+1

cheers, manuel.

On 14 Nov 2009, at 05:44, Tim Ruppert wrote:

+1 - I totally agree. People can always run the release branch if they need to stay on 1.5, but the show must go on!

Cheers,
Ruppert

On Nov 13, 2009, at 9:32 PM, David E Jones wrote:


We had to make this decision before with 1.4 -> 1.5, and if I remember right from 1.3 -> 1.4 too.

Sooner or later we need to deprecate support for 1.5 in order to move on, and to more easily be able to update various libraries and such. The 1.5 -> 1.6 transition is a bit of a pain because they did a few changes that were not backwards compatible. Because of that it would be particularly nice to deprecate support for 1.5 and require 1.6 (or later, as possible/available).

In any case, our criteria before was to deprecate support for the older version of Java as soon as the newer version was widely available. At this point I think 1.6 qualifies well for that... unless there are specific systems people use that still can't run 1.6 then we should just update that requirement.

-David


On Nov 13, 2009, at 1:02 PM, Scott Gray wrote:

What is not clear to me is what are we missing out on by supporting both as we currently do, I think DBCP is an odd case because they obviously have tried to support 1.5 but it seems there are a couple of very minor bugs that prevent that support.

Other than DBCP what would we actually change (and why) that would prevent 1.5 support?

Thanks
Scott

HotWax Media
http://www.hotwaxmedia.com

On 14/11/2009, at 8:46 AM, Jacques Le Roux wrote:

Sun :
Java SE & Java SE for Business Support Road Map (JDKs End of Life)
http://java.sun.com/products/archive/eol.policy.html

1.6 supported systems
http://java.sun.com/javase/6/webnotes/install/system-configurations.html

15 supported systems
http://java.sun.com/j2se/1.5.0/system-configurations.html

Conclusion (I don't list what is better, only what miss from 1.5 to 1.6)
Better for Solaris (no surprises)

Windows x86
no longer : Windows 98 (2nd Edition) and  Windows ME (who cares ?)
no longer : Windows Server 2003, Web Edition (R2, SP1, SP2) and Windows Server 2003, Standard Edition (R2, SP1, SP2) ( a bit more
annoying maybe)
but still : Windows Server 2003, Enterprise Edition (R2, SP1, SP2) and Windows Server 2003, DataCenter Edition (R2, SP1, SP2)

Windows x64 32-bit mode
no longer : Windows Server 2008 Datacenter (SP1, SP2) and Windows Web Server 2008 (SP1, SP2) but still : Windows Server 2008 Standard (SP1, SP2) and Windows Server 2008 Enterprise (SP1, SP2)

Linux x86
no longer : TurboLinux 8.0
but : Turbo Linux 10 (ONLY Chinese and Japanese Locales. No English.)

Linux x64 32-bit mode
new : Turbo Linux 10 (ONLY Chinese and Japanese Locales. No English.)

From this I can see only problems with
Windows Server 2003, Web Edition and Standard Edition
Windows Server 2008 Datacenter and Web Server
TurboLinux 8.0

Should this only stop us to migrate to jdk 1.6? Personnaly I don't think so, now it's the community to decide...

Jacques
PS :
Postgres does not really bring much with JDBC4 drivers (could be : http://www.artima.com/lejava/articles/jdbc_four.html) <<JDK 1.6 - JDBC4. Support for JDBC4 methods is limited. The driver builds, but the majority of new methods are stubbed out. >>
http://jdbc.postgresql.org/download.html
I come to it from this discussion 
http://archives.free.net.ph/thread/20090925.160633.e1923a69.fr.html#20090925.160633.e1923a69

A benchmark while at it 
http://benjchristensen.com/2009/02/12/java-jdk-15-vs-16-performance/


From: "Jacques Le Roux" <jacques.le.r...@les7arts.com>
Thanks Adrian,

So it's the same concern than 3 years ago, I will check...

Jacques

From: "Adrian Crum" <adri...@hlmksw.com>
Running OFBiz on 1.6 is a good idea, because (Sun claims) 1.6 performs
better than 1.5.

The downside is, there might be servers out there that don't have a 1.6 JRE available. I would like to see us move on to 1.6 (especially now
that 1.7 is near), but I'm concerned for existing installations.

-Adrian

Jacques Le Roux wrote:
Hi devs,

I don't remember why we still use 1.5. Here is a thread from 2006
http://markmail.org/message/5j5es63hpsh543ct.
Maybe it's time to think anew about it, don't you think so ?

Jacques










Attachment: smime.p7s
Description: S/MIME cryptographic signature

Reply via email to