I am fine with enhancing existing ECF applications. As you said it is more practically doable within the given time constrains.
I already went through Salvo. Could not find any expected enhancements mentioned in the wiki. Let me know if there are any wanted enhancements in ECF applications. I am really interested to work in this area.
On 4/2/2012 1:26 PM, tishan pubudu kanishka dahanayakage wrote:
Hi Scott,
I am interested in working with Real-Time Collaboration
systems, specially ShareCode. Are there any features that needed
to be implemented or should I go through Bugzilla and find?
I think the first step would be to make direct contact with the ECF
committer that's primarily responsible for ShareCode (if possible),
and that's Marcelo Mayworm. I haven't spoken to Marcelo in a while,
but hopefully we can reengage him.
Also can I have an explanation about entry two(2), full
applications. As I understood I have to implement a complete
application. If it is so are there any proposed application that
needed to be implemented?
It's not strictly necessary that you implement a complete
application...but you could do so if you wanted to (i.e. if that
appeals to you, and you think you can scope it sufficiently for a
google summer of code project). For example, previous gsoc students
have worked on enhancing existing ECF applications...e.g. Salvo,
ECF's twitter client, ECF's existing real-time Eclipse collaboration
application, the real-time shared editing, etc.
So again...it's not strictly necessary that you create an entire new
application...you can/could work on enhancing/adding to (or
integrating) existing applications. That's a completely valid (and
potentially more realistically doable) project.
For a test framework project, you should probably have a
chat/conversation with committer Markus Kuppe...and
probably me as well at some point (as I'm the main author
of the RSA impl). We could do it on this mailing
list...or via some other medium...whatever works best for
all in terms of timing, etc....as I'm on pacific coast of
UI, Markus is in Europe now (I believe), and I'm assuming
that you are in Sri Lanka.
But you should probably identify and articulate what you
are most interested in doing first.
Scott
On 4/2/2012 5:01 AM, Jasintha Dasanayaka wrote:
HI Scott
Thank you for your thoughts/ideas about Google
Summer of Code projects for ECF. I am really
interested in OSGI Remote Service . So I would like
to develop a test framework for remote
service. I Will do some research and post an
update to dev@ ASAP
Here are some thoughts/ideas about Google
Summer of Code projects for ECF. Many of them
are building upon projects that gsoc students
have worked on in the past...and then become
ECF committers (going from being a gsoc
student to an ECF committer is not automatic,
but if that's what you want to do and your
project is sufficiently valuable, it is a
possible path going forward).
Before I describe some individual projects, I
should say that recently (over past 1.5 years
or so), loosely speaking there have generally
been 4 major technical 'themes' for ECF:
1) Remote services - ECF is one of the impls
of the OSGi remote services (RS) standard, and
the remote service admin (RSA) standard [1].
There's been a lot of work done in this area,
and there are, IMHO, potentially several good
gsoc project for people interested in this
area. ECF has a 'provider architecture',
meaning that it's possible replace the
underlying remoting transport (even at
runtime), and still remain completely
compliant with the RS/RSA standard.
Some ideas in this area:
a) New distribution and discovery providers:
I've been working on a Restlet-based remote
services provider [1a], and other providers
are certainly possible...e.g. ones based upon
other rest frameworks (e.g. jax-rs), etc.
Also there is desire to have more/other
discovery providers (our discovery API is
similarly provider-based)...we already have
zookeeper, dnssd, zeroconf, slp...and others
are certainly possible. Further, several of
the existing providers could probably use some
upgrading of versions, etc.
b) A test framework for testing remote
services. Since remote services by definition
involve inter-process communication, there is
very little in the support for unit testing of
network discovery and/or remote services.
c) Assisting in moving the ECF remote services
implementation up to the latest version of the
OSGi specification. I don't know what's
involved here yet (as I haven't examined the
most recent version of the spec closely), but
I expect that there will be some enhancements
to RS/RSA that ECF would like to support.
d) Example applications, tutorials, and
documentation. ECF has a documentation
sub-project [1b]...and it would be very nice
to have some more extended/complete tutorials
and example around remote services. In fact,
it would be very nice to have some example
remote services running all the time...for
example to allow Eclipse (and ECF) users to
access some service....for
communication/collaboration, for remote
testing (perhaps coordinated with 'b'), for
remote compiling/building, etc. Plenty of
other services are possible (e.g. integration
with google, amazon, or other services, etc).
e) ECF has several unique aspects relative to
other OSGi remote services implementations:
i) We have support for asynchronous remote
services built in already [10]....ii) we have
a provider architecture...allowing consumers
to 'mix and match' discovery and distribution
providers as needed.
2) Example...and/or Full Applications - ECF's
Salvo newsreader is a one such application
[2], and there has also been a lot of work
with Real-Time shared editing, including using
the google wave implementation as a provider
[3a].
3) Real-Time Collaboration Systems. ECF has
already been used to implement Coffee [4],
shared code [5], and other real-time developer
collaboration systems. There is certainly a
lot more room for work here, however...and
with the emphasis on remote services over the
past few years, and the changes in Eclipse
(e.g. toward Orion and e4) I think there are a
lot more new opportunities here.
4) There was a lot of work in voice-over-ip
around 2 years ago...but it seems to have
slowed down some. I'm not really sure
why...as I think it would be great to
jump-start the work in VOIP
again...particularly with the smart phones,
etc.
5) Anything having to do with OSGi on
servers...for mobile, iPod, Android, iPhone,
etc as clients...for real-time collaboration,
communication, games (e.g. see [11],[12]) and
simulation [13]).
Here are some other useful links...See section
s: 'Other Cool Stuff' and 'Presentations' on
[6] and [7], [8], [9].
On 3/29/2012 1:08 PM, tishan pubudu
kanishka dahanayakage wrote:
Hi devs,
I am a third year undergraduate at
University of Moratuwa, Sri Lanka. I
have good knowledge in java and have
studied ECF platform a little during
the past few days. I went through your
project site and wiki. And I found the
project is interesting. So I would
like to contribute to your project. I
already went through your API and also
bug list. But still could not find any
lead to a good idea to implement. So
can anybody enlighten me on this.
