cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From ru...@us.ibm.com
Subject Re: JDK 1.2 vs 1.1: my final proposal
Date Wed, 29 Mar 2000 04:40:44 GMT


Stefano Mazzocchi wrote:

>So, while there is an urgent need for evolution, solid reasons not to
>throw away 1.1 compatibility were posed and since, after careful
>evaluatation, I found no specific need to have 1.1 incompatible code in
>cocoon's core, I made my proposal.

+1.  Well said.

Clearly, there will be a day when there is some compelling function
available in a JDK other than 1.1, and that should be the time you switch.
The reason should be more substantial than the developers prefer the new
collection packages (bad example, given that you can graft it onto 1.1, but
you get my point).  It looked like weak references could have been that,
but from what has been reported here, it doesn't look like that will be the
case, at least not with 1.2.2.

There have been comments on this mailing list about IBM's motivation.
Again, I'm not part of the groups developing the JVMs, but I will say that
IBM has some other products that don't support JDK 1.2 yet.  Common answers
when we ask our customers about adding support for JDK 1.2 include:

- adding support for JDK 1.2 is a good idea - as long as you don't remove
support for JDK 1.1.

- perhaps more interesting: many people have no intention of going to JDK
1.2.  The reason?  JDK 1.3 is in beta, and they don't want to pay the
expense of switching twice.

All this said, I  believe that the overwhelming majority of them would
switch in an instant if there was a demonstrable, concrete benefit to them.
If Cocoon2 ran twice as fast on half the hardware with JDK 1.2, I dare say
that no one would complain about having to upgrade their JVM.  And again,
something like weak references has the potential to give those kind of
benefits.

- Sam Ruby



Mime
View raw message