick that will help me not to get distracted. I'm going to get myself an ArgoUML bracelet, just like Uncle Bob has his green wrist band to remember himself that code must be clean and tests green. Maybe I should get a green ArgoUML wrist band!