The Darkstar Preservation Society

[Tim B]

Thanks for setting this up Jeff, I was planning to do something
similar tomorrow but you beat me to the punch.

From the sound of it we have the source code already preserved. It
would be worth seeing if we can get a snapshot of the wiki, forums and
any other documentation as well. Is there anything else we need?

From my point of view the best place for a new subversion would be
google code as that's where our open source projects are hosted. I'm
not familliar with github - what do they provide?

[Jeffrey Kesselman]

They provide git hosting and the usual other features.

The problem with googlecode is they are VERY restrictive on what licenses they will allow you to use.

To me, this violates a fundamental point of open source, which is code freedom.

--
~~ Microsoft help desk says: reply hazy, ask again later. ~~

[Tim B]

Well PDS is GPLv2 already so we are pretty constrained on that unless
Sun want to open things up a bit (I'll not hold my breathe).
Personally I'm happy with SVN so I'd rather not need to get yet
another version control system - but it's not a big deal to install
git if needed.

I know what you mean about google restricting licenses but TBH I agree
with them that the proliferation of open source licenses is not a good
thing. That's extremely off topic though! It does remind me though
that licensing in general is a potential problem now since we may lose
the option Sun were providing to get a commercial license for those
people who need/want it.

TBH as I see it at the moment we have the following things that need
doing:

1. Preserve as much data as possible from the existing PDS sites
against whenever they shutdown.

2. Evaluate the current known bugs and the state of development on
multi-node.

3. Establish just what development work still needs doing and who is
available to do it.

Anyone got anything to add to that?

[Andres Martinez Quijano]

Hey, nothing to add, just welcoming myself to the society

Just one thing: you are happy with SVN because you don't know git!
Trust me! I've been there before :)

[Rich Elswick]

Well the docs are in a wiki and it is worth the ask of the PDS folks
at Sun to do an export of them.

Barring that, copy/paste works.

Rich

--
Sent from my mobile device

Rich Elswick

[Mark]

Hello, I've been lurking on the previous boards and using DarkStar for
a little while now (with a Nintendo DS client)
I just wanted to say that if this project is going to continue on from
here then I want to help support it (with usage, advocacy, and
hopefully code)

Also kudos to the Red Dwarf fork name....very clever

[John Dunlap]

I would like to second that. I too have been lurking for quite some
time. I do not have significant amounts of free time but I do not want
to see this project die. I am a Java developer by profession and I'd
like to help in any way I can.

Cheers!
-John

[nokama]

Hi All, I was a member on the forums, and just want to say how gutted
I am that Sun lost their bottle with this one, it would have been
great to see it complete under the team that have put in so much work.

Anyways, I'm studying a BSc in Computer Science, and was going to use
Darkstar and Slipstream as a basis for my dissertation project next
year, and to be quite honest this news has gotten me a little upset,
but finding this group made me think that my dissertation project was
still possible.

I have access to a very good (and very experienced, he must be in his
late 60's) programming lecturer. I've explained Darkstar to him, and
he seems interested enough to guide me in my exploits, but the
question now is what do I base my study on?

I'm going to have time dedicated to this as part of my study time, and
I wonder if I could take on a little portion of the research we need
to do to get multinode?

I'm offering any help I can, and it'd be a great dissertation piece,
considering the complexity of the issue, and the ground-breaking
nature of Darkstar.

Do we have the conclusions Sun came to at the point of rejection of
the project? Jeff do you think there's a possibility Sun would pass
over all research documentation to this project as well, or are we on
our own in that department?

What I'm a little concerned about is that Sun have tested all the
multi-node work, it hasn't worked out, and the project has been
dropped as a result. I'd really like to know if that is the situation,
because there's no use flogging a dead horse, but there could be good
elements we can re-use.

[Tim B]

If that was the case then I'm fairly sure it would have been announced
as such. Remember that in research (and PDS was a Sun Labs Research
project) a failure is still valid research results.

This is based on absolutely no inside knowledge and I doubt anyone who
has access to such knowledge can comment in a public forum like this
but I'm pretty sure that with the Oracle takeover someone has gone
through every project in Sun and asked awkward questions like "where
is the revenue stream, how is this going to make us money". The answer
to that is always pretty vague when it comes to both an open source
project (unless you have a reasonably sized user base of people
actively paying for support contracts) and a research project. That
means that an open source research project is doubly at risk and this
eventuality has been a real possibility from the moment the takeover
was announced.

I was really hoping they would get multi node finished before the axe
fell though as that is the thing that needs a focussed and dedicated
effort. Ongoing maintanence and bugfix is a lot easier to keep
rolling.

[Jeffrey Kesselman]

I can tell you with confidence that that is not the case.  As others have said, whether you prove something does work or prove it doesn't is all the same in research as long as you publish your results.

This was quite simply Larry Ellison and his team at Oracle deciding they weren't interested in spending any more money on this.

 I think hes being terribly short sighted and could well live to regret that but, of course, I *would* think that ;)

On Feb 3, 8:09 pm, nokama <willbith...@hotmail.com> wrote:

> What I'm a little concerned about is that Sun have tested all the
> multi-node work, it hasn't worked out, and the project has been
> dropped as a result. I'd really like to know if that is the situation,
> because there's no use flogging a dead horse, but there could be good
> elements we can re-use.

[nokama]

Cool, well I'm actually glad at this point they didn't come to a
conclusion then :)

Have we got all the affinity grouping code in our branch? That seemed
to be a main key in it all, and it's something I'd really like to
study. Last I heard it was getting reviewed, but was in a completed
state, but I have had other things going on so haven't paid much
attention.

[Jeffrey Kesselman]

I haven't had a chance to a do a review but we have main trunk plus ALL the branches.  (Took quite awhile to transfer.)

I'm really hoping that some of the team will at least participate on an advising basis.  The rumor I hear from the ones I know personally is that they were VERY close to final integration of all the pieces.  But right now, thats just rumor.

JK

[Matt S]

Just introducing myself. I'm a professional Senior Java SDET for a
Java-backed, browser based, casual MMO developed by a relatively large
gaming company, and though I don't have tons of free time, I'd like to
throw my hat in the proverbial ring for what little I can contribute.
Any chance we can set up some development rooms and (if possible)
capture any previous development discussions?

-M.

[Jeffrey Kesselman]

I think there is definitely that option... and I'm really happy to see the interest in staying together and doing what needs to be done with  PDS.

I understand from folks i know on the (ex) labs team that some of them are still very interested in trying to save the project themselves.

I set this up as a safegaurd in case they fail, but as long as they are trying to make it coem together still I don't want to get too far ahead in terms of striking out on our own.

That seem fair and reasonable to everyone?

[nokama]

That only seems fair Jeff, and their knowledge of the project would be
very benificial.

This project does need to be saved though, in a complete form it would
surpass by far anything that's available to the public for a
reasonable price, if not ultimately.

I'm sure you'll keep everyone informed as to what's going on when you
hear it.

On Feb 3, 10:32 pm, Jeffrey Kesselman <jef...@gmail.com> wrote:
> I think there is definitely that option... and I'm really happy to see
> the interest in staying together and doing what needs to be done with  PDS.
>
> I understand from folks i know on the (ex) labs team that some of them are
> still very interested in trying to save the project themselves.
>
> I set this up as a safegaurd in case they fail, but as long as they are
> trying to make it coem together still I don't want to get too far ahead in
> terms of striking out on our own.
>
> That seem fair and reasonable to everyone?
>

[Zarch AlDain]

Jeffrey Kesselman wrote:
> I think there is definitely that option... and I'm really happy to see
> the interest in staying together and doing what needs to be done with
> PDS.
>
> I understand from folks i know on the (ex) labs team that some of them
> are still very interested in trying to save the project themselves.
>
> I set this up as a safegaurd in case they fail, but as long as they
> are trying to make it coem together still I don't want to get too far
> ahead in terms of striking out on our own.
>
> That seem fair and reasonable to everyone?

Absolutely. The urgent need was to make sure that we had all the source
code and a communications link ready in case the old site disappeared
and that is now done. I'd be very glad to hear that some or all of the
PDS team are still working on the project though and if they are able to
do so then I agree we should try and support them rather than fracturing
the community.

Tim

[nokama]

Got em all, and I'll do a backup. Cheers Jeff, well saved.

[David Jurgens]

Also to introduce myself, I was a summer intern with PDS for three
years and am reasonably familiar with the internals of the code base.
I'm pretty disappointed that it was axed, but I'm willing to help in
whatever way I can to keep the project moving forward.

-David

[Jeffrey Kesselman]

Hi David!

Great to have you around!

JK

[Josh F.]

Count me in as well. I'm an independent developer that has a
production single player game for which we are working on a multi-
player variant using PDS (don't need multi-node... at least not yet),
and if that works out in testing we are doing now (and it certainly
seems it will), then definitely count me in for supporting this
community in whatever way I can with a fairly diverse set of Java
skills and limited wannabe network engineer skills (yeah, probably not
helpful but I'm up for trying). Only caveat is that this would be my
first foray into actual contribution on anything resembling FOSS - so
lots to learn about how these things are run.

PDS was/is the winner in our selection process for a server
technology, and I am willing to help keep it that way.

Thanks for setting this up and keeping it going!

-- Josh

[Federico "Frodo" Bergstein]

Just posting to introduce myself.

I would like to help too, I would be out of my depth (I think) if I
tried to help with the development of new/unfinished features, however
I will be glad to help with bug fixing, simple&boring code monkey work
or even (gulp) documenting...

Saludos.

Frodo.

[Cervator]

Another one joining the crowd here - I'm a professional developer with
a strong interest in gaming and have had PDS in my peripheral vision
for years, and always meant to take more of a look. Caught this news
coincidentally off Google News and came to take a look.

I'm sad to see the project's sponsorship fall aside, but encouraged to
see so many speak up for it and join the group this fast. I cannot
contribute anything personally yet, aside from encouragement, and
you've got that. Remember the old axiom - any publicity is good
publicity - and try to use this event to recruit many more developers
for the project. A speedy re-launch will help take advantage of the
attention before it dies down. The faster you figure out the exact
fate of the official site, forums, code, etc, the better.

[Zarch AlDain]

Good news - some members of the Sun PDS team are setting up a Community
Fork on SourceForge:

http://www.projectdarkstar.com/forum/?topic=1550.0;topicseen

It seems with luck that this list will become obsolete as fast as it was
formed - but still it served it's purpose as an emergency fall-back!

Tim

[Jeffrey Kesselman]

Yup.  Its pushing ahead.

Oh the google code comment.  We might've theoretically been able to use google code (although I believe PDS is GPl + Classpath Exception, which Im not sure

is a google choice.)

i just personally despise the way they are trying to dictate license terms to the world at large.

For that reason I personally have sworn off them... and I set up the git repo. ergo...

But sine there will be an official one, it will likely not be necessary.