Home | History | Annotate | Line # | Download | only in ref 
      1  1.1  agc 3.2 (MPI) doesn't specify what the unused bits should be set
      2  1.1  agc   to. This may be deliberate but I think it should either say they MUST
      3  1.1  agc   be zero (which I prefer) or that their content is unspecified.
      4  1.1  agc 
      5  1.1  agc 4.2 refers to Content Tags, but 4.3 calls them Packet Tags.
      6  1.1  agc 
      7  1.1  agc 5.5.2 doesn't mention V2 keys.
      8  1.1  agc 
      9  1.1  agc In section 9.1, Schneier is given as the reference for DSA - why not refer to FIPS 186-2, which is freely available? Or, indeed, HAC 11.5.1, available here: http://www.cacr.math.uwaterloo.ca/hac/about/chap11.pdf.
     10  1.1  agc 
     11  1.1  agc Similarly 9.2, TripleDES (which, presumably is EDE 3DES - it'd be good
     12  1.1  agc to be specific) is on some FIPS document which I forget or in HAC
     13  1.1  agc chapter 7 (7.32 in 7.2.3 and 7.4.2).
     14  1.1  agc 
     15  1.1  agc ----
     16  1.1  agc 
     17  1.1  agc In 5.2.1:
     18  1.1  agc 
     19  1.1  agc "0x10: Generic certification of a User ID and Public Key packet."
     20  1.1  agc 
     21  1.1  agc Does this mean that the signature is over the User ID packet and the Public Key packet, concatenated, in that order? Or what?
     22  1.1  agc 
     23  1.1  agc Also, what on earth does:
     24  1.1  agc 
     25  1.1  agc        Note that all PGP "key signatures" are this type of
     26  1.1  agc        certification.
     27  1.1  agc 
     28  1.1  agc mean?
     29  1.1  agc 
     30  1.1  agc In 5.2.2:
     31  1.1  agc 
     32  1.1  agc    "The data being signed is hashed, and then the signature type and
     33  1.1  agc    creation time from the signature packet are hashed (5 additional
     34  1.1  agc    octets)."
     35  1.1  agc 
     36  1.1  agc is unclear, suggest:
     37  1.1  agc 
     38  1.1  agc "The concatenation of the data to be signed, the signature type and
     39  1.1  agc creation time from the signature packet (5 additional octets) is hashed."
     40  1.1  agc 
     41  1.1  agc In 5.9:
     42  1.1  agc 
     43  1.1  agc    " - File name as a string (one-octet length, followed by file name),
     44  1.1  agc        if the encrypted data should be saved as a file."
     45  1.1  agc 
     46  1.1  agc but no mention of what if it shouldn't be saved as a file. 0 length,
     47  1.1  agc perhaps?
     48  1.1  agc 
     49  1.1  agc Then:
     50  1.1  agc 
     51  1.1  agc    " - A four-octet number that indicates the modification date of the
     52  1.1  agc        file, or the creation time of the packet, or a zero that
     53  1.1  agc        indicates the present time."
     54  1.1  agc 
     55  1.1  agc I would _guess_ that it means modification date of the file if there's
     56  1.1  agc a filename, the creation time if there isn't. I have no idea what zero
     57  1.1  agc is supposed to mean. Nothing, would be the obvious interpretation -
     58  1.1  agc "the present time" is nonsensical.
     59  1.1  agc 
     60  1.1  agc Once more, when I know what its supposed to mean, I'll suggest
     61  1.1  agc wording.
     62  1.1  agc 
     63  1.1  agc ------
     64  1.1  agc 
     65  1.1  agc 5.2.3.5 Issuer
     66  1.1  agc 
     67  1.1  agc should be:
     68  1.1  agc 
     69  1.1  agc 5.2.3.5 Issuer key ID
     70  1.1  agc 
     71  1.1  agc A tiny point, I know, but it made it hard to find.
     72  1.1  agc 
     73  1.1  agc Key algorithms ... these are used in various contexts, and there's a
     74  1.1  agc list in 9.1 - some of these are clearly unsuitable in some contexts -
     75  1.1  agc for example, one would not expect to see RSA Ecnrpyt-Only (3) in a
     76  1.1  agc signature. But I can't find any language saying anything about
     77  1.1  agc this. Are there any rules?
     78  1.1  agc 
     79