1 files changed, 74 insertions, 0 deletions
diff --git a/TODO.md b/TODO.md
new file mode 100644
index 0000000..12a4a40
--- /dev/null
+++ b/TODO.md
@@ -0,0 +1,74 @@
+This TODO file lists ideas and features for future developments. They are
+more or less ordered according to their benefit, but that is subjective
+to some degree.
+
+Unless written, no activity has been started on the topics.
+
+
+EDI/STI-D Input
+---------------
+Add support for EDI input, so that third-party encoders can be interfaced
+to the multiplexer. Relevant spec: ETSI TS 102 693
+
+It would be beneficial to write and EDI decoding library that can be re-used in
+ODR-DabMod to add EDI input there too.
+
+Stuff in http://git.mpb.li/git/odr-edilib/ could be helpful.
+
+There is experimental support for STI(PI, X)/RTP using UDP, which is used by
+AVT encoders. It does not properly reorder incoming UDP packets nor align
+STI frames with ETI frames.
+
+Also, it doesn't print out nice messages (like ZMQ's "prebuffering complete").
+
+
+Explicit Service Linking
+------------------------
+It is impossible to activate/deactive linkage sets. Commit 5c3c6d7 added
+some code to transmit a FIG0/6 CEI, but this was subsequently reverted
+because it was not tested enough.
+
+
+Inputs for packet data
+----------------------
+It is currently unclear what input formats and sources work for packet data,
+and which ones would make sense to add.
+
+Also, there is no documentation on the possibilites of packet data.
+
+
+Multicast support for the UDP input
+-----------------------------------
+The current UDP input implementation cannot join a multicast group.
+
+
+Improvements for inputs
+-----------------------
+Add statistics to UDP input, in a similar way that ZeroMQ offers statistics.
+This would mean we have to move the packet buffer from the operating system
+into our own buffer, so that we can actually get the statistics.
+
+Also see the STI-D over EDI topic above.
+
+For error handling, there are parts of the code that use return values, parts
+use exceptions. While it works, we could clean it up a bit.
+
+
+Fix DMB input
+-------------
+The code that does interleaving and reed-solomon encoding for DMB is not used
+anymore, and is untested. The relevant parts are src/dabInputDmb* and src/Dmb.cpp
+
+
+Communicate Leap Seconds
+------------------------
+Actually, we're supposed to say in FIG0/10 when there is a UTC leap second
+upcoming, but since that's not trivial to find out because the POSIX time
+concept is totally unaware of that, this is not done. We need to know for EDI
+TIST, and the ClockTAI class can get the information from the Internet, but it
+is not used in FIG0/10.
+
+
+Implement FIG0/20 Service List
+------------------------------
+See ETSI TS 103 176