First, I think that 3.2 should go out the door as soon as JOGL and the
security enhancements are done (barring any major difficulties which
delay one or the other). If we can add audio conferencing in parallel
with those two without delaying them, then I’m for doing it in 3.2. But
if we can’t do that, I think it should be put of until 3.3 or later.
Second, I have some questions about what approach we should take:
A general philosophical point: I’m not a fan of building in
functionality done better by specialized apps. E.g., I hope we never
add an image editor to VASSAL. Cyberboard has one for making tokens,
and it’s crappy in comparison with a real image editor, and won’t ever
be decent unless Cyberboard becomes an image manipulation program
instead of a game engine. Same thing with our chat window. When I play
online, I tend to use my IM client becuase it’s better than our chat
window for chatting—and no reasonable amount of work will make our
chat window as good for that as a dedicated IM client. Similarly, this
is why we shouldn’t build a web browser or email client into VASSAL—
we do much better by calling out to the user’s preferred app for web
browsing and sending mail.
That said, my question is: What would the interface for voice be like?
Would we be able to launch the user’s voice software, or would we be
stuck creating a thrid-rate voice client within VASSAL?
What voice software is there to target? Is Skype the only viable
one?
Integrated audio won’t be ready for 3.2, I’m sure. Like Joel says, we don’t want to write our own, but if we can tie into another product that’d be ideal.
Each release really only needs one major new feature. JOGL or security would be enough to justify a new release. If they’re ready at the same time, they can go in together.
I generally agree with the specialized apps comment. However…
I generally avoid IM clients like the plague. Other than graphical smilies can they do that Vassal doesn’t?
Other voice clients: Team Speak. I generally use that as some people I know get their machine insanely bogged down upon starting Skype.
I actually avoid using a voice client with Vassal. I know it would speed things up a bit, but I like keeping a log of the on-line sessions as a recording of what happened. If I’m speaking instead of typing, I loose most of that.
Whenever the audio component / plugin starts to get attention for adding in
might want to poll userbase for what they would prefer to see.
I know of and have used four: Skype, Teamspeak, Ventrilo, Roger Wilco
(defunct I think now).
Skype is very clear high quality but is weak supporting conf. calls of only
up to 5 users, whereas TS and Vent can support many but their sound quality
is finicky (codec setting dependent) recommending codec/bandwidth be set no
higher than the slowest user’s connection of all users without issues. RW
was plain poor
Pidgin is a multi-protocol IM client (and is open-source). My friends
don’t all use the same IM network, so one useful thing is does is
keeps me from needing to run multiple IM clients.
A ZunTzu importer I would classify as a small job, and when I’m in
development mode (as opposed to bug-fix mode, like I am right now)
I’ll frequently switch to some small job when I’m stuck on, frustrated
with, or tired of the large job I’m working on. So there’s a good
chance that I’ll just do it unscheduled some afteroon.
Also, if we have some spare cycles during 3.2, I have a whole slew of
UI polish tasks which I’d like to see happen, things like making sure
that we show useful error dialogs whenever there’s an error, that we’re
consistent about logging errors, that we use the right default extensions
for new files in all of the file choosers, …
