Registered: Aug 2002
Actually almost all the topics are covered in the two books. I have found some really hard questions (between 5 and 7) that are not covered by any phisical book and/or resource that I have studied, but really demand the real life experience of the candidat. Those were the hardest ones, as the answers are not to be found in the books, but searching throughout the OOA/D experience of each candidate.
My personal favourite questions were those about applying the patterns. I was sure that these kind of questions will come up so here's what I did: I studied the basic patterns(GRASP and GoF, not only those ones covered in Larman's book but some other GoF patterns as well although out of the scope of this exam), the corresponding UML diagrams and their applications. Larman book is the one but also Fowler's website (www.martinfowler.com, check the Analysis Patterns section. It's a work of art and it will really help anybody doing OOA/D). If you get the patterns right (I also try to apply them every day) it's likely that you'll recognize the UML corresponding digarams at the first glance, not to mention the questions in the exam that sometimes claim the response by using the right pattern!
My personal weakness were the questions regarding choosing of the use-cases to support some requirements. The trick that I have used was to apply the EBP rule. If the use-case does not bring _value_ to the actor, then it is not a candidate for a valid use case. I have briefly studied Kruchten's "The Rational Unified Process an Introduction" 2nd Ed. and one of the most valuable things that I have learned was about identifying an use-case. Kruchten talks about some ATM case study and he asks the question: Is inserting your ATM card and validating it a valid use-case? Well, imagine that the ATM machine is happy with your authentication and gives you the card back. Where's the value of this? Would you be happy to receive the card back without doing anything? Well. this is the key concept, the value of a use-case. If it does not bring any measurable value to the primary actor, than this is no primary use-case candidate. Another trick is another quote from Larman's book: "What is the goal of that goal?". But this one is related to the same EBP rule...
I think the most important is for everybody to try to find the weak spots and to try to cover them as good as possible. What I also did was to record all the mock ICE exams that I took (about 4) across my preparation and to try to analyse the weakest points. I have found that some areas were just not climbing at all by the end of my preparation (the same old learning curve) but the techniqe told me what were the areas not enough covered.
The score that I took in the real exam (no 100% at any section but around 85% at each) showed me that I have missed at least one question from each category. Well, I'd like to believe that I'll get to those answers right as I'll go deeper and deeper into OOA/D and I can't wait to get there.
After all, this is the value of all the study we take. Apart of the strong vocabulary you end-up learning and then using, the ultimate value stands in the ability to apply the concepts in the real life projects and this must be the goal of all of us here.
Report this post to a moderator | IP: Logged