Goal: Increased Modularity?

Les Mikesell lesmikesell at gmail.com
Fri Sep 7 04:05:42 UTC 2007

Richi Plana wrote:

>> Do you disagree with the idea that a JVM executing java bytecode should 
>> be very much like a media player playing a media file?  That is, 
>> installing the data should not be dependent on having some particular 
>> player or JVM available that might or might not be used to manipulate it 
>> later.  Why can't I have the app before the JVM or supply my own JVM?
> Probably not the best example. A piece of data (a media file in your
> example) has so many different kinds of applications associated with it,
> some might even be on different systems and the data is exported via a
> number of means (http, nfs, etc.) Java bytecode, unless they're applets,
> are much easier to predict requirements for.

Is there some problem running an application where the bytecode is on a 
different machine than the jvm, using nfs?

> There is a huge probability
> that they will need a JVM, and unlike media players, from outside,
> they're designed to function in the same way.

I'm not sure I understand.  Shouldn't you be able to use any version of 
a jvm, packaged or not, without specific dependencies, just like you 
could use different media players to play the same file? Or even run 
more than one of them at once?

> As examples go, it's closer to say that JVMs are to Java applications or
> even just bytecode what glibc is to native applications.

Things may be that close-coupled in practice but only because of lack of 
conformance to the specification.  It should be more like claiming that 
C source code is dependent on one gcc version or some other specific 
compiler.  In some cases that may be true, but it doesn't seem right to 
encourage such bugginess with permanent workarounds.

> In Fedora, openoffice.org-core-2.2.0 Requires the virtual package java.
> You CAN have the app before the JVM, I suppose, the same way you can
> "rpm --nodeps -ivh openoffice.org-core.2.2.0....", but they'd just be
> bytes taking up space in your filesystem.

Will it work later then if I provide a different JVM than the packaged 

>> No one has given a usable answer for even a single value of JAVA_HOME 
>> for a single packaged jvm version.  Or the location of the document that 
>> provides this necessary information.  I'm sorry if that was too much to 
>> ask.
> Honestly, I tried to answer here
> (https://www.redhat.com/archives/fedora-devel-list/2007-September/msg00396.html), here (https://www.redhat.com/archives/fedora-devel-list/2007-September/msg00414.html) and here (https://www.redhat.com/archives/fedora-devel-list/2007-September/msg00490.html), but I guess I'm misunderstanding the question.

Yes, thanks, and sorry - I missed the one actual answer in the first 
link, but I was confused by the comment somewhere about parts going to 
private and export directories that didn't seem to be under the same top 
level. Is everything expected to be under JAVA_HOME actually still in 
one place?

   Les Mikesell
     lesmikesell at gmail.com

More information about the fedora-devel-list mailing list