Whiteboard and GSoC2008

Hi.
I followed Asterix's advice regarding implementing
whiteboard in Gajim
and GSoC2008 and started a discussion at XMPP Standards
list:
http://mail.jabber.org/pipermail/standards/2008
-March/018084.html

As I summarized in my post (which was later developed by
others):
there are many implementations out there but they are not
interoperable. The closest to current XSF movements in
standarization
of XML shared editing is Psi whiteboard by Joonas Govenius.
AFAIK it
implements most of the specification in this doc:
http://
www.xmpp.org/extensions/inbox/sxe.html

So the question now is whether to apply for GSoC2008 with
this
project. I think this would be really valuable for Gajim
community but
I did not get any direct answers neither regarding 'the best
way' of
implementing whiteboard nor chances of this project to be
accepted for
GSoC2008.

Asterix said mentioned 3 other proposals on MUC
(http://
www.gajim.org/room_logs/2008/2/28.html):
* test framework: some sort of automated tests to test
features in
Gajim to avoid regressions
* plugin system
* BOSH thing

I'll write my point of view of test framework and plugins
system,
because I think these would be also valuable for Gajim
itself and I'd
could apply for them in upcoming GSoC2008.

I'll be glad for any comments regarding whiteboarding (and
also Gajim
projects for GSoC2008 in general).

-- 
Regards -- Mateusz Biliński
_______________________________________________
Gajim-devel mailing list
Gajim-develgajim.org

https://lists.gajim.org/cgi-bin/listinfo/gajim-devel

Mateusz Biliński wrote:
> Hi.
> I followed Asterix's advice regarding implementing
whiteboard in Gajim
> and GSoC2008 and started a discussion at XMPP Standards
list:
> http://mail.jabber.org/pipermail/standards/2008
-March/018084.html

Thanks for the taking the initiative. 

Will Gajim apply on its own or will Gajim-related projects
be mentored
under the umbrella of the XMPP Standards Foundation again
this year?
(Assuming the XSF is accepted... 

> As I summarized in my post (which was later developed
by others):
> there are many implementations out there but they are
not
> interoperable. The closest to current XSF movements in
standarization
> of XML shared editing is Psi whiteboard by Joonas
Govenius. AFAIK it
> implements most of the specification in this doc:
> http://
www.xmpp.org/extensions/inbox/sxe.html

Whiteboarding is a fun feature but given the fact that we (=
the XMPP
developer community) haven't been able to agree on a
solution or even if
this is a problem that the community needs or wants to solve
in general,
it seems like possibly a risky project to purse for the
Summer of Code.
However I am open to changing my mind about that.

> So the question now is whether to apply for GSoC2008
with this
> project. I think this would be really valuable for
Gajim community but
> I did not get any direct answers neither regarding 'the
best way' of
> implementing whiteboard nor chances of this project to
be accepted for
> GSoC2008.
> 
> Asterix said mentioned 3 other proposals on MUC
> (http://
www.gajim.org/room_logs/2008/2/28.html):
> * test framework: some sort of automated tests to test
features in
> Gajim to avoid regressions

Important, but kind of boring. But maybe you get excited
about it.  I
think it's important to work on something that excites you,
not just
something that would be good for the community. Yes it's
hard to balance
those things sometimes...

> * plugin system

Plugins are always cool. 

> * BOSH thing

IMHO BOSH will become more widespread (we hope to launch it
soon at
jabber.org once we install ejabberd 2.0.0 on a new machine)
so support
for it in Gajim would be great.

Other projects I think are interesting (I am not sure if
Gajim already
supports these, my apologies if it does -- I'm a Psi user
but would be
happy to test Gajim on the Mac once that port is stable):

- support for link-local messaging (XEP-0174)

- support for PEP (XEP-0163) -- also to be launched at
jabber.org soon
and it enables lots of fun features like integration with
blogs and
music players and so on

- support for message archiving (XEP-0136) -- this may be
one of the
community's priorities for 2008

I'm sure other people have more ideas.

Peter

-- 
Peter Saint-Andre
https://stpeter.im/



_______________________________________________
Gajim-devel mailing list
Gajim-develgajim.org

https://lists.gajim.org/cgi-bin/listinfo/gajim-devel

2008/3/4 Peter Saint-Andre <stpeterstpeter.im>:
> Mateusz Biliński wrote:
>  > Hi.
>  > I followed Asterix's advice regarding
implementing whiteboard in Gajim
>  > and GSoC2008 and started a discussion at XMPP
Standards list:
>  > http://mail.jabber.org/pipermail/standards/2008
-March/018084.html
>
>  Thanks for the taking the initiative. 

My pleasure 

>  Will Gajim apply on its own or will Gajim-related
projects be mentored
>  under the umbrella of the XMPP Standards Foundation
again this year?
>  (Assuming the XSF is accepted... 

I don't want to speak on behalf of Asterix but he suggested
on MUC
that Gajim will use XSF umbrella this year:
#v+
[15:13] <vArDo> hi all, is Gajim going to apply for
mentoring projects
GoogleSoC2008?
[15:18] <Asterix> vArDo: YES
[15:18] <Asterix> vArDo: through XSF
[15:19] <Asterix> as last 2 past years
#v-

>  [cut!]
>  Whiteboarding is a fun feature but given the fact that
we (= the XMPP
>  developer community) haven't been able to agree on a
solution or even if
>  this is a problem that the community needs or wants to
solve in general,
>  it seems like possibly a risky project to purse for
the Summer of Code.
>  However I am open to changing my mind about that.

It is what I'm afraid of since I've read all the discussions
at
standards list. This my be probably not the best time for
implementing
whiteboarding (in context of standardization), on the other
hand it
would be great to have whiteboard in Gajim which could
interoperate
with i.e. Joonas's implementation in Psi (and after GSoC
work in
parallel on whiteboarding in two independent clients).
Decision will
be probably made by XMPP developers community so I've just
wanted to
know whether there any chances for it on GSoC2008. It seems
that these
are not very big currently 

It would be great if you could ask other people at XSF how
they see
this. Is there any sense for applying with this project or
will it
meet strong rejection.

>  > * test framework: some sort of automated tests to
test features in
>  > Gajim to avoid regressions
>
>  Important, but kind of boring. But maybe you get
excited about it.  I
>  think it's important to work on something that excites
you, not just
>  something that would be good for the community. Yes
it's hard to balance
>  those things sometimes...

I thought of including automatic GUI testing stuff in it so
it would
be more fun. But yes - writing tests might be boring :]

>  > * plugin system
>
>  Plugins are always cool. 

I'll probably write sth longer later, but in short this
would include
help with reorganizing how events work (there's a ticket for
this
already), and creating some hooks in gui (main menu, roster,
chat
window, context menu). IBM's Sametime has a pretty good
documentation
on it as they have generalized Sametime in version 7.5 and
build a
pretty nice framework. I've developed plugin for it during
my
internship at IBM so I have some experience in plugins
system in
(commercial) IM client.

This would be really great to have because I remember that
was really
disappointed by Psi approach. I've been waiting for plugins
system
there for a long time and AFAIK Psi still has some Qt4
issues and
plugins system is further down the road. Gajim is written in
Python so
it's easier to extend it but it's hard to organize these
extensions so
plugins system would be awesome IMHO 

>  > * BOSH thing
>
>  IMHO BOSH will become more widespread (we hope to
launch it soon at
>  jabber.org once we install ejabberd 2.0.0 on a new
machine) so support
>  for it in Gajim would be great.

You mean XMPP over BOSH?

>  Other projects I think are interesting (I am not sure
if Gajim already
>  supports these, my apologies if it does -- I'm a Psi
user but would be
>  happy to test Gajim on the Mac once that port is
stable):
>  [cut!]
>  - support for message archiving (XEP-0136) -- this may
be one of the
>  community's priorities for 2008

And it might be a really great project to implement for
GSoC. I'm
personally interested in making this work in Gajim ASAP as I
use
multiple OSes.

-- 
Regards -- Mateusz Biliński
_______________________________________________
Gajim-devel mailing list
Gajim-develgajim.org

https://lists.gajim.org/cgi-bin/listinfo/gajim-devel

2008/3/5 Peter Saint-Andre <stpeterstpeter.im>:
>  I think it would be great to have an implementation of
SXE whiteboarding
>  in Psi and an implementation in Gajim!
>
>  Unfortunately, we have competing proposals for
whiteboarding over XMPP.
>  The XMPP Council needs to decide how to proceed. The
"least-worst"
>  solution may be for the XSF to publish both as
informational specs and
>  "let the market decide" about the best
approach. If both Psi and Gajim
>  support the SXE approach (I think they are the two
most popular
>  Jabber-only clients) then that would be part of the
market
>  decision-making process. 

Yes, indeed this would be a market decision. The biggest
advantage of
this of would be that most of 'pure-jabber' user could
whiteboard with
each other. Nice perspective 

>  And of course SXE whiteboarding could be written as a
plugin.  But
>  you'll need to chat with Asterix to see if he thinks
this is a good project.

This is Asterix's proposal from Gajim's MUC channel, so
probably he
thinks this one is crucial for Gajim's development.

>  > You mean XMPP over BOSH?
>  Yes. But perhaps Gajim already includes support for
this somewhere in SVN?

BOSH in Gajim is also Asterix's suggestion, so I don't
suppose there
has been made any significant implementation stuff in this
area.

>  > And it might be a really great project to
implement for GSoC. I'm
>  > personally interested in making this work in
Gajim ASAP as I use
>  > multiple OSes.
>
>  This also may be in Gajim SVN. I don't know what's in
Gajim and what
>  isn't, so chat with Asterix about it. 

I found a ticket related to Message Archiving in Gajim's
Trac:
http://trac.gajim.o
rg/ticket/3593
So this is probably open for implementation too.

>  BTW, all reports are that Asterix is a great mentor so
I hope you can
>  find a project that will be accepted!

Great to hear that  I also
hope that we'll be able to find valuable
project for Gajim's community that wil be accepted (and also
will be
FUN :]).

Thanks for your comments.

--
Regards -- Mateusz Biliński
_______________________________________________
Gajim-devel mailing list
Gajim-develgajim.org

https://lists.gajim.org/cgi-bin/listinfo/gajim-devel

Hi,

I am happy to see that you are interested in hacking on
Gajim during
GSoC, Mateusz.

A few days ago Asterix asked whether I have any ideas for
possible GSoC
projects. After a short discussion we came up with the
following list:

1) Plugin system
2) testframework
3) BOSH

In my opinion those are more valuable to Gajim than the
whiteboard you
mentioned. Of course whiteboarding is a cool feature but let
me explain
why I would prefer the plug-in system for now.

Working on architecture is a priority after 0.12. We have to
kill some
complexity but also make it flexible enough to allow
plugins.
That you already have experiences with plug-in system is
great, as we
don't.

I believe it will be much easier to add new features once
the
architecture is reworked. Furthermore testing could be
simplified
because me might be able to test different parts
independently of each
other.

Of course you can start with a whiteboarding plugin if you
have the time
and the passion 
Maybe there has been some progress on the XSF front by then.



Great to have this discussion here. I see lots of possible
great new
features, but still have to keep an eye on the
maintainability of Gajim.


Best Regards,
Stephan Erb

_______________________________________________
Gajim-devel mailing list
Gajim-develgajim.org

https://lists.gajim.org/cgi-bin/listinfo/gajim-devel

On Fri, Mar 7, 2008 at 1:57 PM, Stephan Erb <steve-eh3c.de> wrote:
>  I am happy to see that you are interested in hacking
on Gajim during
>  GSoC, Mateusz.
>
>  A few days ago Asterix asked whether I have any ideas
for possible GSoC
>  projects. After a short discussion we came up with the
following list:
>
>  1) Plugin system
>  2) testframework
>  3) BOSH
>
>  In my opinion those are more valuable to Gajim than
the whiteboard you
>  mentioned. Of course whiteboarding is a cool feature
but let me explain
>  why I would prefer the plug-in system for now.

I fully agree.

>  Working on architecture is a priority after 0.12. We
have to kill some
>  complexity but also make it flexible enough to allow
plugins.
>  That you already have experiences with plug-in system
is great, as we
>  don't.
>
>  I believe it will be much easier to add new features
once the
>  architecture is reworked. Furthermore testing could be
simplified
>  because me might be able to test different parts
independently of each
>  other.

I've talked to Asterix today, he said he had talked to you,
he
proposed plugins system as most wanted and I agreed. So
plug-in system
will be the main project I'll propose in application for
GSoC. As a
second-chance project (if XSF for some reason would reject
this one) I
thought of 'message archiving' because (as stpeter said) XSF
wants to
make it more popular in 2008. What do you think about that?

>  Of course you can start with a whiteboarding plugin if
you have the time
>  and the passion 
>  Maybe there has been some progress on the XSF front by
then.

I'd like to implement it in Gajim for sure some day, but
first things
first. I think Gajim should benefit from GSoC2008 and
plug-ins system
seems to be the right way.

>  Great to have this discussion here. I see lots of
possible great new
>  features, but still have to keep an eye on the
maintainability of Gajim.

Yeap. The faster Gajim gets plug-in system the better. 

We agreed with Asterix that at the end of GSoC2008 apart
from plug-in
system itself there should be two (or maybe one) simple
(example)
plug-ins to demonstrate the way how to create own.

Asterix have one in mind but he had to leave so we'll
probably talk in
the evening.

I thought of Gajim integration with Google maps (where
people in
roster currently are) or Google calendar (add events from
texts in
conversation, get notification of upcoming events - although
I don't
know whether there isn't some kind of a bot made by Google)
or maybe a
simple SMS plug-in.

Do you have any ideas? I think these plug-ins would be a
really nice
(and easy to check) target to achieve during GSoC.

-- 
Regards -- Mateusz Biliński
_______________________________________________
Gajim-devel mailing list
Gajim-develgajim.org

https://lists.gajim.org/cgi-bin/listinfo/gajim-devel

On Fri, 2008-03-07 at 16:33 +0100, Mateusz BiliƄski wrote:
> On Fri, Mar 7, 2008 at 1:57 PM, Stephan Erb
<steve-eh3c.de> wrote:
> snip
> >
> >  In my opinion those are more valuable to Gajim
than the whiteboard you
> >  mentioned. Of course whiteboarding is a cool
feature but let me explain
> >  why I would prefer the plug-in system for now.
> 
> I fully agree.
>
> snip
> 
> I've talked to Asterix today, he said he had talked to
you, he
> proposed plugins system as most wanted and I agreed. So
plug-in system
> will be the main project I'll propose in application
for GSoC. As a
> second-chance project (if XSF for some reason would
reject this one) I
> thought of 'message archiving' because (as stpeter
said) XSF wants to
> make it more popular in 2008. What do you think about
that?

That's a reasonable decision and will be very valuable to
Gajim and its
future. I really appreciate your decision. 

> >  Of course you can start with a whiteboarding
plugin if you have the time
> >  and the passion 
> >  Maybe there has been some progress on the XSF
front by then.
> 
> I'd like to implement it in Gajim for sure some day,
but first things
> first. I think Gajim should benefit from GSoC2008 and
plug-ins system
> seems to be the right way.



> >  Great to have this discussion here. I see lots of
possible great new
> >  features, but still have to keep an eye on the
maintainability of Gajim.
> 
> Yeap. The faster Gajim gets plug-in system the better.

> 
> We agreed with Asterix that at the end of GSoC2008
apart from plug-in
> system itself there should be two (or maybe one) simple
(example)
> plug-ins to demonstrate the way how to create own.
> 
> Asterix have one in mind but he had to leave so we'll
probably talk in
> the evening.
> 
> I thought of Gajim integration with Google maps (where
people in
> roster currently are) or Google calendar (add events
from texts in
> conversation, get notification of upcoming events -
although I don't
> know whether there isn't some kind of a bot made by
Google) or maybe a
> simple SMS plug-in.
> 
> Do you have any ideas? I think these plug-ins would be
a really nice
> (and easy to check) target to achieve during GSoC.

This google integration is a very good idea. By then PEP
GeoIP should be
in place and you could use your plugin to fill it life. I
couldn't think
of a GUI for it yet, you found it 

This could also show people how to modify the GUI: You might
render a
small globe in the roster (next to the avatar) when a
contact has
published his current position and activity.

What do you mean with a SMS plug-in? Do you want to interact
with your
mobilephone or would it be some kind of sms transport?

Another possible plugin I can think of is integration with
the evolution
address book. Afair the conduit guys are working on python
bindings for
some time now.

Of course there is much more work to be done. We have many
existing
features that should be outsourced as plugins. 
(e.g. AWN toolbar integration and maybe some less frequently
used
features like the latex formula rendering).
But imho this porting should be our task, not yours. 

Talk to you soon. 

Best Regards.
Stephan Erb




_______________________________________________
Gajim-devel mailing list
Gajim-develgajim.org

https://lists.gajim.org/cgi-bin/listinfo/gajim-devel

On Sun, Mar 9, 2008 at 11:35 PM, Stephan Erb <steve-eh3c.de> wrote:
>  > I've talked to Asterix today, he said he had
talked to you, he
>  > proposed plugins system as most wanted and I
agreed. So plug-in system
>  > will be the main project I'll propose in
application for GSoC. As a
>  > second-chance project (if XSF for some reason
would reject this one) I
>  > thought of 'message archiving' because (as
stpeter said) XSF wants to
>  > make it more popular in 2008. What do you think
about that?
>
>  That's a reasonable decision and will be very valuable
to Gajim and its
>  future. I really appreciate your decision.

So now that decision is made it's time for hands-on job :]

>  > Do you have any ideas? I think these plug-ins
would be a really nice
>  > (and easy to check) target to achieve during
GSoC.
>  This google integration is a very good idea. By then
PEP GeoIP should be
>  in place and you could use your plugin to fill it
life. I couldn't think
>  of a GUI for it yet, you found it 

Sounds pretty good although Asterix came out with sth that
might be
more appropriate as example plug-in (details below).

>  This could also show people how to modify the GUI: You
might render a
>  small globe in the roster (next to the avatar) when a
contact has
>  published his current position and activity.

OK. Another place to include in GUI hooks of plug-in system.
[Added to my notes]

>  What do you mean with a SMS plug-in? Do you want to
interact with your
>  mobilephone or would it be some kind of sms
transport?

I rather thought of using external services - nothing to do
with XMPP.
Just show that you can use Gajim to connected external data
with
contacts in roster. For example it could be done by using
urllib and
sending data using form on some GSM provider site (in Poland
one of
three operators allow this, there are also APIs for paid
SMS
gateways).

>  Another possible plugin I can think of is integration
with the evolution
>  address book. Afair the conduit guys are working on
python bindings for
>  some time now.

Yeap. I thought of it too. But I suppose we will postpone
this one for
after-GSoC times.

>  Of course there is much more work to be done. We have
many existing
>  features that should be outsourced as plugins.
>  (e.g. AWN toolbar integration and maybe some less
frequently used
>  features like the latex formula rendering).
>  But imho this porting should be our task, not yours.


Probably you're right. I think currently we should focus on
implementing a good plug-in system and then many new doors
will open.

Porting some parts of Gajim to plug-ins would be a pretty
good testing
phase for plug-ins system itself. I mean whether system is
generalized
enough.

   Plug-in that Asterix came up with: Advanced Notification
Control -
it is currently partly implemented in svn, though not
visible. But
probably you're more aware of this than I am 
   For all people that read this mailing list but don't know
what ANC
is. In short: you're able to assign customized notifications
(sounds,
pop-up windows, etc.) with events that happen in Gajim -
some becomes
Online/Offline, someone writes something, someone changes
status
message, etc.
   This plug-in would show how to invoke notifications in
plug-in as
well as connect into data stream and events in Gajim. Maybe
some GUI
hooks would also be used.
   What do you think about it? I agreed with Asterix that
this would
be a pretty good example for future plug-in developers.

   The second example plug-in we thought of is an 'acronym
expansions'
plug-in. This would be of course much simplier than ANC. In
short: you
write a msg with acronym 'IMO' and hit enter. Before sending
message
Gajim would automatically expand acronym to 'in my opinion'.
Of course
these acronym would be customizable. People would be able to
add for
example 'MYBAN' which would exand to hardcoded bank account
number.
   So what do you think about his one? :>

-- 
Regards -- Mateusz Biliński
_______________________________________________
Gajim-devel mailing list
Gajim-develgajim.org

https://lists.gajim.org/cgi-bin/listinfo/gajim-devel