Hi,

I haven't noticed anyone mention that EJB/J2EE is just one in a long line
of technologies that have been applied to problems that may not require
them.  Some applications are built on top of CORBA that could have used
TCP/IP sockets, and some applications are built on a commercial RDBMS
that could have used flat files.

Some of this is caused by vendors that overhype their benefits.  Some of
it is caused by developers who only apply what they can learn quickly,
and avoid capabilities that they don't understand (i.e. if I already know
JDBC and servlets, why bother learning EJB and JSP).  John's point is also
a good one: today's requirements aren't the only driver; tomorrow's
requirements (both real and potential).  Another factor is the tradeoff
between technology and system administration.  One of the major reasons
that many companies try to standardize on a platform is that as the
number of different deployed technologies grow, it becomes difficult
(if not impossible) to find sys admins to keep it all up and running.

One last thought has to do with the Fear, Uncertainty, and Doubt factor.
Which company stands to benefit the most from doubts about the return
on investment available from J2EE technology?  IBM? BEA? HP?  Any of the
open source vendors?  Don't think so.  What about Microsoft who's preparing
for the first release of .Net?  Anyone out there put it past the evil
empire to fund a Gartner study on how well J2EE has fulfilled its promise
over its first 3 years, then permit them to publish it.  I'm not suggesting
that this IS the motivation, just suggesting that it could be.

Charlie


> -----Original Message-----
> From: Winston Gnananayagam [mailto:[EMAIL PROTECTED]]
> Sent: Friday, August 24, 2001 12:54 PM
> Subject: Re: Decline of the EJB civilization?
>
>
> John,
>         I totally agree with you that its the job of sales people to sell
> products and marketing hype, spin are part of it. Yeah, its in the best
> interest of "Sun" to say that EJB, JINI , JDO & others are complimentary and
> are not parallel technologies. This way we buy all of their product lines.
> It suits so well for them to make a marketing pitch like that. Yeah, its in
> the best interests of the commercial EJB vendors to push technologies like
> Entity Beans.It is more complicated to implement and maintain and so they
> earn their money by providing support for it. In fact most of the EJB
> vendors differentiate themselves from others, in the kind of support their
> products give for Entity Beans(including clustering). It definitely suits
> them to make a sales pitch of this sort, targeted towards developers.
>        In the end its upto the Manager/Architect to dig deep through all
> these layers of marketing hype & spin to implement what they actually need.
>         I wouldn't even make an analogy with the commercial vendors as big
> SUV's or trucks and the Open source vendors as the Civics & Mini's. I would
> compare the Open Source vendors more to public transportations like trains,
> buses or aircrafts. Yes, public transports would take everybody in a less
> fancier way to the same destination. But, yeah it would take you there
> faster, cheaper & with much less frustration. Still, some of them would
> insist that they need to hear music in their six speaker stereo system while
> going to work. As you can see both these parallels will always exist in this
> world, just like MS & Linux. No point in taking sides here, its upto us to
> get and use what we exactly need.
> Winston.
>
>
> ----- Original Message -----
> From: "John Bateman" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Friday, August 24, 2001 11:35 AM
> Subject: Re: [EJB-INT] Decline of the EJB civilization?
>
>
> > Hi
> >
> > My comments to this article are that alot of the people buying the
> products
> > may not be sufficiently informed to buy 'only what they need'.
> >
> > Marketing hype and vendor loyalty do go a long way in getting sales.
> Support
> > could be another reason. I KNOW I'm going to get better (but more
> expensive)
> > suport from say IBM than some of the Open Source free projects out there.
> > Calling someone you have a contract with goes a lot farther in ensuring
> you
> > getn an answer then the whim of some group of programmers on an open
> source
> > project.
> >
> > Additionally, do you want to buy just what you need when you have to
> > consider that requirements MAY change in the future? Personally I think
> it's
> > better to buy something you can grow into then to buy something you grow
> out
> > of in no time.
> >
> > Now, I don't beleive you have to buy Websphere or Weblogic if you're
> writing
> > a site that has one dynamic page that shows inventory on your limited 10
> > item database. But if I'm building a corporate wide application with lots
> of
> > data persistance, then trying to get ONLY exactly what I need... ouch, I'd
> > be hangin myself if the requirements changed. (And they always do).
> >
> > Lastly, as for the 'why do people buy something they don't need'... I
> always
> > think 'why do we buy Off road trucks and SUV's simply to ride to and from
> > work in the city'? Heck, most of the population could get by riding Honda
> > Civics, Austin Mini's and Toyota Tercels.. but we go for the beasts.
> >
> > Maybe this could be some form of IT maschismo.. "Oh, you guys only use
> Resin
> > at your place? Well <BOLD>I'M</BOLD> using Websphere 19.0 with distributed
> > clustering on 14 AS/400's" (grunt, grunt grunt)
> >
> > I don't think so Tim!!
> >
> >
> ===========================================================================
> > To unsubscribe, send email to [EMAIL PROTECTED] and include in the
> body
> > of the message "signoff EJB-INTEREST".  For general help, send email to
> > [EMAIL PROTECTED] and include in the body of the message "help".
>

===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff EJB-INTEREST".  For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".

Reply via email to