MyPortal
... professional open source J2EE web portal system ... ...

Brief Introduction
MyPortal is an open source Java/ J2EE web portal system (JSR 168 compliant). It is a fork of the Liferay Enterprise Portal. We aim to build a non-EJB version of the Liferay Enterprise Portal that can be run in a standalone servlet container such as Tomcat.
 
The Project
Status
Roadmap
Functionality
Technology
Why Non-EJB ??
License
About Us
 
Technical BLOG
 
Open Source Process
CVS vs. New Developer
Debugging
 
Development Team
Software Architect
Kelvin Yap, Malaysia
 
Developers
Chew Hock Ping, Malaysia
Lai Yick Hiung, Malaysia
MK Tan, Malaysia
Vincent Wong, Malaysia

 

 


Why Non-EJB ??

Rationale for making a non-EJB version of Liferay


Enquiry from McGovern, James F

Was searching Sourceforge when I came across this list. Curious why you would want to put a lot of effort into converting Liferay to a non-EJB implementation? Isn't this religious in nature?

Wouldn't the time and energy spent be better served by creation of additional functionality that could be used by all?

How about creating better hooks into LDAP or support for WSRP?

Inquiring minds would love to know.
  


Reply from Kelvin Yap

It is not religious, maybe I would say... it is Nationalism...

here is why...

The initial project members are all Malaysians. You know, Malaysians are just too young in the global software industry and the open source world. I want to contribute my little force to my country's software industry, by mentoring the less experience programmers, in the open source project. Our youngsters are all capable, they just need some "mentoring" and a "test bed".

Yes, I make this project a test bed, hence, I have no intention to use this project to replace the Liferay portal. Maybe we can complement the Liferay development by providing a "lite" version, and act as the "many eyes" that can possibly find some weakness of the source code and then provide patches to it. As our project also released under the MIT license, the Liferay crew can always come here to merge our code, if they think appropriate.

Since this is just a test bed for a group of programmers from a particular country, i didn't ask "any" hand from the existing Liferay community. My current project members are all new to the open source world, in other word, "I create new blood".

ok, that's the background story, let's turn to the technical consideration now... why non-EJB??

The major reason is because the learning curve to EJB is too difficult for Malaysian younger programmers. This will stop them to enter J2EE at all !! I have already mentioned, this project is a test bed to "help" Malaysian programmers. So, I must make it suitable for "entry-level" programmers to participate, let them grow from there, then contribute to the country's software industry and possibly start or take part in other open source projects, maybe contribute back to Liferay :-) I hope to inspire some Malaysian younger programmers to be the leaders of the next generation open source projects.

To be honest, I must admit the failure of Malaysian education system, in particular to the Computing Education. Our children are too much "spoon feeded", they can't really "think independently", nor dare to "adventure" nor "invent" nor "innovate". The lecturers in the local colleges and universities are generally getting low pay. So, this doesn't encourage them to do more academic research, hence, they can't deliver high quality Computing Education, often too outdated.
  


Reply from McGovern, James F

Couple of thoughts.

  1. Your rationale as contained in your email would make for great commentary on the home page.

  2. I am a firm believer in open source and the best way to learn to write good code is to start by reading code created by others. The failure in this simple truth also occurs in US based universities. I have been attempting to sway those CS professors from getting people to start writing code after the second day in class. One must learn to read before they can write.

  3. I agree partially with your statement on EJBs as being complicated. I choose to disagree in the method. I would avoid EJBs not because they are complicated but because their usage is shrinking and it is important to a work force to go with the tide.

  4. By focusing on the creation of portlets, you would position developers in your country to not only learn uncomplicated technology but something that is growing in importance. It is important to the careers of those in your country to not only have written code but to know that it is being used in production systems for major Fortune enterprises. Focusing on portlets will help secure jobs quicker than current approach especially in outsourced situations.

  5. Overall, I think your motives are good. Would love to see something happen similar for my own home country of Trinidad. Hopefully I can find a couple of people from there that can work along with you.
      

Reply from Kelvin Yap

Thanks for your input.

You are welcome to introduce more developers to this project :-) There is essentially no border in the Internet and the open source world. Small countries also can make some difference... we all know the little penguin in Finland, don't we?
   
posted at 10:38PM +0800, Wed, 25 Aug 2004 by Kelvin Yap

 


Copyright © 2004 My-JUG.org. All rights reserved.


SourceForge.net Logo