On Tue, May 23, 2006 at 03:14:17PM -0400, Charles Fry wrote: > > > You are proposing two distinct alternatives: > > > > > > - number of non-library packages depending on the VM > > > - if you install a certain VM, how many applications will you be able > > > to run with it > > > > > > The first is easier to measure, while the second would potentially be > > > more useful. > > Errh, agree, but the idea was that it's somewhat correlated (or am I > > missing something!?), and if: > > 1. you trust the non-library package maintainers :-) > > 2. modify the 2nd alternative with "being *sure* to be able to run" > > then it's pretty much the same. > > > > In other words: if the package has a dependency on a specific VM, that > > should mean that the maintainer has tested the application with the VM. > > Cogito ergo sum. > > You are right that popcon would give us a subset of the VMs on which a > package works, but I would expect the actual number to be far larger. > For example, most packages are probably guilt with gcj or kaffe, but > probably work with many other JVMs as well. > > That said, I don't have a sense of how common it is for a package not to > work with a certain JVM.
When a package doesnt work with a specific jvm (or jvm group) it should just be filed as a bug. Either for the package or the jvm. Whereever the bug belongs. On the long run me should try to make it possible to run every application with every runtime. I know its a high goal. Cheers, Michael -- Escape the Java Trap with GNU Classpath! http://www.gnu.org/philosophy/java-trap.html Join the community at http://planet.classpath.org/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
