PuneJava Blog

I had read a couple of chapters of their books online (exerpts from their site) and LOVED them!  I have purchased three of their books (Java, Design Patterns, and Servlets/JSPs).  I'm reading them all at the same time!  Great stuff!  Thanks for pointing out their blog.

--Dave

Java is still a great language for the experienced
developer. However, it does have a steep learning curve compared to many of the
other languages out there. Unless the IDE's make Java coding more VB like it
may continue to loose ground.

JAVA is an open standard with several large investments by industry experts Sun, IBM, HP. Thoses that are being lured into the .NET arena with MS are making a huge mistake. We've seen the consistent effort of MS to force 'proprietary' inroads into 'open standards' over and over. JAVA will continue to mature.

It's not clear what the big fuss about LAMP is since you can just as easily have Linux, Apache/Tomcat, MySQL, and Java as well.  PHP is great if you like scripting and what not, but a true enterprise application is either going to be .NET or Java.  If you pray to Bill Gates every day then .NET is the solution for you, otherwise there is no equal to Java.

Java is a great technology.  I agree with the previous
post stating the learning curve is very high.  This is true, and it
doesn't get easier because then you start throwing other technologies in the
mix such as struts, tiles, hibernate, etc.  I believe this is a shame
because although you can go nuts and make an application extremely complicated
there is no reason to do so.  In general, I believe people working with
LAMP technologies have a let's-build-it attitude while Java developers have a
let's-build-it-right attitude.  The former (in my opinion) leads to
rapidly developed application which get the job done but not in the
"right" way, with "right" being things like design
patterns, code reuse, test-first development, etc.  Where as the
latter (again, in my opinion) leads to hugely complex systems because of trying
to do things "right".  For enterprise systems you need the power
and flexibility doing things "right" gives, but you shouldn't start
there.  Unfortunately, resources teaching Java technology tend to push you
there first.  I believe the Java world
would benefit from instead pushing the simpler let?s-build-it attitude.  Later, if you?re online recipe database
becomes world famous then you can worry about test-first Spring dependency
injection with an EJB business tier and hibernate persistence.

Java, C, C++, C#............Assembly....get the job done, thats it.
Interpreted and or scripted languages do their particular well thats
why we need them. You are seeing a moves towards LAMP on mobile devices
etc ( the Motorola E895 Linux based mobile see ).There are billions of devices out there,
using many different language variants, so 'learn just one at your
PERL'.

LAMP, Small Business Server ...etc are platforms that applications run
on. The language choices used to deliver applications should be
determind by
"what get the job done, on time within budget and is cost effective in
implementation and maintenance".

From me programmers need to have a full language toolbox, and that
means being familiar with as many languages as possible, used across
the many different platforms. Java isnt legacy and neither is Assembly
just look at that mobile in your hand, you will find it has an OS, and
Applications each programmed in different language that was fit for
purpose.

For me the JCP process means developers have some input into language
development and Java pays my bills and will do so for the next couple
of years. That doesnt mean that I ignore C#, C, C++ or Assembly my home
library is evidence of that. So as I said before 'learn just one language at
your
PERL'.

Sun is paying price for ignoring the contemporary thinking such as Agile
Development, Extreme programming. It pushed for sub-standard
technologies which caused pain to developers (Entity Beans is an
example). It called dirty work-arounds as J2EE design patterns.
It forgot that Java is not greater than OOP and J2EE is not equal to
EJB. Instead of carrying forward the great language such as Java (which
was much ahead of C++ when it started), it played in hands of big
companies such as IBM, HP etc to make compromises on specifications. It
is learning from its mistakes but a lot of ground has been lost and I
think top position in programming language is vacant and it remains to
be seen what happens in next 1 or 2 years.

Java is doomed because it has become a victim of its own
success. It is too hard for Newbs to get going in Java to the point
that they be productive on complicated, enterprise-class
systems. Not only do they have to learn the language, but they have to
learn the framework (Struts/Shale/Spring/etc), the DAL (Hibernate),
threading, unit testing, etc., before they are even productive.

IDEs help a little (if you're still writing getter/setter
classes by hand or living without auto code-completion, then grow
up and start using Eclipse), but they can't help the Newbs
understand the concepts that are essential for writing Java apps.

When an application is built and goes into maintenance mode,
this is where doing things "right" pays off. By adhering to the MVC
model, making changes to business logic or the presentation layer
is much easier than trying to wade through spaghetti code without
messing things up. LAMP projects are easy to roll out, but fall
down on the maintenance for this very reason, which is why they are
not suitable for enterprise development.

Things like Ruby/Rails looks promising, because it combines the
ease of coding with the "right" way of doing things. Yes, it is
still possible to make a poorly-constructed application using
Ruby/Rails--which is why one hires talented developers instead of
Schlubs.

I picked up Java about 9 years ago, and to me it is like a
well-worn, comfortable shoe. It will be a hassle to give it up as
my primary development tool, but I see the writing on the wall:
Java won't go away, but as the majority of Newbs gain experience,
it won't be in Java, so the emphasis will shift to something else.
As the Pragmatic
Programmer advised, "Learn at least one new language a year."
This is very good advice.

Although all the comments are right in my opinion, the fact that we are even discussing who is better in the programming language is a sign that not one of them can suit every need, each one has come to be because someone out there decided that the tools available lacked certain features, too complex, etc. In my opinion  Java and .Net are in the same category competing. Perl, PHP and others are competing in another category. And history has shown us that each language has evolved in it's own group, improving and borrowing ideas from others, ie .Net borrowed a lot of concepts from java, java borrowed from C++ and so on. In summary I think java has a bright future, so does .Net PHP, and others...

Although all the comments are right in my opinion, the fact that we are even discussing who is better in the programming language is a sign that not one of them can suit every need, each one has come to be because someone out there decided that the tools available lacked certain features, too complex, etc. In my opinion  Java and .Net are in the same category competing. Perl, PHP and others are competing in another category. And history has shown us that each language has evolved in it's own group, improving and borrowing ideas from others, ie .Net borrowed a lot of concepts from java, java borrowed from C++ and so on. In summary I think java has a bright future, so does .Net PHP, and others...

Although, the learning curve can be steep with JAVA there are just too many holistic systems out there to even consider branding JAVA as legacy software.

For example, in our Enterprise environment we use a middle-ware product to create JAVA wrapper classes to expose properties of 4GL Natural Programs that run on a Z/OS based mainframe. (64 bit processing) We then access our mainframe data residing in ADABAS via an Intranet web front end. The web front-end portion is scripted in Java Server Pages (JSP) and uses the JAVA wrapper classes to communicate with a middle-ware broker running on our mainframe. I also forgot to mention that data received back from our mainframe is written to our web server as XML using a JAVA routine.

I think many technologies suffer from lack of proper marketing and listening to the valued consumer. Just look at the similarities in syntax between C# and JAVA? Now ask yourself this. Which company provides free technical training on their products and even gives away gifts?

So the answer is yes, JAVA is legacy software when it comes to understanding the customer and giving them what they perceive as a necessity. The good part is that the tech industry is volatile and competition drives product lines. So it?s a coin toss at whose product is better one day or another?

want the power of Java without the complexity ? Consider Adobe/Macromedia ColdFusion! Its a scripting language at it simplest like LAMP, but can form the the presentation layer of a J2EE app if you need to do it 'right'.

It can talk to .NET via COM and it it runs on top of java so you have all the power of java, with the simplicity of a scripting language and powerfull integration with flash.

dont get me wrong, there are times when EJB's and a full model 2 app makes sense, but then there are times when its just overkill. i feel cold fusion gives u the best of both worlds.

Although, the learning curve can be steep with JAVA there are just too many holistic systems out there to even consider branding JAVA as legacy software.

For example, in our Enterprise environment we use a middle-ware product to create JAVA wrapper classes to expose properties of 4GL Natural Programs that run on a Z/OS based mainframe. (64 bit processing) We then access our mainframe data residing in ADABAS via an Intranet web front end. The web front-end portion is scripted in Java Server Pages (JSP) and uses the JAVA wrapper classes to communicate with a middle-ware broker running on our mainframe. I also forgot to mention that data received back from our mainframe is written to our web server as XML using a JAVA routine.

I think many technologies suffer from lack of proper marketing and listening to the valued consumer. Just look at the similarities in syntax between C# and JAVA? Now ask yourself this. Which company provides free technical training on their products and even gives away gifts?

So the answer is yes, JAVA is legacy software when it comes to understanding the customer and giving them what they perceive as a necessity. The good part is that the tech industry is volatile and competition drives product lines. So it?s a coin toss at whose product is better one day or another?

I was a Java developer for about a year before I was a .NET developer.  When you're immersed in a technology and everyone you're working with in a technology, it feels like that technology is the center of the IT world.  It was the same then with Java as it is for me now with .NET.

Saying either technology is doomed seems foolhardy.  Both technologies are strong right now, and any prediction at this point is pure conjecture.  It's true one of both of these technologies may die at some point, but when?  With the strong showing both technologies are making right now, I would not be surprised to see both of them alive in 10 or 20 years.  One of both of them may even become the COBOL or C++ of the new era of software tools, outlasting all others and outliving their expect lifespan.

With that said, I'd like to state my personal preference and say that I love .NET.  In the last month, I've written my first web application, a small application in VS2005, and began work on my first mobile application.  When a tool has a "steep learning curve" like Java, while another tool gives you the same capabilities (on the Windows platform) then the easier tool is the one I will use.  When working in Java, there were lots of minor quirks that drove me crazy (albeit mostly with the WSAD and JDeveloper IDEs).  I don't have those minor annoyances anymore in .NET.  IMO, .NET is really a superior tool.

My prediction is that both are going to be around for many years, and that's good.  I wouldn't want to live in a world where there was only one software company to choose from.

Java Hard?

When I started using Java, I had been doing C/C++ from the commercial software
business.  Difficult of a language or
technology is relative to where you start. 
What people are comparing Java to is what is referred to as scripting
languages.  :-)

PHP is great. 
However, it is not meant for everything. 

I'm not the programmer I used to be when developing interpreters and
compilers at Watcom over 20 years ago - my exposure is more in the
rolling out of solutions. I have worked with LAMP - easy to implement
because the people doing the implementation know it. My stuff tends to
still be the ancient tools - shell scripts, C, C++, Fortran, Pascal and
when I want to ensure some portability, Java. If I had a compiler for
it available, I'd possibly even resort to Cobol for certain tasks. That
said, that is for stuff that I want to use, and don't want to have to
do by hand - not business critical. Not like the days when the
University of Waterloo used Fortran (WatFor) to write a precursor to
their C compiler.

That said, I still find that when things need to be secure from the
user's desktop to the client's server, and more responsive than just a
form, Java is what gets used. The only .NET I've encountered in any
solution outside Microsoft is actually the Beagle search system on
Linux, written that way because it is from the Ximian group of Novell,
who developed Mono, the Linux implementation of .NET. Java is what
seems to be used to reach beyond the limitations of scripting and
browser when desktop applications, or worse yet, Citrix distribution of
applications (lets bog down the central box with multiple instances of
a large and often not that efficiently written application, and the
network with sending the visual interface to dozens of desks.)

My experience is primarily driven currently by in-house applications -
putting things out over the Internet to consumers or business clients
raises different priorities - quick loading and quick response rather
than an application kept active for hours at a time. If your only tool
is a hammer, every problem looks like a nail - to lose access to tools
that best handle the task relative to the most popular current fad is
to lose out on efficiency, power and effectiveness - otherwise Java
would have replaced all of the Cobol code coming up to Y2K.