Advice to those planning to take the SCBCD

[Francis Amanfo]

Hi all,
If you take a look at the discussions going on at http://www.theserverside.com/news/thread.tss?thread_id=25779 and
http://www.theserverside.com/news/thread.tss?thread_id=25804
you'll notice that EJB is going to be totally overhauled. Most of it's features are borrowed from the Spring and Hibernate frameworks.
For those of you planning to take the exams, I'll advise you to think twice. I'll suggest spending time instead on Spring and Hibernate. By the time EJB 3.0 specs is complete, implemented by the vendors and embraced by the industry, you have a head lead since it contains features borrowed from these frameworks. Taking an exam for EJB 3.0 would be also easier.
Use your time, energy and resources wisely and stop learning things that there is evidence that it's going to disappear.
Thats my 01 cent.
Francis.
Looking for links to these frameworks?
Hibernate: http://www.hibernate.org
Spring: http://www.springframework.org
Tapestry is a great MVC framework for your front end. Check it at:
Tapestry: http://jakarta.apache.org

[Amitava Basu]

Hi Francis,
You have posted your advice at the right pale everybody here is related to SCBCD either passed or preparing.
I am preparing for the SCBCD for last 2 weeks and planing to take the exam before end of May. Studying hard after 8 hors office work....
According to you This is NOT worth spending energy & Time? You made my day...
Thanks, Let me know more...

[Alfred Harre]

Thanks a lot for information, I am also studying really hard for SCWCD. I would really appreacite comments from other senior members of this commuinty.
Regards

[Roger Chung-Wee]

Whether or not to use EJBs is an important subject for discussion - but not in this forum. Now, Francis made one post on this subject, and we can accept it. But making two posts is just unacceptable. I don't know what your problem is, Francis. You ordered Head First EJB in October 2003, maybe you then found it too difficult and now you are slagging off EJBs. But whatever your reasons, you are looking like a troll who is pissing off people who are participating in this forum in order to help them pass the SCBCD exam.
If you want a discussion on whether or not to use EJBs, please go to another forum.

[Sivasundaram Umapathy]

EJB 3.0 is now in the initial stage. It would atleast take 1 more year to have the final draft. And it would atleast take couple of months for the vendors to implement it, 4 or 5 months for Sun to create a new exam and may be an year for the businesses to start adopting EJB 3.0 solutions.Meanwhile EJB 2.0 will continue to stay here and won't be wiped away overnight by EJB 3.0
So doesn't sound like a valid point to me.

[Goan Balchao]

I'm not sure whether Francis has been awarded a marketing contract from Spring and Hibernate , but I wouldn't go that far this soon anyway. I can't see Spring and Hibernate replacing the big enterprise applications anywhere in the near future.
No doubt they are competing technologies and no doubt Hibernate is a huge improvement (in terms of development) over Entity Beans, but one should not turn a blind eye towards the features that EJB advocates too - scalability and versatility. Plus the fact that today's app servers support mutitude of features and are pretty robust.
So unless a company is going to start cutting costs radically and wants to shift over to these as yet "unproven" technnologies, I got my money still on EJB and am going to delve right into EJB 3.0
That's just my thoughts...

[Valentin Crettaz]

Before going on, please read what Kathy Sierra said once about which version gets covered in the exam and why things are the way they are:
https://coderanch.com/t/159240/java-EJB-SCBCD/certification/SCBCD-or-SCBCD
Francis, if you are so against EJB, why do you recommend using OpenEJB then?

[Francis Amanfo]

Originally posted by Roger Chung-Wee:
Whether or not to use EJBs is an important subject for discussion - but not in this forum. Now, Francis made one post on this subject, and we can accept it. But making two posts is just unacceptable. I don't know what your problem is, Francis. You ordered Head First EJB in October 2003, maybe you then found it too difficult and now you are slagging off EJBs. But whatever your reasons, you are looking like a troll who is pissing off people who are participating in this forum in order to help them pass the SCBCD exam.
If you want a discussion on whether or not to use EJBs, please go to another forum.

Roger,
Just to correct you, I didn't and don't find EJB difficult at all I have 2 1/2 years development experience with EJB. I'm pointing out something that is of great concern to everyone here on this forum. The issue is whether it is worth the effort to spend time and energy on now, especially after knowing this thing would be obsolete in 2 years or so time. Can you address that? It is not a matter of being a troll. It's too easy to call someone a troll and not argue with facts. I'm only a messenger. Don't persecute the messenger.
Best Regards,
Francis

[Francis Amanfo]

Originally posted by Akasmat de Goa:
I'm not sure whether Francis has been awarded a marketing contract from Spring and Hibernate , but I wouldn't go that far this soon anyway. I can't see Spring and Hibernate replacing the big enterprise applications anywhere in the near future.
No doubt they are competing technologies and no doubt Hibernate is a huge improvement (in terms of development) over Entity Beans, but one should not turn a blind eye towards the features that EJB advocates too - scalability and versatility. Plus the fact that today's app servers support mutitude of features and are pretty robust.
So unless a company is going to start cutting costs radically and wants to shift over to these as yet "unproven" technnologies, I got my money still on EJB and am going to delve right into EJB 3.0
That's just my thoughts...

Akasmat,
Like I've already mentioned in a previous post, EJBs are suitable for a certain class of enterprise applications. But that class comprises of a small number of applications. For the majority of enterprise applications a combination of Spring and Hibernate based solutions is known to do better. Moreover with this solution, there is no more deployment descriptors hell! No more JNDI hell! No more entity beans hell! No more paying of big bucks for the big boys to get even richer!
Have you questioned yourself why Sun is copying ideas from these frameworks to create the EJB 3.0 specs?
Best regards,
Francis

[Francis Amanfo]

Originally posted by Valentin Crettaz:
Before going on, please read what Kathy Sierra said once about which version gets covered in the exam and why things are the way they are:
https://coderanch.com/t/159240/java-EJB-SCBCD/certification/SCBCD-or-SCBCD
Francis, if you are so against EJB, why do you recommend using OpenEJB then?

Valentin,
As you may know we all get wiser as the years go by and turn to do things diffrently and better thereafter. In my 2 1/2 years of EJB development I've learned a lot in practice. I've learned that certain things in EJB are just evil. The specs developers made bad design decisions and they know it now. Hence the radical overhaul of EJB in version 3. They've seen how bad it was and are correting themselves.
Why can't I make a change of mind along the way?
Best regards,
Francis.
[ May 11, 2004: Message edited by: Francis Amanfo ]

[Nicholas Cheung]

The specs developers made a bad design decisions and they know it now. Hence the radical overhaul of EJB in version 3. They've seen how bad it was and are correting themselves.

I think new stuffs are sometimes (or even always) come out by trial an error. We think something is good, we wanna make it out, and then we improve it when better approach comes up.
I did appreiate that EJB did improve a lot, from EJB 1.x's bad CMP, to EJB 2.0's improved CMP.
In the past, when I worked with VAJ 3.5, we usually use BMP, as CMP was not mature and reliable at that time. But now, I am working with WSAD 5.1 with CMP happily.
I think things will be improved and improved, as time goes by. Thus, I can forsee that EJB 3.0 will be better. The approach of this is just like some iterative development processes. So, what's the problems for improving?
Nick

[C Chavan]

Entity EJBs in past had serious problems and have been improvized in later versions. Session beans and MDBs have proven to be great in distributed enterprise solutions. So, EJBs are here to stay.
As far as Spring/Hibernate or any similar frameworks goes, they have great features and should be considered as viable tool as per project needs and resource availability. But existense of competetive frameworks does not rule out EJBs and certifications in it.

[Francis Amanfo]

Originally posted by C Chavan:

As far as Spring/Hibernate or any similar frameworks goes, they have great features and should be considered as viable tool as per project needs and resource availability. But existense of competetive frameworks does not rule out EJBs and certifications in it.

I think you missed my point. What I'm trying to say is that the specification for EJB 3.0 is so different and shows a radical change from the current version of EJB. So I'm advising people to NOT waste their time and resources to learn and take certification for something that is obviously going to disappear in a short period of time.
That's just my 0.2 cents.
Best regards,
Francis

[Nicholas Cheung]

I'm advising people to NOT waste their time and resources to learn and take certification for something that is obviously going to disappear in a short period of time

In such sense, will you advice people not to take SCJP 1.4, becos J2SE 1.5 beta has been released, and the semantics of 1.5 has been changed a lot?
Obviously, SCJP 1.5 beta, according to Kathy, may come out at the end of this year, or the beginning of next year, while SUN has no plans on releasing EJB 3.0 exam, and even there is NO beta version of the software released.
Nick

[Goan Balchao]

I think we (people like me) maybe taking this in the wrong spirit. What Francis is so strongly advocating may not seem right to us - because of the fact that we have been using this for qute some time and a shift is always inherently unpleasant however viable it may be.
No doubt there are situations where Hibernate and Spring make sense - but I woudn't use them right now .. althoughj I have no qualms about learning about them.
Also Francis as to your query as to why Sun may have copied over things from the Hibernate and Spring frameworks ( I don't have knowledge of my own that they have done so), but trying to incorporate new things into existing proven frameworks to improve on them isn't a bad thing is it ? Necessity is the mother of invention as they say.
Obviously there were areas that could be immproved on and Sun or the JCP is trying to improve on that. Dosen't mean that Hibernate and Spring have superceeded all the app servers out there. Morever if you are concerned about vendor lock in - wanting to pay the big boys big bucks -- there's always Tomcat and Jboss to consider ...
If you are actively using Hibernate and Spring and in your experience you find it to be more perfomance oriented and reliable and scalable than the app servers out there today, please share this information with us so that we can imake an informed decision too !
Thanks.

[Valentin Crettaz]

As you may know we all get wiser as the years go by and turn to do things diffrently and better thereafter. In my 2 1/2 years of EJB development I've learned a lot in practice. I've learned that certain things in EJB are just evil. The specs developers made bad design decisions and they know it now. Hence the radical overhaul of EJB in version 3. They've seen how bad it was and are correting themselves.
Why can't I make a change of mind along the way? --Francis Amanfo
You are free to, of course I never had the intention to prevent you from changing your mind...
On the other hand, 2 1/2 years of practice with EJB should have taught you that companies are sometimes not able to evolve as rapidly as the technology they use. You know that story of the big tanker sailing across the sea. Once it's in movement, it takes a substantial amount of time to make it take a radical right turn...
I know of plenty of organizations whose applications are still implemented in EJB 1.x. When they decided to migrate to 2.0, 2.1 way already out and now that they are almost over with the 2.0 migration, 3.0 shows its nose... If you have a certain sense of pragmatism, you should realize that even when EJB 3.0 will be finalized, it will take a certain amount of time for people to evaluate whether they need to go for EJB 3.0 or not.
People learning for the SCBCD exam as it stands now will most certainly not waste their time. At least, they will have gained plenty of knowledge about EJB and most important of all, they will know WHY the spec has evolve to a 3.0 version and what was it that the 1.x and 2.x specs didn't have.
Bottom line: In order to solve problems efficiently, it is not only important to know the technology as it is today, it is also important to know how it was yesterday to better understand it!!
PS: I thank you for giving me some blog matter for today
[ May 11, 2004: Message edited by: Valentin Crettaz ]

[Francis Amanfo]

Originally posted by Valentin Crettaz:
As you may know we all get wiser as the years go by and turn to do things diffrently and better thereafter. In my 2 1/2 years of EJB development I've learned a lot in practice. I've learned that certain things in EJB are just evil. The specs developers made bad design decisions and they know it now. Hence the radical overhaul of EJB in version 3. They've seen how bad it was and are correting themselves.
Why can't I make a change of mind along the way? --Francis Amanfo
You are free to, of course I never had the intention to prevent you from changing your mind...
On the other hand, 2 1/2 years of practice with EJB should have taught you that companies are sometimes not able to evolve as rapidly as the technology they use. You know that story of the big tanker sailing across the sea. Once it's in movement, it takes a substantial amount of time to make it take a radical right turn...
I know of plenty of organizations whose applications are still implemented in EJB 1.x. When they decided to migrate to 2.0, 2.1 way already out and now that they are almost over with the 2.0 migration, 3.0 shows its nose... If you have a certain sense of pragmatism, you should realize that even when EJB 3.0 will be finalized, it will take a certain amount of time for people to evaluate whether they need to go for EJB 3.0 or not.
People learning for the SCBCD exam as it stands now will most certainly not waste their time. At least, they will have gained plenty of knowledge about EJB and most important of all, they will know WHY the spec has evolve to a 3.0 version and what was it that the 1.x and 2.x specs didn't have.
Bottom line: In order to solve problems efficiently, it is not only important to know the technology as it is today, it is also important to know how it was yesterday to better understand it!!
PS: I thank you for giving me some blog matter for today
[ May 11, 2004: Message edited by: Valentin Crettaz ]

Vincent,
If you've followed the discussions going on around the net about the changes coming up in EJB 3.0, as our own Kathy characterised in one of her posts as BIG, you'll know that there are very few areas where EJB 3.0 will resemble the previous versions. Why would you encourage people to go and waste their resources on the current version? To mention a few, entity beans is dissapearing! JNDI is dissapearing! The deployment descriptor is also dissapearing! These are also areas the certification seriously examines. Why should you take people through these hells if in a year or two time they won't need them?
You also said, I quote:

In order to solve problems efficiently, it is not only important to know the technology as it is today, it is also important to know how it was yesterday to better understand it!!

I agree with this comment. But in this context it's not completely true because you don't need that extensive study of the current version, as the certification require, before you can understand and use EJB 3.0. I'm sure before you started with EJBs you had no prior knowledge of EJB Just a brief introduction to the current version and its problems would be sufficient. And I know many books that will teach EJB 3.0 would devote some time and space for this introduction.
The point is that EJB 3.0 has a whole different design. You may also need a whole new way of thinking to be successful with it.
I've already said it and I'll repeat it. Go learn Spring and Hibernate and you'll be certain that your time is well spent since most of the ideas in EJB 3.0 are borrowed from these frameworks. If later you decide you still want to go with EJB 3.0, learning it would be just a snap.
Happy Springing (http://www.springframework.org) and Hibernating (http://www.hibernate.org)
Best regards,
Francis
[ May 12, 2004: Message edited by: Francis Amanfo ]

[Valentin Crettaz]

Vincent
Valentin, thanks
Why would you encourage people to go and waste their resources on the current version?
"Waste" is not the verb I used.
If you've followed the discussions going on around the net about the changes coming up in EJB 3.0
I have followed those discussions and I'm still following them, don't worry. I'm very well aware of all the good stuff EJB 3.0 will bring us and I don't deny that
If you look at JSR 220 (EJB 3.0) carefully (section 2.11), you'll see that EJB 3.0 won't be out until J2EE 1.5 will be released. If you go see JSR 244 (J2EE 1.5), you notice that they have just started working on the spec and they don't expect the final release before late summer 2005 (section 2.13) which means more than a year from now provided no major issues delay the whole stuff.
Don't think that I spit on Spring and/or Hibernate, I very much like them, but my point is that if people want to get certified NOW, they will be able to *enjoy* their certificate for more than a year and maybe even find jobs with them (I'm not related in any way to Sun Microsystems Inc. )
I agree with this comment. But in this context it's not completely true because you don't need that extensive study of the current version, as the certification require, before you can understand and use EJB 3.0.
The problem I raised is not related to being able to understand EJB 3.0, but more to being able to understand what EJB 3.0 does that previous versions didn't. This is technological culture as I like to call it. When I talk to people, I appreciate when they know what they are talking about and when they are able to bring some historical flavors into the debate. My opinion is that a certification is primarly targeted at professionals, and to me, a professional is not someone who wants to learn for an exam as quickly as possible in order to get one more sheet of paper on the wall. Maybe I'm asking too much, but to be qualified as a professional, one has to understand all the whereabouts of the technology, and to achieve this, one must go through an extensive study of the technology, knowing the previous versions with all their flaws, issues, bugs, design problems, benefits, etc. At least, this is what I expect from any job applicant I have to evaluate in the frame of my job.
So, to me, learning EJB 2.0 now is neither a lost of time nor money, it's part of your job as an IT professional. I'm a strong believer of the idea that we will better evolve in the future if we are capable of understanding and assimilating the success and failures of the past.
I'm sure before you started with EJBs you had no prior knowledge of EJB
Which is true of anybody starting to learn any new technology. You have to start somewhere. In my case, I started reading the very first EJB spec introduced in 1998 even though the next one was already out. I just wanted to know how all the story started, that's all.
[ May 12, 2004: Message edited by: Valentin Crettaz ]

[Francis Amanfo]

Originally posted by Valentin Crettaz:
Vincent
Valentin, thanks
Why would you encourage people to go and waste their resources on the current version?
"Waste" is not the verb I used.
If you've followed the discussions going on around the net about the changes coming up in EJB 3.0
I have followed those discussions and I'm still following them, don't worry. I'm very well aware of all the good stuff EJB 3.0 will bring us and I don't deny that
If you look at JSR 220 (EJB 3.0) carefully (section 2.11), you'll see that EJB 3.0 won't be out until J2EE 1.5 will be released. If you go see JSR 244 (J2EE 1.5), you notice that they have just started working on the spec and they don't expect the final release before late summer 2005 (section 2.13) which means more than a year from now provided no major issues delay the whole stuff.
Don't think that I spit on Spring and/or Hibernate, I very much like them, but my point is that if people want to get certified NOW, they will be able to *enjoy* their certificate for more than a year and maybe even find jobs with them (I'm not related in any way to Sun Microsystems Inc. )
I agree with this comment. But in this context it's not completely true because you don't need that extensive study of the current version, as the certification require, before you can understand and use EJB 3.0.
The problem I raised is not related to being able to understand EJB 3.0, but more to being able to understand what EJB 3.0 does that previous versions didn't. This is technological culture as I like to call it. When I talk to people, I appreciate when they know what they are talking about and when they are able to bring some historical flavors into the debate. My opinion is that a certification is primarly targeted at professionals, and to me, a professional is not someone who wants to learn for an exam as quickly as possible in order to get one more sheet of paper on the wall. Maybe I'm asking too much, but to be qualified as a professional, one has to understand all the whereabouts of the technology, and to achieve this, one must go through an extensive study of the technology, knowing the previous versions with all their flaws, issues, bugs, design problems, benefits, etc. At least, this is what I expect from any job applicant I have to evaluate in the frame of my job.
So, to me, learning EJB 2.0 now is neither a lost of time nor money, it's part of your job as an IT professional. I'm a strong believer of the idea that we will better evolve in the future if we are capable of understanding and assimilating the success and failures of the past.
I'm sure before you started with EJBs you had no prior knowledge of EJB
Which is true of anybody starting to learn any new technology. You have to start somewhere. In my case, I started reading the very first EJB spec introduced in 1998 even though the next one was already out. I just wanted to know how all the story started, that's all.
[ May 12, 2004: Message edited by: Valentin Crettaz ]

Valentin,
I apologise for calling you Vincent in my previous post
Based on your opinions noted above, would you advise your brother or sister or whatever, who want to start programming in C/C++, Java or.NET, to first begin with Assembly, Cobol, Pascal, Lisp, [...fill the rest in here]? I'm curious to hear your answer
Regards,
Francis

[jeff mutonho]

Hi guys.I posted a question on sun's EJB forum about all this EJB 3.0 story and got a response from someone on the J2EE team.There is his response, which I think makes very good sense.

Hi Jeff,
In this industry, there's always a newer version on the way :-; I wouldn't hold off on EJB 2.x just because we are making improvements to EJB. First of all, the J2EE 1.5 umbrella JSR that will include EJB 3.0 was just filed, so it will be a while before the platform is released. Secondly, despite the fact that there will be some new features, EJB 3.0 will be focused on ease of use and ease of development. Developers will still *always* be able to use all aspects of EJB that were part of the 1.1 spec through the 2.1 spec. That is guaranteed by the J2EE platform, in the same way that all existing J2SE programs will work on J2SE 1.5.
Furthermore, the SCDCD aims to test your grasp of the core EJB concepts and programming model. It's never a bad thing to have that understanding.
--ken
J2EE Team
SUN Microsystems

[Valentin Crettaz]

Based on your opinions noted above, would you advise your brother or sister or whatever, who want to start programming in C/C++, Java or.NET, to first begin with Assembly, Cobol, Pascal, Lisp, [...fill the rest in here]? I'm curious to hear your answer
Yes, I would advise him/her to have a look at those languages and other older technologies... That d**n technological culture again
Otherwise, I guess ken's answer makes plain sense to me (Thank you Jeff for providing it )
[ May 12, 2004: Message edited by: Valentin Crettaz ]

[Francis Amanfo]

Originally posted by jeff mutonho:
Hi guys.I posted a question on sun's EJB forum about all this EJB 3.0 story and got a response from someone on the J2EE team.There is his response, which I think makes very good sense.

Jeff,
Of course you wouldn't expect one of the top people at Mc Donalds to tell you that their junk food is bad for your health? Similarly, since Sun needs cash so badly I did not expect them (Ken) to tell you that taking the SCBCD now is a bad idea.
So judge for yourself. EJB in its current form is doomed. Open your mind and read the discussions going on around the net from some of the brightest minds in our industry. Don't waste too much time though. Jump on the Spring/Hibernate bandwagon. You'll never regret it. Currently, it's the best framework for enterprise developments.
Best regards,
Francis
[ May 12, 2004: Message edited by: Francis Amanfo ]

[Roger Chung-Wee]

I am glad that Ken has affirmed one of the basic principles of the EJB 3.0 spec, which is that it is backward compatible all the way to the EJB 1.1 spec. We should therefore ignore uninformed statements about "stop learning things that there is evidence that it's going to disappear".
And a main thrust of the EJB 3.0 spec is to make it easier for developers. I'm pleased, for instance, to see much work being done on entity beans. New features include:
- a new type of persistent POJO entity bean (to provide transparent object persistence by standardising an object-relational framework with features borrowed from TopLink and Hibernate)
- enhanced EJB-QL by adding bulk update and delete, subqueries, projection, outer joins; the generation of dynamic queries at runtime instead of requiring all queries defined at deployment; and an escaping mechanism to allow direct use of your database�s SQL dialect where desired.
I summary, I believe that anyone who needs to learn about EJBs should study for the SCBCD exam now. There is litthe or nothing that you will learn which will be wasted, and you will be in a good position to take advantage of the new features of EJB 3.0.

[Kathy Sierra]

Just wanted to say that I agree with what Roger said, and everything that Valentin has said in this thread is *exactly* my position.
Francis, you make very compelling points--I think where we would argue with you is on the *time scale* in which these things will happen. If someone takes a *very* long term view, then it might not make sense to prepare for the exam, but at Sun, we know that both vendors and customers take an extremely long time to migrate to new specifications, regardless of how significant and beneficial the changes are.
There will not be a new EJB certification for *at least* one year from now, and most likely that will be much longer before there is a 3.0 exam!
Your point, though, is whether anyone should become certified on a version that will be going away, well, it isn't going away. It isn't going away because the 3.0 spec will *not* be an all-out replacement, but a significant enhancement. But that's not the main point--the main point is that EJB 1.1 and EJB 2.0 are what customers are using *today*, and will be continuing to use (although they might be using EJB 2.1 if they're doing Web Services), for potentially another 18 months to two years or MORE.
A lot of people would like to become certified *now*, in order to be doing what customers are using *today* and for the next year. Let's say a new exam comes out in 18 months from now... it won't be very difficult to learn the new stuff and take the short *upgrade* exam to take your certification from EJB 2.0 to 3.0.
As cool as the new stuff is, including the new frameworks, it just won't be mainstream for a long time, and at Sun we want people to certify in something that is practical and useful for at least the next 18 months, which is about the length of time that the industry has estimated is the valid lifespan of technical knowledge. In other words, most technical knowledge becomes obsolete in 18 months.
Given that EJB 1.1 and EJB 2.0 are currently in widespread use, despite the complaints, issues, and workarounds, it can still make sense to certify in it if you have any plans to *use* EJBs in the next year, or potentially work for a company that's using EJB today.
The chances of someone applying for a job and hearing the company say, "You mean, you only know EJB 2.0??" are quite small In fact, we are MORE concerned that an employer will say, "You only know 2.0? We're still on 1.1... do you have 1.1 experience?"
This doesn't mean that you can't also be learning about the new frameworks!!
I think there are two levels of technology:
* The thing you need to use TODAY
* The thing you'll be learning for the FUTURE
Learning and certifiying in what is in use TODAY doesn't waste your time... you can gain valuable knowledge and experience that you can then apply two years from now when 3.0 fully hits critical mass.
This is a good and interesting discussion, though!
cheers,
Kathy

[Santosh Ramachandrula]

I have been reading for SCBCD certification for around 2 months and now the EJB 3.0 comes up....
I have gone to the internet and read about the changes(I guess major changes) that are being made to EJB 3.0 .
Though it was a tough decision for me to stop preparing for SCBCD certification exam. I had to stop.
I have prepared hard for the exam for the last 2 months (I work full time and then study), but the knowledge I gained in preparing for this exam will not go waste.
I started preparing for the SCDJWS(Web Services)with following points in my mind (these are my personal, am not trying to discourage anybody from taking any exam)...
1. New technology.
2. Web Services has lot of other technologies in itself like XML, SOAP, JAXB, JAXP, Architecture, security, some EJB also.... so in the process of preparing for Web Services exam I can gain wider range of knowledge.
3. Very few people have good knowledge about it, I want to be one among them.
4. By the time I graduated from school with my Master's EJB and some other technologies had already matured and I think I can be with Web Services right from its start.
The knowledge of EJB2.0 that I gained in preparing will be with me untill EJB 3.0.
Kathy and Bert am looking forward to Web Services Certification book from you guys...
[ May 12, 2004: Message edited by: Santosh Ram ]

[Roger Chung-Wee]

I'd like to echo what Kathy and others have said: study for what you need today. I may be wrong, but it sounds to me as if you are trying to pick a winner. Even if web services does take off, there will be far more demand for people with web component and EJB skills than web services skills.
In my department, web services development is typically done be server-side developers. We don't recruit someone who knows about web services, it's far more likely that the existing developers on the project will just have to learn about web services and how WebLogic implements it. And I could say the same thing about UML, Rational Rose, korn shell scripting, Javascript, HTML, XML, SQL, ORACLE, etc, etc.
So, it's not a simple case of being either a web services developer or an EJB developer ...

[Jim Bracks]

Expert One-on-One J2EE Development without EJBby Rod Johnson
Another great book by the same author "Expert One-on-One J2EE Design and Development"here(detailed discussion on EJB.)
INTERVIEW
[ May 12, 2004: Message edited by: Jim Bracks ]

[Valentin Crettaz]

Thank you Kathy and Roger for your wise opinions.
Based on your opinions noted above, would you advise your brother or sister or whatever, who want to start programming in C/C++, Java or.NET, to first begin with Assembly, Cobol, Pascal, Lisp, [...fill the rest in here]? I'm curious to hear your answer --Francis Amanfo
Yes, I would advise him/her to have a look at those languages and other older technologies... That d**n technological culture again --Valentin Crettaz
Maybe I should also say why I would give such an advice based on my past experiences and education. Before starting to learn Java and C++, our professors at EPFL taught us languages such as Lisp, Prolog, ASM, Modula, Pascal, Ada, Smalltalk, etc. Not in their depths, but enough to understand the important concepts underlying them. That way we knew on which basis Java had been created and why the language was the way it was, with its advantages and drawbacks.
How would you teach someone the pros and cons of the object-oriented languages without talking about predecessor languages? How would someone understand why the software engineering community came up with the object-oriented paradigm without knowing which ideas and concepts were dominating before? To me, a technological school that teaches Java (or C# for that matter) in its software engineering curriculum without educating its students by giving them some prior historical background does not deserve to be called a technological school.
So, my answer would be: YES, I would definitely encourage anyone to learn past technologies before starting new ones. The problem with this assertion being of course the time pressure due to the ever-growing competition on the job market. We have to learn more, better and faster!!!
[ May 12, 2004: Message edited by: Valentin Crettaz ]

[Francis Amanfo]

Originally posted by Valentin Crettaz:

Maybe I should also say why I would give such an advice based on my past experiences and education. Before starting to learn Java and C++, our professors at EPFL taught us languages such as Lisp, Prolog, ASM, Modula, Pascal, Ada, Smalltalk, etc. Not in their depths, but enough to understand the important concepts underlying them. That way we knew on which basis Java had been created and why the language was the way it was, with its advantages and drawbacks.

[ May 12, 2004: Message edited by: Valentin Crettaz ][/QB]

Valentin,

Yes, I can see from this comment of yours that we more or less agree with each other. I said that people don't need extensive understanding (which the SCBCD requires) of current EJB to be able to understand and successfully use EJB 3.0. I also mentioned that just a brief introduction to the previous EJB and the problems they had, which any decent book that would teach EJB 3.0 would do, must be sufficient. Your professor didn't require you to study the full details (syntax, semantics, language grammer, detailed programming assignments, etc) of the languages you mentioned above? On the same note, we shouldn't encourage people to go through the hells of EJB in its current form if we know it'll disappear in a year or two. It is just waste of resources. I believe this way can be also counter productive because not only would people have to learn the new thing, but they also have to put in a lot of effort to unlearn the way of thinking they had already developed.

I sincerely think the best way is for people to start delving into Spring and Hibernate now. With that knowledge and expertise, when you later on decide to switch to EJB 3.0 the transition would be far smoother than coming from the current version of EJB. Because, boy, the changes coming are massive. That's why it's not EJB 2.* but EJB 3.0!

The fact is that technology is moving very fast. There is pressure to be proficient in new things quickly. There is so much to learn. Don't waste time learning things that would disppear soon. To stay ahead of the competition you have to make wise choices. That might entail taking short cuts. And the short cut in this case is starting with the source where EJB 3.0 is coming form. The source is Spring and Hibernate.

That's just my 0.2 cent.

Regards,
Francis
[ May 13, 2004: Message edited by: Francis Amanfo ]

[jeff mutonho]

Jeff,
Of course you wouldn't expect one of the top people at Mc Donalds to tell you that their junk food is bad for your health? Similarly, since Sun needs cash so badly I did not expect them (Ken) to tell you that taking the SCBCD now is a bad idea.

So judge for yourself. EJB in its current form is doomed. Open your mind and read the discussions going on around the net from some of the brightest minds in our industry. Don't waste too much time though. Jump on the Spring/Hibernate bandwagon. You'll never regret it. Currently, it's the best framework for enterprise developments.

Best regards,

Francis

Well , if he didn't work for Sun , what would your response be?What I'm trying to say is that Ken is making a valid point , irrespective of whether or not he works for Sun.What Kathy said is very true.Most companies take a long time to adopt to the "latest" releases.Where I work , we're doing our development with Websphere and all our EJB's are based on the 2.0 spec.

Open your mind and read the discussions going on around the net from some of the brightest minds in our industry.

I'm sure the fact that I'm searching for ideas on this issue means I have OPENED MY MIND.

Don't waste too much time though. Jump on the Spring/Hibernate bandwagon.

Been using those already.No need to "jump on the bandwagon".Anyway , that line sounds like a very familiar one from M$.

Ok I've said enough...let me get back to my studying for the SCBCD(based on EJB 2.0)....


jeff mutonho

[Valentin Crettaz]

I can see from this comment of yours that we more or less agree with each other

Well, that's your interpretation of it. I wouldn't claim that, though.

I won't go further into this discussion because all what needed to be said has been said. There are two camps (as always), this is not a sacred battle (at least not for me), so there is no point for me to keep going on. I appreciate your input and I'm sure people will be able to make a decision when reading this thread.

See ya 'around the ranch (obviously not in this forum )

[Amitava Basu]

Hi All,

Nice discussions/debates. I was observing closely. Finally I made up my mind regarding SCBCD. But before that a brief introduction on my programming career.

I am CS & E graduate, started learning programming with FOTRAN (1994) then Pascal -> COBOL –> LISP –> PROLOG –> Small Talk –> C –> C++ -> Java.
(the path advocated by Valentin)
Till C++ it was collage and academic projects. Started programming in Java in commercial projects science 1998. Got SCJP(1.2) in 1999 end. After that was involve in many commercial projects based on J2EE technology. Built a framework with a team of 4 people (not like Stratus but quite different, Neither open source nor intended for selling, Intellectual Property of the organization I work for). We built B2B applications based on this framework which provides $70,000 revenue per day, 3000 users per day…..running on free App Servers on open source OS flawlessly (no big boys, no big money spent, our team supports all the applications).

In the course of all this I learned may things but what I missed is, learning new technology’s like EJB. I never coded EJB until I started preparing for SCBCD. I thought SCWCD will be comparatively easy for me, and decided to go for SCBCD first. In our organization EJB is not yet an accepted technology (in fact my boss who is technically very very good don’t like it all, he advocates Web Services instead, [in fact we achieved distributed computing using web services]Cost Reduction ….???).

Still I will go for the SCBCD this month. My personal reasons below. Correct me if I am wrong.

1. My experience in other J2EE technologies + EJB -> Is a Plus. SCWCD (if required) Will be considered by a larger set of employer.
2. Many are now already started using EJB 2.0 and I believe will continue doing so. Even if EJB 3.0 includes/does not includes all aspects of EJB 2.1
2. Even if EJB 3.0 spec is radically different, learning EJB 3.0 will be easier/quicker with prior EJB 2.0 knowledge.
3. When EJB 3.0 accepted by the market I may not be coding, but will be in a position to design guide developers (expecting that will be my next role), knowing EJB 2.0 will have advantage (I am assuming?)
4. Sun needs cash so badly - $150 Will help them …

NB:- I really don’t know anything about Spring/Hibernate bandwagon. Will keep my eyes open.

-Amitava

[Kathy Sierra]

Originally posted by Amitava Basu:
4. Sun needs cash so badly - $150 Will help them

ev

Well, the idea is nice... but only a very VERY small portion of your certification fee goes to Sun. Most goes to the third-party companies that help develop, audit, analyze and administer the exams.

Sun's not in certification for the money. If they were, the fees would have to double.

But taking an exam does help support *Java* in general, and that's always a good thing for all of us.

cheers,
Kathy

[Kathy Sierra]

Originally posted by Francis Amanfo:
Don't waste time learning things that would disppear soon.

You make a great point... but I interpret the word *soon* quite differently. Two years, in my opinion, is not *soon*, but hey -- that's fuzzy logic anyway

Francis, I applaud your evangelism skills . I'm going to pay more attention to Spring and Hibernate thanks to your enthusiasm here. I think you've made your point quite clearly and articulately, and I definitely believe you did the right thing by posting your opinion and perspective on this, but now I really do hope you'll understand that many of the folks here have chosen (and will continue to choose) to stay on the current path because they simply cannot afford to learn something they can't apply today, and this forum is for encouragement and support. The information was valuable, but I don't believe it's helping anyone now to continue to tell people they're wasting their time. If they've made their choice, please, we want to be encouraging. This forum is for getting people through the exam; there are other forums more appropriate for this discussion now.

I appreciate what you're saying about there being so much to learn, and not to waste time... but I think you're not giving people here enough credit for being quick learners. Anyone in this business MUST be. I'm sure they'll have no trouble when the time is *right* to learn the new stuff, and they'll be well-grounded in the current technology. And once again, I'm MORE concerned that people are learning EJB 2.0 when even THAT is still in its infancy at many companies.

The jump from 2.0 to 3.0 is NOT going to be some terrible, hard thing to grasp! In fact, it'll be positive and fun, given that the whole thrust behind 3.0 is to be more developer-friendly

I'm looking forward to 3.0, but I've put that in a very deep background thread in my brain, because now is not the time. In fact, to really try to learn this NOW, in my opinion, would also be a waste of time. We do not yet know how even the new frameworks are really going to play out. Personally, I will wait until things are stabilized and have reached critical mass before I try to go there. I learned this the hard way studying the draft specifications of EJB 2.0. Remember, even very close to the end, there was a profoundly different CMP mechanism in the EJB 2.0 spec, and it was very painful to learn. People complained, and shortly before final release, the CMP 2.0 was changed DRAMATICALLY and I kicked myself for studying something that was beta and subject to drastic change or even complete elimination even close to final release.

I like to wait until things have settled down. It's just a different perspective, but with your same idea--there is only so much time to learn, so where do you spend your time? I prefer a more practical approach rather than studying the cutting edge that hasn't yet been adopted. The benefit to *waiting* is that more and more resources become available, more and more people are using the new stuff, and it becomes easier and quicker to learn after a certain amount of stabilization and maturity of the technology. So I'll come back with my argument that studying something to *early* can be just as much a waste of time as studying something too *late*.
But this is always the game we play with this stuff... and there *are* some things where I DO get excited and try to get a really early jump on it, so I appreciate what you're saying. I guess enterprise technologies just don't fall into that category for me because of how lumberingly, painfully slow most companies are to update

Thanks for the good discussion!

cheers,
Kathy

[Bert Bates]

Another cautionary note!

Think twice about buying a computer any time soon - word is out that if you wait 12-18 months you'll get double the speed, more RAM and a bigger hard drive, all for the same amount of money!

[Valentin Crettaz]

Francis,

As Kathy suggested, this forum has been created for supporting and encouraging people willing to take the SCBCD certification exam. You made your point, we are aware of it and we thank you for that. Now, this discussion has taken a direction that is more geared towards the future of EJBs than towards the value of the certification.

We had a small pow-wow with other sheriffs and one of them suggested that we move this discussion to the EJB forum, which is more inclined to handle this topic. So, people, rearm your gun and get ready for the second round

I'm happy to introduce you to Chris, our famous bartender in the EJB saloon. Chris, please meet Francis, Francis meet Chris.

You can continue this discussion at https://coderanch.com/t/314041/EJB-JEE/java/Advice-those-planning-SCBCD

Thank you for your comprehension.
[ May 14, 2004: Message edited by: Valentin Crettaz ]

[Francis Amanfo]

Originally posted by Kathy Sierra:


You make a great point... but I interpret the word *soon* quite differently. Two years, in my opinion, is not *soon*, but hey -- that's fuzzy logic anyway

Well Kathy, for me 2 years in our industry is indeed a very short time, at least when you look at it in the context of projects. How many EJB projects can an average size IT shop do in 2 years? Not so many.

Francis, I applaud your evangelism skills .

Thanks Kathy. Actually, I have a PhD in Evangelism . No, just kidding

... I really do hope you'll understand that many of the folks here have chosen (and will continue to choose) to stay on the current path because they simply cannot afford to learn something they can't apply today, ...

Kathy, infact, I hate to disagree with you but I can't resist it here . Hibernate is widely used in enterprise mission critical applications. It's even becoming the standard for the persistence layer of enterprise applications, in most situations replacing entity beans.
You may be a bit right in the case of Spring. But as you may know, EJB 3.0 has stolen ideas not only form Hibernate but Spring as well. So I can comfortably conclude here that these 2 frameworks have great ideas and are proving to be successsful.

The jump from 2.0 to 3.0 is NOT going to be some terrible, hard thing to grasp! In fact, it'll be positive and fun, given that the whole thrust behind 3.0 is to be more developer-friendly

I know this. But my point is people need not wait long or waste time and resources because the right meat is already here. And the meat is Hibernate and Spring- the source of EJB 3.0 .

We do not yet know how even the new frameworks are really going to play out. Personally, I will wait until things are stabilized and have reached critical mass before I try to go there.

Like I've already mentioned, Hibernate is a success story. It's proved itself already. In the enterprise world, the critical mass have long started moving in there. So guys, no more waiting. Jump onto the bandwagon!

I learned this the hard way studying the draft specifications of EJB 2.0. Remember, even very close to the end, there was a profoundly different CMP mechanism in the EJB 2.0 spec, and it was very painful to learn. People complained, and shortly before final release, the CMP 2.0 was changed DRAMATICALLY and I kicked myself for studying something that was beta and subject to drastic change or even complete elimination even close to final release.

Kathy, the specs developers for EJB 3.0 have indirectly admitted that they had it wrong and are now copying ideas from frameworks they know are successful and have been accepted by the community to replace their earlier bad design. So if you bet again this time, I can assure you your reward would be huge. I mean here betting on Spring and Hibernate. And if you ever decide to write a book on these frameworks, I'll be the first to grab it from the bookstore. Because my bet on your books have always been worthwhile .

I provide the links once again for your convenience.
Happy springing (http://www.springframework.org) and happy hibernating (http://www.hibernate.org)

Best regards,

Francis.

[ May 14, 2004: Message edited by: Francis Amanfo ]
[ May 14, 2004: Message edited by: Francis Amanfo ]

[C Chavan]

Francis,
I am not familiar with Spring or Hibernate. Could you plz provide some features of these frameworks that are better (in terms of ease of use, features-wise, API-wise) than Session EJBs and MDBs.

Also, could you plz highlight features from these 2 frameworks that are "stolen" into EJB 3.0 draft spec.

Note: I personally find Entity EJBs limiting in certain aspects and would prefer a pluggable architecture in J2EE to do persistence.

[Suresh Yadav Ramamurthy]

Please give a idea about SCJP-1.5 beta version exam.
bye
Suresh