We have “byte-for-byte” and “character-for-character” in the BRs, which means same in current BRs.
To unify expressions and reducing confusion, I have two proposals.
- My first proposal is changing text as below.
We have 55 byte-for-byte, and 1 character-for-character.
OLD:
If the value is a Fully‐Qualified Domain Name or Wildcard Domain Name, then the value MUST be encoded as a character-for-character copy of the dNSName entry value from the subjectAltName extension.
NEW:
If the value is a Fully‐Qualified Domain Name or Wildcard Domain Name, then the value MUST be encoded as a byte-for-byte copy of the dNSName entry value from the subjectAltName extension (excluding the tag, length, and number of unused bits)”
(2) My second proposal is related with definition of “byte”
Although “byte” most commonly consist of 8bits, and defined as 8bits in 2008[2], there are some (I believe) historical definitions for other numbers of bits.
It is because of "byte" was originally meant “a group of bits used to encoded a character or the number of bits transmitted in parallel to and from input-output unit” [1].
For example, in my understanding, (I believe it is not common at all but) byte can be interpreted to 7, 9, 10 bits in historical enviroment, and 16 bits or 32 bits in some Unicode environment.
In my understanding, word “octet” was invented to mean 8 bits, to deal with that situation.
I believe we have some options.
a) define “byte” being 8 bits in BRs.
b) Change “byte-for-byte” to “octet-for-octet”.
c) do not change language related with byte-for-byte.
I think a) is reasonable, I do not have strong opinion about that.
[1] Planning a Computer System : Project Stretch (computerhistory.org), pp.40, https://archive.computerhistory.org/resources/text/IBM/Stretch/pdfs/Buchholz_102636426.pdf,
[2] ISO/IEC_80000-13:2008
We have “byte-for-byte” and “character-for-character” in the BRs, which means same in current BRs.
To unify expressions and reducing confusion, I have two proposals.
We have 55 byte-for-byte, and 1 character-for-character.
OLD:
If the value is a Fully‐Qualified Domain Name or Wildcard Domain Name, then the value MUST be encoded as a character-for-character copy of the dNSName entry value from the subjectAltName extension.
NEW:
If the value is a Fully‐Qualified Domain Name or Wildcard Domain Name, then the value MUST be encoded as a byte-for-byte copy of the dNSName entry value from the subjectAltName extension (excluding the tag, length, and number of unused bits)”
(2) My second proposal is related with definition of “byte”
Although “byte” most commonly consist of 8bits, and defined as 8bits in 2008[2], there are some (I believe) historical definitions for other numbers of bits.
It is because of "byte" was originally meant “a group of bits used to encoded a character or the number of bits transmitted in parallel to and from input-output unit” [1].
For example, in my understanding, (I believe it is not common at all but) byte can be interpreted to 7, 9, 10 bits in historical enviroment, and 16 bits or 32 bits in some Unicode environment.
In my understanding, word “octet” was invented to mean 8 bits, to deal with that situation.
I believe we have some options.
a) define “byte” being 8 bits in BRs.
b) Change “byte-for-byte” to “octet-for-octet”.
c) do not change language related with byte-for-byte.
I think a) is reasonable, I do not have strong opinion about that.
[1] Planning a Computer System : Project Stretch (computerhistory.org), pp.40, https://archive.computerhistory.org/resources/text/IBM/Stretch/pdfs/Buchholz_102636426.pdf,
[2] ISO/IEC_80000-13:2008