- For T4954: SOS representation and improvements in GnuPG, I realized that: ideally speaking, T4956: agent: Discrepancy of handling MPI for the interpretation of signed and unsigned and T4964: ecc: Discrepancy of handling MPI for the interpretation of signed and unsigned should be all fixed. But it's too much, very confusing, and possibly will introduce regressions (for SSH and GpgSM for signed data).
- The capacity of my brain is too small to find/identify/fix all possible problems for this.
- So, I'm focusing possible minimal change(s) for real use case of X448/Ed448.
- Tried to flash BC card for 3.4, and failed because of the card is too old. Asked Achim for sending new one.
- Side effect: confirmed that TTXS reader works well on Windows.
- Backport T4934 to 2.2
- Created: T4957: OpenPGP card protocol 3.4 with Yubikey
- continue on ECDH with X448, minimal change for SOS
- Data object FA support in scdaemon, and possible change of gpg-card/gpg--card-edit to show supported key attributes.
- Worked on gpg--card.
- Added changes for the new card specs 3.4.
- Started work on a new feature to store the public key on ther CERT-3 DO of an OpenPGP card: This wil help to get the same user experience as with X.509 cards.
- New Libgpg-error release.
- Work on the new feature.
- What shall we do with T4541 - I have seen no more information about this and would thus propose to close this as not-a-bug.
- Mostly "technical sales"
- Added support to GpgOL that "gpgsm --learn" is called if an OpenPGP card is inserted with keys for which there are no keystubs. -> Nice hack :-)
- Looked into reported performance problems in Outlook with GpgOL
- Rewrite BCC / OpenPGP and S/MIME combined code in GpgOL to allow for error handling.
- The usual non-code work.