The user who is talking is shown in green but screen-readers cannot detect this, so there should be a keyboard shortcut to speak out a user's state.

This is not an error

Currently the VU-meter does an average of the samples recorded. This should be replaced by calculating the number of dB. lnbsndfile might have a way to do this: http://www.mega-nerd.com/libsndfile/

http://www.bearware.dk/phorum/read.php?1,4090

Make it so the volume sliders "jump" to the default level when you're moving the slider close to default (in the same was VLC does).

Currently all TTS events are hard-coded to English.

Fragment activity which shows video sessions (part of MainActivity).

TeamTalk JNI DLL currently doesn't support extracting video bitmaps.

Limit amount of data traffic can be used on a single user account (to prevent abuse)

(in Pro-edition)

Maybe use the bProtected field to tell whether password should be changed. On UserAccount blank password is no change?

http://www.bearware.dk/phorum/read.php?1,3929

automatically position users seems to be broken, although I have to have a greater sample size before I know this. it seems as though it doesn't seem to continuously update the positioning as time goes by, and you have to keep going into the position users dialogue and hit default, over and over and over. Sounds like a bug.

http://www.bearware.dk/phorum/read.php?1,4094

An activity which is used to configure the Android client. Basically port what is in the TeamTalk 5 Qt client's preferences dialog.

Linux, Mac and Raspbian SDKs should also have Java samples included. Just copy Android's classes to the other SDKs and make a small sample.

If only 'self' is in channel then don't transmit voice, unless 'self' is subscribing to SUBSCRIBE_VOICE (to conserve bandwidth).

The chat and files tabs can not be navigated accessibly. TalkBack reads Chat/Files/Desktop but these items can not be clicked. It would be nice if the app has a few separate tabs such as chat, files, desktop, history etc.

Reported here: http://bearware.dk/?p=279

The incoming messages does not start a new window, so we can not understand whom the message was recieved from.

Reported here: http://bearware.dk/?p=279

Currently it's only possible to ban at server level.

"Preferences -> Sound System" sound input/output devices should have "No Sound Device" option (start client without sound).

Currently voice is transmitted to a channel even if it doesn't have any users (besides oneself).

In Preferences->General there should be an option to automatically disable voice transmission after a certain period of time in case Push-to-talk is stuck or voice activation is never deactivated.

TeamTalk5.acquireUserDesktopWindow() requires a bit of CPU so it should be run in a thread instead.

Main thread currently decodes all video sessions. Each video widget (uservideowidget.cpp) could decode video frames in a separate thread by calling TT_AcquireUserVideoCaptureFrame().

I can not find the settings for microphone and speaker. I have found sound system on preferences but there is no setting to change.

Reported here: http://bearware.dk/?p=279

Some users change settings and cannot figure out to revert them

Currently working directory is first checked for TeamTalk5.ini afterwards %APPDATA%

Fragment activity in MainActivity which shows shared desktops.

TeamTalk JNI DLL currently doesn't support extracting desktop bitmaps.

Channel activity currently only allows setting name and topic of channel. Remaining settings in TeamTalk 5 Qt client should be put in Android client as well.

Currently minimum is 8-bit

TeamTalk uses an old API for video capture on Linux, i.e. V4L. It should be replaced by a new video capture API.

A channel in a .tt file is always created as "default channel type". Classroom, no voice activation, etc channel types should be possible to specify.

http://msdn.microsoft.com/en-us/library/windows/desktop/dd407023(v=vs.85).aspx
http://msdn.microsoft.com/en-us/library/windows/desktop/dd390172(v=vs.85).aspx
http://msdn.microsoft.com/en-us/library/windows/desktop/dd406916(v=vs.85).aspx

Writing Transform Filters
http://msdn.microsoft.com/en-us/library/ms788165.aspx

VMRWindowlessControl9::GetCurrentImage method:

http://msdn.microsoft.com/en-us/library/dd390541(v=vs.85)

The BEST : Programming Microsoft DirectShow for Digital Video and Television : Mark.D.Pesce, Microsoft Press
And chapt11 is"Using the Sample Grabber Filter"...

DSUtil:
http://svn.aegisub.org/branches/aui/vsfilter/dsutil/

Echo channels have been removed. Users should, however, still be able to hear themselves. Currently this can be done by doing Ctrl+4 on one self but there should be an easier way for users to find this option.

Currently only a single window can be shared

Apparently some video devices output top-down bitmaps to the TT video encoder. It only supports bottom-up.

Currently these only work for single selection.

See if PrintWindow() can be used for capturing desktop windows on Windows. Maybe this fixes the problem with overlapping windows.

Is it possible to ban an entire subnet?
For example 88.135.233.x
Where x can be ANY number?

Keep a map of channelid->password so user doesn't have to retype passwords

If view all users is disabled on a user account then it's still possible to do TT_DoUnsubscribe() on a user ID which cannot be seen. E.g. by unsubscribing to one self. The server should report "user not found" error.

Could be a configurable option in Preferences/Display-tab

TeamTalk JNI DLL currently doesn't support this.

Transmit video is currently disabled whenever application starts

Currently only a limited set of video resolutions are supported. TT_InitVideoCaptureDevice() should be able to accept any user-specified resolution