Post by BuchPost by pnicholsNo but who would the manager be? That is the problem.
Obviously, IBM :))
That is what they would like for sure. I do not believe it would be wise,
prudent, nor justifiable for one company to invent something, take all of
the R&D cost on themselves, and then hand it over to another company. I
surely would not do that, without some major bucks flowing my way :)
Post by BuchPerhaps, IBM remote controlled detachment of their people, like Eclipse.
I do not WANT Java like Eclipse. As stated previously, I HATE the Eclipse
IDE. It is something akin to "Lets go back to the stone ages." If Eclipse
was the best IDE Java had to offer, well I might have to say, "DOT.NET
anyone?" :)
Post by BuchThat is the way I like it. Team should have a leader. Collaboration does
not imply anarchy.
Yes, without tight control, Java becomes multi-flavors and loses the beauty
of Java itself, namely "write once using common APIs and libs, run
anywhere."
Post by BuchThen you should choose one, or use both, keeping in mind situations where
usage of one is better than another.
Having more than one choice is alwasy better than not having a choice.
There is a time when there are not any more place for improovements in one
framework, so another should be created.
True, but considering this (Struts and JSf) as an example. They furnish the
same type of basic framework structure. It is kind of like saying we will
have sixteen different frameworks, that basically do the same thing from
sixteen different places. No thanks.
Post by BuchLook at this like situation with database middleware in Delphi. It would
be bad if only BDE existed.
Yes, but for a time, the BDE was standard. No one has stopped or is stopping
others from coming up with additional frameworks for Java, but you cannot
couple everything into Java, nor should it be that way. I do not want a Jre
that is 50 meg in size, to support a plethora of services that are
basically duplicative in nature. Where would it end?
Post by BuchNobody said that everything should be in core. Core is core, and if it is
under reasonably tight control, like you said, it would be ok.
Perhaps we will have "Java distros", some including all, some including
subset. But as long there is one common denominator, it is OK.
God, I hope not. I do not want a Suse Java, a Mandrake Java, a Red Hat Java,
and/or an IBM Java, a Sun Java, a MS Java, a Debian Java, etc. To be
personally frank with you. if this is the way Java was today, I would not
be using Java and would have no plans to adopt it at all. If I desired a
multi-flavored programming language and compiler scenario, I would totally
adopt C/C++.
I use Java because it allows me to code once for many platforms and because
it is consistent across platforms, If it becomes an everybody has their own
distro, type of programming language and platform, I would definitely
migrate to something else.
I do not want Java to become like DOT.NET either. I do not want an MS
version and a Mono or GNU version of DOT.NET, which may or may not be
compatible, either. I want what I have with Java today;-- one Java, one set
of APIs and one consistent set of libs, that runs the same across all
platforms. Who makes the JVM or jre is not important. What is important is
that regardless of whether I choose to use an IBM jre, a Sun jre, or a
Blackdown jre, my code changeth not.
Post by BuchPost by pnicholsI do not mind these alternate frameworks being available, but I do not
want all of them part of the core of standard Java. That is the point. Do
you want 15 different ways to implement Data Abstraction cores to be part
of
No. Perhaps it would be good idea to remove ejb and else from core and
make it more optional ...
EJB is essential to Java. Without it, Java is woefully incomplete, like
Dot.Net is today. Read my post of Delphi NG, about Dot.NETs inefficiency in
terms of not having a persistent framework like EJB. I agree, EJBs are
sometimes overhyped and made to sound like a cure all, but for Enterprise
type development, they are essential, or at the least something similar to
them.
Post by BuchBut where did you saw this "throw it all" approach. I have seen quite the
opposite .. "keep core under control" kind of.
So, isn't that what the JCP is about currently? Why change from the JCP to
GPL? That is what I just don't get.
If we want to remove Sun's tight control, Ok. But I see the JCP as a good
working model. Amend it perhaps, but no need to start all over again.
Post by BuchIf you look it that way, then Java, and IT in general, are already
bloated. We are talking about Sun letting its control over Java to some
kind of open source community.
Frameworks are already "included", in a sense.
Yes, and THAT IS THE PROBLEM. We know MS is NEVER going to play with the
rest of the world, by choice. The only way MS would ever agree to having a
level playing field is if they totally and completely control it. There is
no need for DOT.NET for instance. The only reason for DOT.NET is that MS
could not control Java. DOT.NET is nothing more than MS' version of Java.
That doesn't make DOT.NET all bad, it does a lot of things well, and
improvements to the Java core have been made. But that was not the purpose
of DOT.NET and still is not. DOT.NET is all about MS controlling what the
development community uses, so that all roads will lead to Redmond, WA.
It is when we embrace proprietary solutions, that we are going to be forced
to deal with complexity. If we were all using one OS that was open (like
Linux), then the APIs, GDI, and kernel would all basically be the same.
There would have been little need for Java, or Web Services. We would be
free to remove complexity from the application core, because the OS core
would be homogeneous. However, I never see this happening until or unless
the developer and user community demand it. You can bet that MS would never
accept this without being forced by the developer and user community to
accept this. To be fair, Oracle, IBM, Sun, etc. would not desire all of
their systems to be this open either (at the Database or Server Application
Level).
Post by BuchIt is. Look at it this way: Big companies often change top management,
because after 2-3 years original one implemented their ideas, and
perhaps/most likely has not knew one.
Fresh blood. Well, its about time to get another management. :)
However, there is a great deal to be said about the "creator's genius" and
the ownership pride of success. IBM for instance, is not known for making
the complex simple. Ever tried to read one of their manuals? :)
Post by BuchPost by pnicholsThe only thing Eclipse has going for it, is it is free, open, and darn
well supported through the plugin community. What is has against it, is
that it is ugly, unnatural, and beotch to navigate with. It reminds me too
much of Visual Age.
I'm experimenting adapting JDT (that is visual libraries used to display
Eclipse0s IDE) for database interface.
I really like how it looks, and has lot of eye candy that can attract
customers.
Customers like eclipse's look, because of swt. I was careless enough with
laptop running eclipse, while waiting for a meeting.
When they saw it, they asked if I can make their apps looking like that
... they are addicted to windows l&f. Looks like mee too.
SWT is nice, it is the IDE that makes me wish for Borland IDEs :)
Post by BuchPost by pnicholsWhat you said about Sun is dead on. Sun has always been an engineering
company. Great on implementation, but slow on marketing.
Well, they finally made good steps, like PCs and desktop OS...
Post by pnichols"baby out with the bathwater", either. I do not want Java GPLed at all!! I
can guarantee you, this will kill it in the Enterprise.
You can't, I'm sorry - Linux is GPL, and its getting more and more
enterprise.
Why? Do you ship the Linux OS as an application? Do you really think
organizations are going to want to ship the source code to their internal
and external application customers who use them? I can guarantee you, they
will not.
It is much different at the OS level, where the OS is a commodity. In other
words, the OS is in use by many and is non specific between organizations.
Applications can be general use applications, such a Word Processors,
graphics programs, etc., but most are environment and business specific.
I would not want my Order Entry system for XYZ company to be open for all
other companies to use and exploit. In fact I know of no major business
that would say, "Take all of our applications that we have developed, and
any company or user than use, exploit, or extend our framework." Security
and specificity are null when this happens.
That is what I meant by their being a huge difference between an Open OS and
an open application development platform. Would you want to make all of the
source code available for every application you write and/or distribute? If
it were to ever come to this, start looking for a new career. If the OS
community were to try and force this, MS would win, for sure.
Post by BuchJava going open source is part of larger picture.
If you remember, Gates said (few months ago) that he predicts very cheap
hardware.
Would that not be nice for Microsoft, they sell software at the same price
as today, put people buy more of it because hardware is chepaer, therefore
more affordable.
Certain 3 letter companies :) , on the other hand, wants very cheap or
free software, and same price for hardware as they are today, for the same
reasons.
That is why IBM and others (Intel too, but more covertly :) ) support
Linux, and thats why they want open source Java.
Not all of them WANT Java open sourced. Look today at JBoss announcement for
instance. They are voting for tight control over Java, not necessarily Open
Source.
Post by BuchPost by pnicholsDEFINITELY BE RULED OUT.. So what type of license should an open source
Java carry? Another major concern I have.
Mozzila? (mozilla, mozzilla, mozzarella? :) )
Something that allows free use and does not demand releasing application
extensions, I can live with. GPL, for an application framework, I cannot
live with.
Post by BuchPost by pnicholsPersonally, I see this as the beginning of the end for future software
development. Are we going to trade all professionals and modest profits for
Why? There will be still need for customizations, integration, transitions
to new platforms, frameworks ... plenty of space for application software
developer.
I disagree. If all software is free and developers are not paid to write it,
why would a company hire a bevy of programmers, when they can just send
request to OS community at large?
Post by BuchPost by pnicholsLet's face it a guy with a ninth grade education is not going to become a
neurologist, heart surgeon, nor a very good programmer. Without incentives,
Yes. However, best programmers I know have high school only, without
college.
Well, I would disagree with this as well. Most people with HS only diplomas,
are hackers, since their exposure is not in an organizational sense. Of
course I would agree that experience in software development, especially
within the framework of a corporate IT structure, is more valuable than a
formal degree in many situations.
However, most engineers have some type of an engineering background, or
formal education which prepares them for their specific tasks. But I did
not mean this in terms of academic attainment, but rather in terms of
aptitude and intellect. Education can come in many forms;-- formal and
self-instruction.
I would agree that many BS and BA holders from formal university settings,
are not as well prepared, nor as equipped, right out of school, as some of
their tech type school counterparts. Tech schools usually place their
emphasis in the area of concentration, not on academia in general.
Post by BuchInteresting thing, they could not pass college at all, although they were
more intelligent and knowledgeable about programming then their
professors. I'm currently few months behind graduation, after many years I
finally found mental strength to endure - will that make me less good
programmer? (lol) (How did you ever managed to get Ph.D., you look too
smart for that :)) )
LOL!! I agree that many academic programs push for academia over
practicality and real world practical usage. It is also true that many
professors are woefully lacking in good IT skills. The reason that this
disparity exists, is understandable however.
Look at the salary that a college or university IT professor receives,
verses what a senior level programmer or R&D Team lead makes. No rocket
science required here. :) This is why many tech type school students are
better prepared than their liberal art counterparts, as well.
Many of the professors in a tech school environment, are still engaged in
the real IT world, and are not full time professors, or most of them been
IT professionals before they were engaged in formal academic endeavors.
Therefore, what these professors know and convey is not based upon an ivory
tower "religious" experience, but rather in the real world where "elitism
and liberal musings", do not function well. It is fine to try and
understand why 2+2 has to always equal four or to change the world of
constants in your ivory tower, but if you spend your time in such dribble
while attempting to be productive in the real world, you will seldom, if
ever, get anything useful or productive done.
Many professors are so busy being ethereal and esoteric minded, that they
are useless in the real world where things are as they are, not as you wish
they could be.
Have a good one..
BTW, my Doctorate is not in IT. Is is in psychology, which does help me to
understand myself as well as others. Basically, I discovered, any
programmer is a little off to begin with, starting with myself. :):)