felix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Richard S. Hall" <he...@ungoverned.org>
Subject Re: [proposal] Oscar OSGi Project
Date Mon, 25 Jul 2005 19:05:26 GMT
Niclas Hedhman wrote:

>Many people want to see Oscar thrive and succeed. The fact that Richard 
>recognizes the fragility of his own position, should give him all possible 

Yes, this is very much the case. I would like people to have my back on 
implementing the core. I love to do it, but I cannot be solely 
responsible all the time. I am an academic researcher, I actually need 
to do research occasionally. :-)

>Richard is here to learn the Apache Way and make himself redundant in the 
>longer-term :o)  which is good for everyone involved. Several people at ASF 
>is interested in becoming more involved in OSGi core development, learn the 
>internal tricks and be part of the OSGi specification process. We can all do 
>that by "leaving" ASF and join Richard, or he (and others) can join this 
>large and joyful gang on well-proven grounds.

That is essentially my view too. If starting an Apache OSGi project 
brings more people into the OSGi fold, then it seems like a good thing 
to me.

>Also important note to remember in respect to contributions;
>Richard mentioned that we are talking Oscar Core, and not the various bundles 
>that has been contributed via the Oscar Bundle Repository. Also, we are 
>talking about Oscar 2.0 which "doesn't really exist", at least not in terms 
>of OSGi specification, since the R4 spec is not out yet. So, theoretically we 
>can't work on the codebase, since there is no specification to implement, or 
>be more precise, only Richard can work on it and not show it publicly. IMHO, 
>this is also an issue that we need to address to some extent. What can and 
>can not be done?

Actually, that is partially correct. It is definitely true that non-OSGi 
members cannot read the R4 spec at this point, since it is not released. 
Thus, it would be difficult for anyone except members to implement R4 
features (but this is probably true of many spec processes). Oscar 2.0 
source, however, is publicly available; I just don't claim that it is 
compatible with anything, it is just part of my experimentation with how 
to extend R3.

>For the sake of perfection; Let's seek explicit grants from the individuals in 
>question. After all, if you are willing to hand out a chunk of code over 
>mail, the Apache ICLA shouldn't be much of a problem.

It certainly wouldn't be an issue if necessary, since there are probably 
only about three people that would qualify.

>We will also need to see what comes out of the OSGi Alliance in respect to the 
>R4 specification and licensing terms for independent implementations in 
>general and ASF/ALv2 in particular.
>Further down the road I envision ASF, possibly together with Eclipse 
>Foundation, stepping up to the OSGi Alliance, and "help them" become more 
>OpenSource friendly, as these two implementations will probably take a lion's 
>share of "the market". But this is way down the road.

I think this is happening slowly already.

-> richard

View raw message