Daniel Dimov writes:

IGC format is pretty old format with many
problems and needs upgrade. I worked
recently on a software for validity checking
IGC files and have fresh impressions of
these problems. They are:

• No support for Unicode characters - for
  example you are not allowed to put pilot
  name "Jörg Ewald" because the symbol
  ö is forbidden.


• Date and times are separated - you can
  assume when the time goes back - the
  date increases with one, but this is
  just assumption and introduces
  uncertainty. Different software can do
  different thing in this case. Some
  manufacturers write additional date
  header in the middle of the file, but
  this is not allowed by the standard.


• The year is represented with only 2
  digits.


• No standard way to fully describe all
  variants of task or record you are
  flying - again different manufacturers
  found various ways to describe the
  things that the IGC standard is missing.


• No support for logging additional
  external data outside of signature
  scope. For example if you connect your
  flight recorder with external sensor by
  cable/bluetooth and you just want to
  store the external data into the
  tracklog without having to sign it.
  There is no way to do this with the
  current IGC standard.


• No support for multiple signatures (for
  example one signature per block of
  data). Now one bit error in any part of
  the file will render it invalid
  (signature validation will fail).


• No standard algorithm/rules for feeding
  the file data into the hashing function.
  Every manufacturer probably did this in
  a different way.


• No standard hashing and signature
  algorithms. Now the only way to validate
  "the signature" (I'm quoting the
  signature because for some manufacturers
  it's not a real cryptographic signature,
  but some in-house developed hashing
  function) is through vali-xxx.exe.

I hope the new tracklog format will solve
all these problems.