We have a need to access QM from Java. We can set up a project to do
this in-house but it seems the sort of thing that the open source
community might like to do (if they haven't already done it).
Exactly. Go to the top of the class!
This is why we are not abusing your GPL release of OpenQM by using it
in a commercial environment.
What's source for the goose, is source for the gander!
Ashley
> Exactly. Go to the top of the class!
>
> This is why we are not abusing your GPL release of OpenQM by
> using it in a commercial environment.
>
> What's source for the goose, is source for the gander!
Which shows exactly why we have no belief that the open source community
have any intention of participating in collaborative development.
Haha, that's funny. Made me laugh out loud!
If you want a collaboration, it's usual to expect a equitable
relationship. Quid pro quo. If you want to get BSD licensed
contributions, then release your code under BSD.
It really is that simple.
Ashley
>
> Hi Ashley,
>
>> Exactly. Go to the top of the class!
>>
>> This is why we are not abusing your GPL release of OpenQM by
>> using it in a commercial environment.
>>
>> What's source for the goose, is source for the gander!
>
> Which shows exactly why we have no belief that the open source community
> have any intention of participating in collaborative development.
>
*pop*
What the fuck do you think the whole IPv6 update was about, you smarmy
little bitch? Glen & I worked really hard on that update and it was
offered to you under a dual license. You didn't have the balls to step up
to the plate and even acknowledge the contribution.
You constantly and consistantly treat people like they're all personally
after your little pot of gold and then have the audacity deny it and point
out how nobody contributes.
From the _instant_ you released a GPL source tree, you've done _nothing_
to foster community participation. You put the main download behind a
password wall that essentially threatens anyone that wants the source.
You consistantly treat anyone that's interested in the GPL version like
they're some kind of digital thug and then act all surprised when you get
called on it.
Frankly, I'm glad you don't participate. Until I started working on
op_skt.c, I was convinced that I was the worst C programmer on the planet.
If you worked for me and wrote code like that, I'd fire you with
trebuchet.
Oh, and what is this "we" shit? You're not royalty there sparky, despite
your delusions to the contrary.
Congrats Martin, you got me to blow a gasket. Now you can go around and
look smug to all your little buddies and crow about how right you were.
Asshat.
g.
--
Proud owner of F-15C 80-0007
http://www.f15sim.com - The only one of its kind.
ScarletDME - The red hot Data Management Environment
A Multi-Value database for the masses, not the classes.
http://www.scarletdme.org - Get it _today_!
> Would Lesser GPL be acceptable Martin?
The need for Java support has come from a third party software tool vendor
who wishes to interface with QM in a commercial product.
The requirements as I see them are:
* It must be available for all major platforms on which commercial QM runs.
* It must be distributable by the tool vendor to his clients. To make it
more generally useful rather than restricted to this one application, it
should be distributed with QM even if we don't own it or have rights to
modify it.
* It must be released in binary form so that the user does not need to build
it.
* There must be no separate licence charges.
* It must be fully supported.
Within these requirements, there is no reason why it cannot remain a third
party product but I can see that the vendor might require some assurance
that it will not go away.
If the above (which I admit asks a lot) is not acceptable, we will have to
do this in-house, duplicating your work.
>
> Whatever Martin did in the past, he is now tentatively discussing putting up
> a subversion repository to aid contributions such as your IP6 work. If he
> does this then all your issues with him are essentially resolved and we can
> forgive and forget.
>
Nah, I'm done with him. Fool me once and all that. The dual license for
the IPv6 work is still in place, no hidden stick.
I wish you enormous amounts of luck Steve. Frankly I think my backside
will become a porcine aircraft carrier long before he'll actually step
up to the plate.
>
> Gene is right though, the IPv6 layer was designed for the new D3
> implementation.
> This is our proffered and intended future development.
>
Actually, the IPv6 feature was a side benefit of updating the 2.6-6 socket
code to comply with the 2.8 changes. Cedric had used a 2.8 flag in his
version of the JD3 server code. I never did get back to seeing if it
worked after re-writing op_skt.c :)
g.
--
Proud owner of F-15C 80-0007
http://www.f15sim.com - The only one of its kind.
ScarletDME - The red hot Data Management Environment
> Whatever Martin did in the past, he is now tentatively discussing
> putting up a subversion repository to aid contributions such as your
> IP6 work. If he does this then all your issues with him are essentially
> resolved and we can forgive and forget.
In the light of the highly abusive postings from Ashley and Gene, we are
seriously considering abandoning the open source collaborative develoment
completely. It is clear that the developer community does not share our
views on what this is about. We will need to discuss this further internally
and with some of our key business partners before taking action.
There will continue to be some form of open source offering but it will, as
before, be just the core multivalue engine and we may impose some more
restrictive licensing conditions to control how it is used. We will not
necessarily expect to get anything back but, equally, we will not post any
new developments except those that are contributed.
We are also considering a source option for licensed commercial users who
want to extend the product within their own environment but not for
distribution. There are some interesting issues to resolve before we can do
this but it has possibilities.
Incidentally, despite Gene's rantings, we are unaware of his IPv6
development being offered to us. There are no references to it within either
of the OpenQM Google groups and I can find no references within any of our
email boxes. Certainly we have not received a submission document. It is a
trivial development that has been low on our wish list for a long time and
we will probably do it ourselves in the not too distant future.
How can you abandon something you never participated in? Regardless, I'm
one person, not a community. Taking punative action on many because of
the words of one is....typical.
I'll make you a deal Martin, you go right along and "participate" as much
as you're willing. I'll shut up. You won't hear another ranting thing
from me. Hell, even if you take your ball and run home, I'll shut up.
It's much better for my blood pressure either way.
> There will continue to be some form of open source offering but it will, as
> before, be just the core multivalue engine and we may impose some more
> restrictive licensing conditions to control how it is used. We will not
> necessarily expect to get anything back but, equally, we will not post any
> new developments except those that are contributed.
>
Yeah, good luck with that sparky. You can't un-GPL a code base. You'd
have to *gasp* issue the sources for an updated version.
> We are also considering a source option for licensed commercial users who
> want to extend the product within their own environment but not for
> distribution. There are some interesting issues to resolve before we can do
> this but it has possibilities.
>
That would be good for the customer. Maybe you're not a total loss.
> Incidentally, despite Gene's rantings, we are unaware of his IPv6
> development being offered to us. There are no references to it within either
> of the OpenQM Google groups and I can find no references within any of our
> email boxes. Certainly we have not received a submission document. It is a
> trivial development that has been low on our wish list for a long time and
> we will probably do it ourselves in the not too distant future.
>
If I cared enough at this point, I'd dig up the thread where it was
discussed. I don't exactly recall which list.
I'm surprised you consider what was essentially a complete rewrite of
op_skt.c to be "trivial". Oh that's right. Outside efforts are
"trivial", inside efforts are "valuable". Got it. Thanks for that.
Really.
I don't recall doing that. If I did, then I do apologize unreservedly
for offending you.
What specifically are you referring to, so I can be sure I do not
repeat the offence?
Ashley Chapman
Hi Diccon,
The need for Java support has come from a third party software tool vendor
> Would Lesser GPL be acceptable Martin?
who wishes to interface with QM in a commercial product.
The requirements as I see them are:
* It must be available for all major platforms on which commercial QM runs.
* It must be distributable by the tool vendor to his clients. To make it
more generally useful rather than restricted to this one application, it
should be distributed with QM even if we don't own it or have rights to
modify it.
* It must be released in binary form so that the user does not need to build
it.
* There must be no separate licence charges.
* It must be fully supported.
Within these requirements, there is no reason why it cannot remain a third
party product but I can see that the vendor might require some assurance
that it will not go away.
If the above (which I admit asks a lot) is not acceptable, we will have to
do this in-house, duplicating your work.
> We have a discussion going on in Scarlet started by you about moving to
> requiring bsd licence for submissions but you seem to be suggesting that
> here that we do not. I am now officially confused.
As this is a Scarlet discussion, I'll answer your question on that forum
http://groups.google.co.uk/group/scarletdme/browse_thread/thread/6d56f828801f8770?hl=en
Ashley