Stefan Guilhen

Hi all,

I agree with Barry when he says that there is a crescent number of people who are taking the certification without the appropriate knowledge. I've been around in the SCJP forum for about 1.5 years now and I must say I saw a lot of posts made by people who really need a lot more experience with the language before aiming at the certification. Some people are not even able to write a simple program to test their doubts. They rather post their doubts and if you reply sending a link to an external reading that could help they say "oh no, plz, be straight... that document is too confusing for me".

However, phrases like "This is the main reason I doubt the value of the SCJP certification" really saddens me. I have 5+ years of experience with the Java language, developed several projects and I am constantly learning and enhancing my knowledge. For me, the SCJP certification was a recognition given by Sun regarding my knowledge and experience with the language. That should count as a plus in my resume, not as a point of doubt.

Are there people who pass the exam and don't have the adequate knowledge? Yes, there are. Unfortunately there is no flawless test and you can't prevent this from happening. Maybe the tests could be harder (I personally think that some of them are easy) but still this will happen.

Maybe this is the problem Sam is having. Fortunately you don't hire someone just based on his/her resume. You have the interviews where you can better analyse people and select the ones whose profile matches your need. Sam, I really understand your point and I appreciate how you are encouraging your colleagues to study for the certification as well. After all, the value of what the knowledge you gain in the process is priceless. At least for those who really want to lean something and do not treat certifications just as a way to add value to their resumes.

Cheers,
Stefan

Replies to all

What's next?

Hi Nicholas, I'm now back to the java certification track. Planning to take SCWCD in one month, as I already read HF Servlets and I'm now in revision process. I probably could take this test sooner, but I like to read the specs

may i ask if these 2 are the correct books and editions, that you mentioned:

1) UML Distilled: A Brief Guide to the Standard Object Modeling Language, Third Edition (Paperback) by Martin Fowler

2) Applying UML and Patterns: An Introduction to Object-Oriented Analysis and Design and the Unified Process (2nd Edition) (Hardcover) by Craig Larman

Hi Jesus,
Yes, these are the two books and editions that I used. However, keep in mind that the recommended edition for this test of UML Distilled is the Second one. The third edition cover the 2.0 version of the UML and introduces new notations and diagrams. I had no problem reading this edition because the book explains what changed since the last version of the UML. So you just have to be aware of these differences for the exam.
Botton line is: if you can get (borrow) the second edition it may be more appropriate for this test. But, if you are thinking about buying this book, you can buy the third edition as it covers the current UML and you should not have any problems as long as you identify what changed.

What is your expereince with OOAD+UML exlcuding reading Fowler and Larman's book?

Hi Viki,
I've been using the UML for about 4-5 years now, but my experience in OOAD was limited to my own projects as I was not responsible for the analysis and design of the systems developed by the company I worked for. However I am always trying to improve the quality of my designs and I developed an increased interest in OOAD and Patterns. I think this may have helped me to get a good score

Regards,
Stefan

Hi ranchers,

I took this exam last thursday (30/6) and passed with 83%. Many thanks to Hari Vignesh Padmanaban for his tutorials. They helped a lot and actually they were my starting point and main resource. As it is a tutorial, nothing is explained in detail, but you will learn the main concepts and what is important to know for the exam. I used XML in a Nutshell and other tutorials on DOM and SAX to get a deeper knowledge in the important areas of the exam.

PS: Hari, did you ever consider taking your tutorials as the starting point for a XML certification book? You did a great job and I feel that if you explore the sections in more detail and provide more exercises it would be the only needed resource for this exam.

Regards,
Stefan
[ July 04, 2005: Message edited by: Stefan Guilhen ]

Hi ranchers,

I took the exam 484 on friday 24/6 and the 486 exam on monday 27/6 and got 77% and 95% respectively. Also took the 141 (XML) test 3 days later but that result I'll post in the appropriate forum. Below I share my experience:

Enterprise Connectivity - Exam 484

Overall it is not a hard exam, but you can expect some hard questions. I read both HF-EJB and HF Servlets and they cover the majority of the test. For the connectivity services I used the tutorials posted in this thread.
Time should not be a problem. I had about 35 minutes left to revise the questions.

OOAD with UML - Exam 486

I read both Larman's and Fowler's books thoroughly. It is important that you know the UML constructs and when to aply them, and UML Distilled is very good at it. But for this test it is more important that you know the iterative process and how to derive artifacts from the requirements and another artifacts, so Larman's book is the ultimate resource for this exam. For revision, you can use the notes from Ezequiel Cuellar available here.

Personaly, I didnt use his notes because sometimes you need to remember the context in which the concepts were introduced and you can't get that from revision notes, so there is nothing like re-reading the concept from the book itself. As Larman's book is large, and most people (like me) won't read it twice thoroughly, I adopted a different strategy: I revised his book just by re-reading the boxes that contain all the important concepts. All chapters have one or more of these boxes, and you can rapidly remember the context surrounding the concepts as you read them.

Anyway, I included Ezequiel's notes here because some people may feel confortable using them as a revision resource. I just didn't work for me, but may work for you.

Finally, I recommend Mikalai Zaikin's Magnet Mocker engine. Now it has 3 quizes besides the ICE exam and they are very useful to test your skills. I would like to thank MZ for the time and effort he put on this.

Regards,
Stefan

Congrats, Mr. Sato

Great Score!

Stefan

The Decorator pattern is a well known pattern written by the Gang of Four (GoF). You can find further information about this and other very useful patterns checking their book:

Design Patterns - Elements of Reusable Object-Oriented Software
Erich Gamma, Richard Helm, Ralph Johnson and John Vlissides
Addison Wesley

This pattern provide an alternative to subclassing for extending functionality, that is, you can add responsabilities to an object that can be withdrawn later without using subclassing. The decorator acts as a wrapper that forwards requests to the decorated object and may optionally perform additional operations before or after forwarding the request.

Regards,
Stefan

Hi again,

As I don't have the book anymore to check, I tried to remember what kind of expression was the question refering to and I thought it was about LIKE expressions. My bad. Now I know it refers to the IN expression. The question does not make it clear, but given the correct answers we see it was refering to the IN expression used in the WHERE clause. I say this because page 229 (botton) of the specification states the rules for IN expressions used in WHERE clauses:

"The syntax for the use of the comparison operator [NOT] IN in a conditional expression is as follows:
single_valued_path_expression [NOT] IN (string-literal [, string-literal]* )
The single_valued_path_expression must have a String value."

So again, it expects a String value.

Stefan

Hi all,

The question Meg posted (page 430 - Q17) refers to Like expressions. If you check the EJB2.0 specification, page 230 says the following:

"The syntax for the use of the comparison operator [NOT] LIKE in a conditional expression is as follows:

single_valued_path_expression [NOT] LIKE pattern-value [ESCAPE escape-character]

The single_valued_path_expression must have a String value."

Although the question is not very clear, its asking whether the single_valued_path_expression used in LIKE expressions should result in String or not. As we can see from the specification it should, so the spoted alternative is indeed correct.

Hope this helps,
Stefan

Congrats, Binoj

A perfect score!

Stefan

Page 130 of the specification says: "Once the primary key for an entity bean has been set, the Bean Provider must not attempt to change it by use of set accessor methods on the primary key cmp-fields. The Bean Provider should therefore not expose the set accessor methods for the primary key cmp-fields in the component interface of the entity bean."

Page 134 completes saying that "If the Bean Provider attempts to reset a primary key value, the container must throw the java.lang.IllegalStateException."

Regards,
Stefan

The answer is yes for both questions. The beans participating in a CMR must define local interfaces and the accessor methods that you define in the bean class use the local interface as parameter and return type. As a result, you cannot expose those methods in the remote interface because you are not allowed to expose local interfaces to remote clients.

Nothing prevents you from exposing them in the local interfaces, though. Also, the remote interfaces can expose the non-relationship field accessor methods. But keep in mind that if you are using CMP your accessor methods will be abstract and if you choose to expose them in the component interface you cannot perform additional steps like validation when they are called because the container will provide the implementation of the methods. For that reason, it is recommended that you define and expose your own pair of accessor methods that perform any additional steps and then call the abstract accessor methods.

Regards,
Stefan
[ April 18, 2005: Message edited by: Stefan Guilhen ]

Hi,

Try this one

Regards,
Stefan

Hi Amol

Congrats!! Well done!

Stefan

I dont think you will get both options in a question because one could argue that both are right and the exam questions are very clear, they avoid such type of confusion.
If I get such a question I would choose NoSuchObjectException because this is what the container is supposed to throw. (see page 379 of the spec). But again, it is a subclass of RemoteException, so it is not wrong to say that the remote client will see a RemoteException.

Regards,
Stefan

You are right. I corrected my post in the other thread. I just wonder why this is not true regarding the method-permission element. I checked the section 21.3.2 of the spec (Application Assembler responsibilities) regarding the method-permission and it does not say that all the methods must be from the same bean, althought the example shown in page 445 shows this arrangement - notice the role employee appearing in two method-permission elements. Any ideas?

Regards,
Stefan