Re: River and Backward Compatibility

Brian Murphy <[email protected]> Thu, 6 Sep 2007 09:25:42 -0400
Newsgroups gmane.comp.java.sun.javaspaces
Message-ID <27cff3750709060625w54024p893575d63b8d7f2b__14445.2704037123$1189085167$gmane$org@mail.gmail.com>
On 9/5/07, Mark Brouwer <[email protected]> wrote:

> allows me to ask you what you
> exactly mean with your "+1 for backward compatibility" as the original
> question was "How much backward compatibility does everybody expect in
> this first release?".

I think Mike Morris probably expressed my hope for
the first release with respect to backward compatibility.

>On 2007-9-5 00:42, Mike Morris wrote:
>> change the absolute _minimum_ that /must/ be changed
>> (package names, what else?) to get the release out.  Nothing else.  As
>> close as possible to 100% backwards compatibility.

> Should I read it as 100% compatible even with respect to
> the directory layout of the distribution

Yes, that would be a fair read.

Related to this, for as long as the jsk-* jar files have been
available, I've been a little surprised by how many people
have stated on this and similar threads that they still develop
and/or deploy using jini-core.jar, jini-ext.jar, and sun-util.jar files,
rather than converting to using jsk-platform.jar, jsk-lib.jar, and
jsk-dl.jar. I certainly understand that most people still don't read
the documentation -- in particular, the release notes -- but I
would have thought that by now most would have made the
change.

So this is the sort of thing that makes me think it may be wise
to minimize, as much as possible, differences in that first release
with the current jsk release. In a perfect world, the only difference
for me as a user would be that I now obtain a zip file from the
Apache River site rather than the Jini page on java.net. Of
course, I realize this is not a perfect world. I realize that the
apache community as a whole has its own agenda that
emphasizes possibly different goals and/or priorities than
the pre-river Jini/JavaSpaces communities emphasized.
So I understand that there will probably have to be trade offs
made in order to balance the requirements imposed by the
"Apache Way(TM)", and the current and future needs of the
established user base.

With apologies to the apache faithful, I would hope that the
first river release would tip the scale a little more toward the
needs of the established user base; so as to make the transition
to river for that class of user as straightforward as possible.
Although people seem to think there aren't that many users
of Jini and JavaSpaces out here (I suspect because they don't
see it in the press or blogs), my experience has shown me
that that's not exactly true. Thus, if the first river distribution is
not as "drop-in" as possible, then I fear that river will lose a
good base of users on which to build its community because
those users will decide that's it's simply easier to keep using
what is already working well for us.

Anyway, I guess what I'm trying to say above is that I agree
with what Dennis Reedy said in his reply to this thread; modulo
any deal-breaker requirements imposed by the Apache process.

Again, just my 2 cents,
Brian

===========================================================================
To unsubscribe, send email to [email protected] and include in the body
of the message "signoff JAVASPACES-USERS".  For general help, send email to
[email protected] and include in the body of the message "help".

To view past JAVASPACES-USERS postings, please see:
http://archives.java.sun.com/archives/javaspaces-users.html