|Alexander Broekhuis||Jan 10, 2012 10:30 pm|
|Marcel Offermans||Jan 10, 2012 10:55 pm|
|Pepijn Noltes||Jan 11, 2012 12:04 am|
|Martim||Jan 11, 2012 4:32 am|
|Sascha Zelzer||Jan 11, 2012 4:41 am|
|Sascha Zelzer||Jan 11, 2012 5:37 am|
|Alexander Broekhuis||Jan 12, 2012 2:14 am|
|Pepijn Noltes||Jan 13, 2012 2:23 am|
|Marcel Offermans||Jan 13, 2012 10:13 am|
|Marcel Offermans||Jan 13, 2012 10:23 am|
|Sascha Zelzer||Jan 17, 2012 12:55 pm|
|Pepijn Noltes||Jan 18, 2012 12:52 am|
|Alexander Broekhuis||Jan 23, 2012 1:09 am|
|Subject:||Re: Poddling status|
|From:||Alexander Broekhuis (a.br...@gmail.com)|
|Date:||Jan 12, 2012 2:14:39 am|
Thanks for the replies. I'll try to summarize what has been posted by now:
- Thales is going to use Celix in a research project and is actively developing Celix and with Celix. Part of their development is an implementation of the Device Access Specification, which will be donated to Celix. I think this is great news! Pepijn: Are you willing to maintain the Device Access implementation if it is donated to Celix?
- Use Celix as an alternative for JNI, which provides a more robust solution. The processes are separated, and one side crashing won't take down the other. Does anyone have a specific use case or interest in this that can be used as a showcase?
- During many discussions C++ is mentioned, also seing the replies now again there seems to be quite a lot of interest in an OSGi implementation in C++. - There are several C++ OSGi like implementations, collaboration with these projects could benefit both.
Seeing this interest in C++, I think it would be a good starting point to try and reach a broader community.
The following C++ frameworks are mentioned: - nOSGi: http://www-vs.informatik.uni-ulm.de/proj/nosgi/ - SOF: http://sof.tiddlyspot.com/ - CommonTK Plugin Framework: http://www.commontk.org/index.php/Documentation/Plugin_Framework
Areas where I think collaboration might be interesting are: * Bundling * Metadata * API (how to map the OSGi specification to C/C++)
Does anyone have any ideas/suggestions regarding this? What would be a good starting point?
Also I think it is interesting how the current Celix framework can be extended so that it can support C++. If possible I would like to keep a C only framework, with specific extensions if used with C++.
Again, any ideas are welcome! My knowledge about C++ isn't that great, so any help would be appreciated.
If I misunderstood or forgot something, feel free to correct me.
2012/1/11 Sascha Zelzer <s.ze...@dkfz-heidelberg.de>
It also has a very nice paper explaining their approach. Maybe we can get the devopers of all these frameworks together to share their requirements, ideas, visions, etc.
Just my two cents.
On 01/11/2012 01:41 PM, Sascha Zelzer wrote:
I am following the Celix efforts with interest, but I am also more interested in C++.
In my opinion, Celix could try to reach out to other projects and their (probably small) community implementing a OSGi - like environment. If efforts could be concentrate, or some kind of interoperability be achieved, this would be awesome. Projects in my mind are Poco (commercial), SOF, and CTK.
I tried to start some discussions about that a while ago, but unfortunately did not get any replies:
http://incubator.markmail.org/**search/+list:org.apache.** incubator.celix-dev#query:**list%3Aorg.apache.incubator.** celix-dev%20from%3A%22Sascha%**20Zelzer%22+page:1+mid:** yrsceyy3ovisbhkh+state:results<http://incubator.markmail.org/search/+list:org.apache.incubator.celix-dev#query:list%3Aorg.apache.incubator.celix-dev%20from%3A%22Sascha%20Zelzer%22+page:1+mid:yrsceyy3ovisbhkh+state:results>
On 01/11/2012 01:32 PM, Martim wrote:
I think that the c++ point is a good one. Given celix as a universal osgi, how language bindings enter in scene? Other languages communities could benefit by having a osgi implementation. Object oriented languages tend to have a community more open to this kind of development (service oriented) than c community. As a c++ developer I would love if I could use a good osgi implementation with a good community support and Apache as the infrastructure provider in my daily work (currently we are using SOF (http://sof.tiddlyspot.com/), which is good but with almost zero support). Python, Ruby, .net world, all that could benefit too.
Em 11/01/2012 06:04, Pepijn Noltes escreveu:
On Wed, Jan 11, 2012 at 7:55 AM, Marcel Offermans <marc...@luminis.nl> wrote:
On Jan 11, 2012, at 7:30 AM, Alexander Broekhuis wrote:
On the incubator list a discussion is going on about slow/small
poddlings and how to handle these. Celix is one of these poddlings.
For Celix I see the following problems (at least): - No community growth (or not visible..) - No new committers - No releases made
I think a plan is needed to see how we can move Celix towards graduation, how we can get a community, more committers etc.
Agreed. The Incubator PMC in particular is interested in learning how the community intends to address the issues that prevent a podling from graduating. They feel that, especially after being in the incubator for over a year, a graduation plan should be drafted.
Concerning the Celix community growth, at Thales Netherlands we are currently working on a research project in which Celix plays a big part. We are exploring the opportunity to use Celix as a middleware solution - replacing our propertairy solution - in our Radar development. We strongly feel this is going to be a succes and are the moment busy developing the OSGi Device Access Specification in Celix. We are planning to send a patch in the coming weeks. Hopefully this helps in supporting Celix :)
Looking at the three items, the first two will be the most difficult,
and require the most attention. We have been working on visibility, going to the EclipseCon, ApacheCon and several other smaller local community events.
Again, I agree. While it is time to make a release as well, and this might actually help people who want to take a look at Celix, the main issue that needs to be addressed is the size and diversity of the community.
At these events you mention, no doubt you have talked directly to a bunch of people. Could you somehow summarize their feedback?
What other communities can we possibly reach out to, and how?
I'd like to hear what people following this list think of these
problems, and how we can move towards a healthy community. What is holding you back? What is needed to be able to adopt Celix? Feel free to express any concern or opinion you have. Either technical, documentation etc.
One question I've been getting occasionally is: "does Celix also work with C++?".
From a use case point of view, I think we could explore the scenario where you have Java application that uses native code. Mostly, this is now done using JNI which has the downside that it can take the whole JVM down if something goes wrong in the native code. From a robustness point of view, this is unacceptable, and I think Celix can be used to provide a better implementation. With this use case, we can target any Java project that uses native libraries, which in turn might help growing our community. Such use cases are probably interesting enough to write about on all kinds of Java sites.
Good idea. I also think a working scenario with Celix instead of JNI is worth presenting to different Java user groups.
-- Met vriendelijke groet,