You are viewing a plain text version of this content. The canonical link for it is here.
Posted to fop-dev@xmlgraphics.apache.org by Glen Mazza <gr...@yahoo.com> on 2004/03/02 23:55:40 UTC

Cocoon appears to be switching to 1.4

They're currently voting on the Cocoon side[1] to set
1.4 as the minimum JDK for their next 2.2 release.  So
far it looks good for approval.

This would be good news for us, as it would remove one
of the major risks (incompatibility with Cocoon)
involved with adopting 1.4 on our side.

Glen

[1] http://marc.theaimsgroup.com/?l=xml-cocoon-dev&w=2&r=1&s=entry+level+jsdk+1.4&q=b

Re: Cocoon appears to be switching to 1.4

Posted by Clay Leeds <cl...@medata.com>.
On Mar 3, 2004, at 1:39 AM, Bertrand Delacretaz wrote:
> Le Mercredi, 3 mars 2004, à 10:27 Europe/Zurich, Chris Bowditch a 
> écrit :
>> Glen Mazza wrote:
>>> They're currently voting on the Cocoon side[1] to set
>>> 1.4 as the minimum JDK for their next 2.2 release.  So
>>> far it looks good for approval.
>>
>> I'm not so sure it does, look at the 3rd mail in the thread:
>>
>> http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=107813002510299&w=2
>>
>> and this sparks off a debate about how many users still require 1.3, 
>> etc....
>
> You're right, this is only being discussed at the moment.
>
> And if the switch to 1.4 happens, it will only be (AFAIK) with the 
> Cocoon 2.2 release, for which there is not set date at the moment.

I assume this would be akin to 0.20.5 (and any updates--however minor 
they may be) continuing to operate under the current minimum system 
requirements, but 1.0+ requiring 1.4.

> Also, there is a push to keep the requirement at 1.3 for the Cocoon 
> core, leaving blocks free to require either 1.3 or 1.4.

That sounds like it might be a nice idea for a solution, although I 
can't imagine what portions would continue to require 1.3+ (1.2+?), and 
what would require 1.4+.

> So it might be better not to let this influence the FOP decision.
>
> -Bertrand

[brief anecdote]
The owner of my company would rather spend extra money/time developing 
a product so they make the lives of users easier, rather than make the 
lives of programmers easier. In other words, a switch to requiring 1.4 
should not be done merely to make the lives of fop-dev'ers easier, but 
instead would be done to make the lives of our fop-users easier.
[/brief anecdote]

I guess I still don't know what the benefits of switching to JDK 1.4+ 
are to _users_ (see below).

BTW, if we don't raise the bar to 1.4 (min Sys Req remains 1.3 or 1.2) 
might a user running under 1.4 still get some of the benefits of 1.4? 
Or, alternatively, would it be worth the effort to maintain 1.2/1.3 
compatibility & also include support for 1.4 features.

For my benefit (and others?) here's a brief list of "features" I think 
we can expect from a jump to Java SDK 1.4:

- performance enhancements (similar to the HashTable => HashMap change 
allowed by jump from 1.1 to 1.2+)[1]
- nestable exceptions which will make problem tracking much easier[2]
- the BIDI support[2]
- other AWT fixes and extensions[2]
- JCE by default (people will still have to get a RC4 provider though 
:-/)[2]
- And 1.4 has java.util.logging and java.util.prefs[2]

What other 'benefits' should we expect from making the jump to 1.4?

Web Maestro Clay

[1]
http://marc.theaimsgroup.com/?l=fop-dev&m=100106134012939&w=2

[2]
http://marc.theaimsgroup.com/?l=fop-dev&m=107353819024636&w=2


Re: Cocoon appears to be switching to 1.4

Posted by Bertrand Delacretaz <bd...@apache.org>.
Le Mercredi, 3 mars 2004, à 10:27 Europe/Zurich, Chris Bowditch a écrit 
:

> Glen Mazza wrote:
>
>> They're currently voting on the Cocoon side[1] to set
>> 1.4 as the minimum JDK for their next 2.2 release.  So
>> far it looks good for approval.
>
> I'm not so sure it does, look at the 3rd mail in the thread:
>
> http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=107813002510299&w=2
>
> and this sparks off a debate about how many users still require 1.3, 
> etc....

You're right, this is only being discussed at the moment.

And if the switch to 1.4 happens, it will only be (AFAIK) with the 
Cocoon 2.2 release, for which there is not set date at the moment.

Also, there is a push to keep the requirement at 1.3 for the Cocoon 
core, leaving blocks free to require either 1.3 or 1.4.

So it might be better not to let this influence the FOP decision.

-Bertrand


Re: Cocoon appears to be switching to 1.4

Posted by Chris Bowditch <bo...@hotmail.com>.
Glen Mazza wrote:

> They're currently voting on the Cocoon side[1] to set
> 1.4 as the minimum JDK for their next 2.2 release.  So
> far it looks good for approval.

I'm not so sure it does, look at the 3rd mail in the thread:

http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=107813002510299&w=2

and this sparks off a debate about how many users still require 1.3, etc.

> 
> This would be good news for us, as it would remove one
> of the major risks (incompatibility with Cocoon)
> involved with adopting 1.4 on our side.

I know Ive moaned about this in the past, but I'm concerned that 
upgrading FOP to JDK 1.4 will make FOP unavailable to x% of the 
potential users. Now if x% is a significant figure then FOP becomes in 
danger of failing in a manner similar to xmlroff.

I hope Tony Graham and any one else out there whos using xmlroff doesnt 
take offence at me saying this, but Ive been following the xmlroff 
project, and there seems to about 1 mail/month on their user list. I 
believe this is because xmlroff is unavailable to a significant % of the 
potential user base, i.e. it doesnt run on windows. I can tell you now 
if it did run on windows our company would be using it, and possibly 
contributing towards it.

Anyway, my point is simply that if x% is significant then we should hold 
off on this, and I will vote -1. However, if x% is very small then i'll 
vote +0. So before making a decision I propose we do some research to 
find out what x% is?! Which is the conclusion that cocoon-dev appear to 
come to, in the discussion thread above.

Chris