It's worth noting that if a customer wants to deploy River on sparc, it
will be relatively easy for them to do so, they'll be able to take up
the baton, build and run the tests.
Maintaining a sparc development computer takes time and resources, you
need a license and support contract for Solaris, hardware is expensive
and 3rd party vendors non existent or drying up, you can't buy new sparc
workstations either. I've been waiting 3 years for Sun/Oracle to
realise that workstations are an essential, albeit small market, surely
they could have put an M3000 inside a tower case with 3D ATI graphics
support by now. In fact if they did that, they'd be able to sell Sparc
Visualisation servers too. All the other platforms you've mentioned are
readily available.
We don't test on IBM's Power or HP's Itanium architectures, these are
Sparc's competitors.
I think we can get a better bang for buck, testing on other JVM's, like
Apache Harmony, OpenJDK, JRockit and IBM J9 by eliminating our
dependence on proprietary implementation details of Sun's JVM, but we're
not there yet.
River's currently in a refactoring phase, our goal is to make
development more efficient so we can ensure long term survival. Because
we're all volunteer developers, we need to make it possible to have
smaller develop/test cycles. We're all very passionate, from many
backgrounds and countries. We don't always agree on everything and in
fact have more in common even when we do disagree, what matters is that
we respect each others views, if you feel passionate enough about
something, jump in and join in the fun.
Peter.
Peter Firmstone wrote:
Perhaps you might be interested in helping us fix some bugs or
checking the release documentation?
We're all just volunteers here, I've made attempts to identify the
source of the bug and lack the time needed to figure it out. Patricia
has offered to help. Feel free to jump in and get your hands dirty.
If someone does have time to nut this one out, send me your public ssh
key & I'll set you up a user account.
Cheers,
Peter.
Jason Pratt wrote:
i am stating that the first "graduated" release should work for everyone
period. after that if you want to release with known bugs and reduce/not
support sparc or whatever other platform you in your infinite wisdom
deem
irrelevant , great. river should have one release it can point at for
anyone
wanting to try/use it.
regarding supporting systems, if ubuntu, amd, redhat, intel, apple, etc.
don't send you equipment and software to test on, are you going to drop
support for them as well? doesn't apache have these things
internally? have
you asked?
you can be as antagonistic as you like, i deal with children everyday
;-)
On Fri, Apr 1, 2011 at 9:46 PM, Dan Creswell <[email protected]>
wrote:
Why should it be valid for everyone?
So far the conversation has mostly amounted to "save the SPARC user"
at any
cost and that cost includes "penalise all other users".
Is that latter cost something we think we should be asserting? And
for how
long?
Further, if those using SPARC can't lend us a box to test on, how
serious
are they about their investment in River? Simply and brutally,
they're not
supporting us (an opensource co-operative effort), why would we in turn
support them?
Last up, if we are to support these SPARC users and they won't
provide the
kit, we have to obtain it and maintain it etc - how are we doing
with that?
How well can we do that going forward?
My point overall then is that there's a balance and as yet, I don't
see a
proper discussion about that balance just what would be ideal in a
perfect
world.
Yeah, I am being a little antagonistic,
Dan.
On 1 April 2011 09:18, Jason Pratt <[email protected]> wrote:
sparc was a key architecture when jini was being promoted, maybe
not so
much
now. however, at least for a graduation release it should be valid for
everyone sparc included. afterwards if justified it can be phased out.
antagonism aside, for its first release let give the masses something
that
works...
jason
On Fri, Apr 1, 2011 at 1:51 PM, Patricia Shanahan <[email protected]>
wrote:
On 3/31/2011 5:41 PM, Niclas Hedhman wrote:
On Wed, Mar 30, 2011 at 5:48 AM, Jason Pratt<[email protected]>
wrote:
please don't release anything with failures. i've been a big
fan/user
of
jini since it was released. now that it is alive again via river, a
good
release history will be key to success/survival
Assume for a second that the failure is related to the SPARC JVM
implementation; For how long do you intend us to hold off a
release to
the other 99.9654% of the users, searching for a compatible way to
work around the problem?
Sometimes, "known bugs" are just that. The "unknown bugs" is in most
cases a longer list, and should that also hold off a release
until we
find and correct them?
I hope I don't come across as too antagonistic... ;-) ... just want
to provide some perspective.
Here's a key question for the future. Are there users that need River
on
SPARC? Do we go on supporting it? Does anyone care enough to make a
SPARC development environment available?
Patricia
