• docs/doordrop_ini.md

    From Rob Swindell (on Debian Linux)@VERT to Git commit to main/sbbs/master on Thursday, October 01, 2026 13:35:41
    https://gitlab.synchro.net/main/sbbs/-/commit/0248569e08a9dcb57719369a
    Added Files:
    docs/doordrop_ini.md
    Log Message:
    docs: DOORDROP.INI draft 0.7, renamed from DROPFILE.INI

    The customary file name is now DOORDROP.INI and the environment variable DOORDROP_INI. DROPFILE.INI was too close to the DROPFILE.### of DTS-0001,
    the 1992 named-value proposal the draft cites as prior art, and the
    rename comes before more implementations hardcode the old name (one
    host already shipped against draft 0.3 under it).

    The document moves to docs/doordrop_ini.md. The status section records
    the old names so readers of earlier drafts and that implementation can
    connect the two, and the naming open question is closed.

    Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Debian Linux)@VERT to Git commit to main/sbbs/master on Thursday, October 01, 2026 13:35:41
    https://gitlab.synchro.net/main/sbbs/-/commit/4faf457476657584a07bd244
    Modified Files:
    docs/doordrop_ini.md
    Log Message:
    docs: DOORDROP.INI draft 0.7: add ATASCII

    ATASCII joins the character set names for COMM_CHARSET and TERM_CHARSET
    and the TERM_TYPE tokens, modelled on PETSCII: it is a glyph set and a control-code set at once, with 0x9B as end of line. The can't-use-this-
    charset message rule notes that CR LF shows as graphics on an ATASCII
    terminal. Closes the ATASCII open question; Synchronet never writes it.

    Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Debian Linux)@VERT to Git commit to main/sbbs/master on Thursday, October 01, 2026 13:35:41
    https://gitlab.synchro.net/main/sbbs/-/commit/943ea2fe7aea9d2cc06b76cf
    Modified Files:
    docs/doordrop_ini.md
    Log Message:
    docs: DOORDROP.INI draft 0.7: why FILE_UTF8 rather than FILE_CHARSET

    Close the "other file encodings" open question with the reason: a bool
    the host may only set for a door that declared UTF-8 support keeps the
    decoding burden on the host, which can always transcode to CP437 or
    UTF-8, whereas an open-ended charset name would oblige every door to
    handle or refuse whatever a host might write.

    Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Debian Linux)@VERT to Git commit to main/sbbs/master on Thursday, October 01, 2026 20:45:34
    https://gitlab.synchro.net/main/sbbs/-/commit/c3fc2fd1248277547a68e89a
    Modified Files:
    docs/doordrop_ini.md
    Log Message:
    docs: DOORDROP.INI draft 0.7: FILE_TIME is a datetime, not ascii

    Add a datetime type, YYYY-MM-DDThh:mm:ss with an optional UTC offset in
    RFC 3339 form, and give FILE_TIME that type instead of carrying its
    shape in prose under an ascii type. Nothing a producer writes changes.
    The date type stays date-only; a reader wanting only the date of a
    datetime takes the text before the T, as NuSkooler suggested on the
    GitHub mirror.

    Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Debian Linux)@VERT to Git commit to main/sbbs/master on Thursday, October 01, 2026 20:45:34
    https://gitlab.synchro.net/main/sbbs/-/commit/806df3e58bf8190825904e41
    Modified Files:
    docs/doordrop_ini.md
    Log Message:
    docs: DOORDROP.INI draft 0.7: COMM_TYPE is what the door sees

    Say that an emulator's TCP connection back to the host is not the
    door's comm type: a DOS door on a DOSBox null-modem port still sees a
    COM port, so its type is fossil or uart. Narrow the connect-back open
    question to a native door that connects itself, note that no known
    door or host does, and list what such a type would need.

    Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net