In this errata, we briefly summarize several errata to RFC 4695, reported by Alfred Hoenes. The listed errata either fix normative problem in RFC 4695, or describe editorial errata that may be particularly confusing to the implementor. The presentation below is brief. See the main text of for a bis version of RFC 4695 that integrates the fixes listed below in a more-complete way, and also includes many other editorial fixes. John Lazzaro, lazzaro@eecs.berkeley.edu --- [1] In Appendix C.1 and Appendix C.2.3 of RFC 4695, an ABNF rule related to System Chapter X is incorrectly defined as: = "__" ["_" ] "__" The correct version of this rule is: = "__" *( "_" ) "__" [2] In Appendix C.6.3 of RFC 4695, the URIs permitted to be assigned to the "url" parameter are not stated clearly. URIs assigned to "url" MUST specify either HTTP or HTTP over TLS transport protocols. In Appendix C.7.1 and C.7.2 of RFC 4695, the transport interoperability requirements for the "url" parameter are not stated clearly. For both C.7.1 and C.7.2, HTTP is REQUIRED and HTTP over TLS is OPTIONAL. [3] Both fmtp lines in both session description examples in Appendix C.7.2 of RFC 4695 contain instances of the same syntax error (a missing ";" at a line wrap after "cm_used=2M0.1.2"). [4] In Appendix D of RFC 4695, all uses of "*ietf-extension" in rules are in error, and should be replaced with "ietf-extension". Likewise, all uses of "*extension" are in error, and should be replaced with "extension". This bug incorrectly lets the null token be assigned to the j_sec, j_update, render, smf_info, and subrender parameters. [5] In Appendix D of RFC 4695, the definitions of the and incorrectly allow lowercase letters to appear in these strings. The correct definitions of these rules appear below: command-type = [A] [B] [C] [F] [G] [H] [J] [K] [M] [N] [P] [Q] [T] [V] [W] [X] [Y] [Z] chapter-list = [A] [B] [C] [D] [E] [F] [G] [H] [J] [K] [M] [N] [P] [Q] [T] [V] [W] [X] [Y] [Z] A = %x41 B = %x42 C = %x43 D = %x44 E = %x45 F = %x46 G = %x47 H = %x48 J = %x4A K = %x4B M = %x4D N = %x4E P = %x50 Q = %x51 T = %x54 V = %x56 W = %x57 X = %x58 Y = %x59 Z = %x5A [5] In Appendix D of RFC 4695, the definitions of the , , and are incorrect. The correct definitions of these rules appear below: nonzero-four-octet = (NZ-DIGIT 0*8(DIGIT)) / (%x30-33 9(DIGIT)) / ("4" %x30-31 8(DIGIT)) / ("42" %x30-38 7(DIGIT)) / ("429" %x30-33 6(DIGIT)) / ("4294" %x30-38 5(DIGIT)) / ("42949" %x30-35 4(DIGIT)) / ("429496" %x30-36 3(DIGIT)) / ("4294967" %x30-31 2(DIGIT)) / ("42949672" %x30-38 (DIGIT)) / ("429496729" %x30-34) four-octet = "0" / nonzero-four-octet midi-chan = DIGIT / ("1" %x30-35) DIGIT = %x30-39 NZ-DIGIT = %x31-39 [6] In Appendix D of RFC4695, the rule is incorrect. The correct definition of this rule appears below. hex-octet = %x30-37 U-HEXDIG U-HEXDIG = DIGIT / A / B / C / D / E / F ; DIGIT as defined in [5] above ; A, B, C, D, E, F as defined in [4] above [7] In Appendix D of RFC4695, the rules and are defined unclearly. The rewritten rules appear below: base64-unit = 4(base64-char) base64-pad = (2(base64-char) "==") / (3(base64-char) "=") ---