RFC 733
This document is obsolete. Please refer to RFC 822.




RFC #   733
NIC # 41952

Obsoletes:  RFC #561  (NIC #18516)
            RFC #680  (NIC #32116)
            RFC #724  (NIC #37435)







                   STANDARD FOR THE FORMAT OF

                  ARPA NETWORK TEXT MESSAGES(1)




                        21 November 1977






                               by


                        David H. Crocker
                      The Rand Corporation

                         John J. Vittal
                  Bolt Beranek and Newman Inc.

                        Kenneth T. Pogran
              Massachusets Institute of Technology

                   D. Austin Henderson, Jr.(2)
                  Bolt Beranek and Newman Inc.





_________________________________________________________________
(1)This work was  supported  by  the  Defense  Advanced  Research
Projects Agency of the Department of Defense, under contract Nos.
N00014-75-C-0661, MDA903-76-C-0212, and DAHC15-73-C0181.

(2)The authors' postal  addresses  are:   D.  Crocker,  The  Rand
Corporation,  Information  Sciences  Dept.,  1700 Main St., Santa
Monica, California 90406; J.  Vittal  &  D.  A.  Henderson,  Bolt
Beranek & Newman, 50 Moulton St., Cambridge, Massachusetts 02138;
and  K.  Pogran,  MIT  Laboratory  for  Computer   Science,   545
Technology  Square, Cambridge, Massachusetts 02139.  The authors'
ARPANET addresses are:  DCrocker at  Rand-Unix,  Vittal  at  BBN-
TenexD, Pogran at MIT-Multics, and Henderson at BBN-TenexD.

                              -iii-





                             PREFACE



     ARPA's  Committee  on  Computer-Aided  Human   Communication
(CAHCOM)  wishes  to promulgate a standard for the format of ARPA
Network text message (mail) headers which  will  reasonably  meet
the  needs  of  the  various  message  service  subsystems on the
Network today.  The  authors  of  this  document  constitute  the
CAHCOM  subcommittee charged with the task of developing this new
standard.

     Essentially, we specify a revision to  ARPANET  Request  for
Comments (RFC) 561, "Standardizing Network Mail Headers", and RFC
680, "Message Transmission Protocol".  This revision removes  and
compacts  portions  of  the  previous  syntax  and  adds  several
features to network address  specification.   In  particular,  we
focus  on  people  and  not  mailboxes  as  recipients  and allow
reference to stored address lists.   We  expect  this  syntax  to
provide  sufficient  capabilities  to  meet most users' immediate
needs and, therefore, give developers enough  breathing  room  to
produce  a new mail transmission protocol "properly".  We believe
that there is enough of a consensus in the Network  community  in
favor  of such a standard syntax to make possible its adoption at
this time.  An earlier draft of this specification was  published
as  RFC  #724, "Proposed Official Standard for the Format of ARPA
Network Messages"  and  contained  extensive  discussion  of  the
background and issues in ARPANET mail standards.

     This specification was developed  over  the  course  of  one
year,  using  the ARPANET mail environment, itself, to provide an
on-going forum for discussing the capabilities  to  be  included.
More   than   twenty   individuals,   from  across  the  country,
participated in this discussion and we would like to  acknowledge
their  considerable  efforts.   The  syntax  of  the standard was
originally specified in the Backus-Naur Form (BNF) meta-language.
Ken  L.   Harrenstien,  of SRI International, was responsible for
re-coding the BNF  into  an  augmented  BNF  which  compacts  the
specification and allows increased comprehensibility.



                               -v-





                            CONTENTS



PREFACE..................................................... iii


Section
   I.  INTRODUCTION.........................................   1

  II.  FRAMEWORK............................................   2

 III.  SYNTAX...............................................   4
       A. Notational Conventions............................   4
       B. Lexical Analysis of Messages......................   5
       C. General Syntax of Messages........................  13
       D. Syntax of General Addressee Items.................  15
       E. Supporting Constructs.............................  15

  IV.  SEMANTICS............................................  17
       A. Address Fields....................................  17
       B. Reference Specification Fields....................  22
       C. Other Fields and Syntactic Items..................  23
       D. Dates and Times...................................  24

   V.  EXAMPLES.............................................  25
       A. Addresses.........................................  25
       B. Address Lists.....................................  26
       C. Originator Items..................................  26
       D. Complete Headers..................................  28


Appendix
   A.  ALPHABETICAL LISTING OF SYNTAX RULES.................  31
   B.  SIMPLE PARSING.......................................  35

BIBLIOGRAPHY................................................  37


Standard for the Format of Text Messages                        1
I. Introduction





                        I.  INTRODUCTION






     This standard specifies a syntax for text messages which are
passed between computer users within the framework of "electronic
mail".  The standard supersedes the informal standards  specified
in  ARPANET  Request  for  Comments  numbers  561, "Standardizing
Network Mail Headers", and 680, "Message Transmission  Protocol".
In  this  document,  a  general framework is first described; the
formal syntax is then specified, followed by a discussion of  the
semantics.  Finally, a number of examples are given.

     This specification is intended strictly as a  definition  of
what  is  to  be  passed between hosts on the ARPANET.  It is NOT
intended to dictate either features which systems on the  Network
are  expected  to support, or user interfaces to message creating
or reading programs.

     A distinction should be made between what the  specification
REQUIRES  and  what  it ALLOWS.  Messages can be made complex and
rich with formally-structured components of information or can be
kept small and simple, with a minimum of such information.  Also,
the standard simplifies the interpretation  of  differing  visual
formats in messages.  These simplifications facilitate the formal
specification and indicate what the OFFICIAL  semantics  are  for
messages.   Only  the  visual aspect of a message is affected and
not the interpretation of information  within  it.   Implementors
may choose to retain such visual distinctions.




Standard for the Format of Text Messages                        2
II. Framework





                         II.  FRAMEWORK




     Since there are many message systems which exist outside the
ARPANET environment, as well as those within it, it may be useful
to consider the general framework, and resulting capabilities and
limitations, provided by this standard.

     Messages are expected to  consist  of  lines  of  text.   No
special provisions are made, at this time, for encoding drawings,
facsimile, speech, or structured text.

     No significant consideration has been given to questions  of
data   compression   or   transmission/storage  efficiency.   The
standard, in fact, tends to be very free with the number of  bits
consumed.   For  example, field names are specified as free text,
rather than special terse codes.

     A general "memo" framework is  used.   That  is,  a  message
consists  of some information, in a rigid format, followed by the
main part of the message, which is text and whose format  is  not
specified  in this document.  The syntax of several fields of the
rigidly-formated  ("header")   section   is   defined   in   this
specification;  some of the header fields must be included in all
messages.  The syntax  which  distinguishes  between  headers  is
specified  separately  from  the  internal  syntax for particular
headers.  This separation is intended to allow  extremely  simple
parsers  to operate on the overall structure of messages, without
concern  for  the  detailed  structure  of  individual   headers.
Appendix B is provided to facilitate construction of these simple
parsers.  In addition to the fields specified in  this  document,
it  is  expected  that  other fields will gain common use.  User-
defined header fields allow systems to extend their functionality
while  maintaining  a uniform framework.  The approach is similar
to that of the TELNET protocol,  in  that  a  basic  standard  is
defined  which  includes  a  mechanism for (optionally) extending
itself.  As necessary, the authors of this document will regulate
the  publishing  of  specifications for these "extension-fields",
through the same mechanisms used to publish this document.

     Such a  framework  severely  constrains  document  tone  and
appearance  and  is  primarily useful for most intra-organization
communications  and  relatively   structured   inter-organization
communication.   A more robust environment might allow for multi-
font, multi-color, multi-dimension encoding  of  information.   A
less  robust  environment,  as  is present in most single-machine
message systems, would more severely constrain the ability to add
fields  and the decision to include specific fields.  In contrast
to paper-based communication, it is interesting to note that  the

Standard for the Format of Text Messages                        3
II. Framework



RECEIVER  of  a  message  can exercise an extraordinary amount of
control over the message's  appearance.   The  amount  of  actual
control  available  to  message  receivers is contingent upon the
capabilities of their individual message systems.


Standard for the Format of Text Messages                        4
III. Syntax





                          III.  SYNTAX



     This  syntax  is  given  in  five  parts.   The  first  part
describes  the  notation  used  in the specification.  The second
part describes the base-level lexical analyzers  which  feed  the
higher-level  parser  described  in the succeeding sections.  The
third part gives a  general  syntax  for  messages  and  standard
header  fields;  and  the  fourth  part  specifies  the syntax of
addresses.  A final part  specifies  some  general  syntax  which
supports the other sections.



A.  NOTATIONAL CONVENTIONS



These specifications are made in an  augmented  Backus-Naur  Form
(BNF).  Differences  from  standard  BNF  involve  the  naming of
rules, the indication of repetition and of "local" alternatives.


1.  Rule naming



Angle brackets ("<", ">") are not used, in general.  The name  of
a   rule  is  simply  the  name  itself,  rather  than  "<name>".
Quotation-marks enclose literal text (which may be  upper  and/or
lower case).  Certain basic  rules  are  in  uppercase,  such  as
SPACE,  TAB, CRLF, DIGIT, ALPHA, etc.  Angle brackets are used in
rule definitions, and in the  rest  of  this  document,  whenever
their presence will facilitate discerning the use of rule names.


2.  Parentheses:  Local alternatives



Elements enclosed in parentheses are treated as a single element.
Thus,  "(elem  (foo  /  bar)  elem)" allows "(elem foo elem)" and
"(elem bar elem)".


3.  * construct:  Repetition



The character "*" preceding an element indicates repetition.  The
full form is:

          <l>*<m>element

indicating at least <l> and at most <m> occurrences  of  element.
Default values are 0 and infinity so that "*(element)" allows any
number, including zero; "1*element" requires at  least  one;  and
"1*2element" allows one or two.

Standard for the Format of Text Messages                        5
III. Syntax
  A. Notational Conventions



4.  <number>element



"<n>(element)" is  equivalent  to  "<n>*<n>(element)";  that  is,
exactly  <n>  occurrences of (element).  Thus 2DIGIT is a 2-digit
number, and 3ALPHA is a string of three alphabetic characters.


5.  # construct:  Lists



A construct "#" is defined, similar to "*", as follows:

                  <l>#<m>element

indicating at least <l> and at most <m> elements, each  separated
by  one or more commas (",").  This makes the usual form of lists
very easy; a rule such as '(element *("," element))' can be shown
as  "1#element".   Wherever this construct is used, null elements
are allowed, but do not  contribute  to  the  count  of  elements
present.   That  is,  "(element),,(element)"  is  permitted,  but
counts as only two  elements.   Therefore,  where  at  least  one
element  is  required,  at  least  one  non-null  element must be
present.


6.  [optional]



Square  brackets  enclose  optional  elements;  "[foo  bar]"   is
equivalent to "*1(foo bar)".


7.  ; Comments



A semi-colon, set off some distance to the right  of  rule  text,
starts  a  comment which continues to the end of line.  This is a
simple way  of  including  useful  notes  in  parallel  with  the
specifications.



B.  LEXICAL ANALYSIS OF MESSAGES




1.  General Description

A message consists of headers and, optionally,  a  body  (i.e.  a
series of text lines).  The text part is just a sequence of lines
containing ASCII characters; it is separated from the headers  by
a null line (i.e., a line with nothing preceding the CRLF).


Standard for the Format of Text Messages                        6
III. Syntax
  B. Lexical Analysis



a.  Folding and unfolding of headers

    Each header item can be viewed as a single, logical  line  of
    ASCII characters.  For convenience, the field-body portion of
    this conceptual entity can  be  split  into  a  multiple-line
    representation  (i.e.,  "folded").   The general rule is that
    wherever there can be linear-white-space  (NOT  simply  LWSP-
    chars), a CRLF immediately followed by AT LEAST one LWSP-char
    can instead be inserted.  (However, a header's name  and  the
    following  colon  (":"),  which occur at the beginning of the
    header item, may NOT be folded onto multiple  lines.)   Thus,
    the single line

       To:  "Joe Dokes & J. Harvey" <ddd at Host>, JJV at BBN

    can be represented as

       To:  "Joe Dokes & J. Harvey" <ddd at Host>,
            JJV at BBN

    and

       To:  "Joe Dokes & J. Harvey"
                        <ddd at Host>,
        JJV at BBN

    and

       To:  "Joe Dokes
        & J. Harvey" <ddd at Host>, JJV at BBN

    The  process  of  moving  from  this   folded   multiple-line
    representation   of   a  header  field  to  its  single  line
    representation will  be  called  "unfolding".   Unfolding  is
    accomplished  by  regarding  CRLF  immediately  followed by a
    LWSP-char as equivalent  to  the  LWSP-char.

b.  Structure of header fields

    Once header fields have been unfolded, they may be viewed  as
    being  composed  of  a  field-name followed by a colon (":"),
    followed by a field-body.  The field-name must be composed of
    printable  ASCII  characters  (i.e.,  characters  which  have
    values between 33.  and  126.,  decimal,  except  colon)  and
    LWSP-chars.   The  field-body  may  be  composed of any ASCII
    characters (other than  an  unquoted  CRLF,  which  has  been
    removed by unfolding).

    Certain field-bodies of  header  fields  may  be  interpreted
    according  to  an internal syntax which some systems may wish
    to parse.  These fields will be referred to  as  "structured"
    fields.    Examples   include  fields  containing  dates  and

Standard for the Format of Text Messages                        7
III. Syntax
  B. Lexical Analysis



    addresses.  Other fields, such as "Subject"  and  "Comments",
    are regarded simply as strings of text.

    NOTE:  Field-names, unstructured field bodies and  structured
    field  bodies  each  are  scanned  by  their own, INDEPENDENT
    "lexical" analyzer.

c.  Field-names

    To aid in the creation and reading of field-names,  the  free
    insertion  of  LWSP-chars  is  allowed in  reasonable places.

    Rather than obscuring the syntax specification for field-name
    with  the explicit syntax for these LWSP-chars, the existence
    of a "lexical" analyzer is assumed.  The analyzer  interprets
    the  text  which  comprises  the  field-name as a sequence of
    field-name atoms (fnatoms) separated by LWSP-chars

    Note that ONLY LWSP-chars may occur between the fnatoms of  a
    field-name and that CRLFs may NOT.  In addition, comments are
    NOT lexically recognized, as such, but parenthesized  strings
    are  legal  as  part  of  field-names.  These constraints are
    different from what is permissible  within  structured  field
    bodies.   In  particular,  this means that header field-names
    must wholly occur on the FIRST line of a folded  header  item
    and may NOT be split across two or more lines.

d.  Unstructured field bodies

    For  some  fields,  such  as  "Subject"  and  "Comments",  no
    structuring is assumed; and they are treated simply as texts,
    like those in the message body.  Rules of  folding  apply  to
    these  fields, so that such field bodies which occupy several
    lines must therefore have the  second  and  successive  lines
    indented by at least one LWSP-char.

e.  Structured field bodies

    To aid in the creation and reading of structured fields,  the
    free  insertion  of linear-white-space (which permits folding
    by inclusion of  CRLFs)  is  allowed  in  reasonable  places.
    Rather  than  obscuring  the  syntax specifications for these
    structured fields  with  explicit  syntax  for  this  linear-
    white-space,  the  existence of another "lexical" analyzer is
    assumed.  This analyzer does not apply for field bodies which
    are  simply unstructured strings of text, as described above.
    It provides an interpretation of the unfolded text comprising
    the  body  of  the  field  as  a sequence of lexical symbols.
    These symbols are:

            -  individual special characters
            -  quoted-strings

Standard for the Format of Text Messages                        8
III. Syntax
  B. Lexical Analysis



            -  comments
            -  atoms

    The first three of these symbols are self-delimiting.   Atoms
    are  not; they therefore are delimited by the self-delimiting
    symbols and by linear-white-space.  For the purposes  of  re-
    generating sequences of atoms and quoted-strings, exactly one
    SPACE is assumed to exist and should be  used  between  them.
    (Also,  in  Section  III.B.3.a,  note  the  rules  concerning
    treatment of multiple continguous LWSP-chars.)

    So, for example, the folded body of an address field

            ":sysmail"@   Some-Host,
            Muhammed(I am   the greatest)Ali   at(the)WBA

    is analyzed into the following lexical symbols and types:

            ":sysmail"              quoted string
            @                       special
            Some-Host               atom
            ,                       special
            Muhammed                atom
            (I am   the greatest)   comment
            Ali                     atom
            at                      atom
            (the)                   comment
            WBA                     atom

    The cononical representations for the data in these addresses
    are  the  following  strings  (note that there is exactly one
    SPACE between words):

                :sysmail at Some-Host

    and

                Muhammed Ali at WBA



2.  Formal Definitions

The first four rules, below, indicate a meta-syntax  for  fields,
without  regard to their particular type or internal syntax.  The
remaining rules define basic syntactic structures which are  used
by the rules in Sections III.C, III.D, and III.E.

field       =  field-name ":" [ field-body ] CRLF

field-name  =  fnatom *( LWSP-char [fnatom] )


Standard for the Format of Text Messages                        9
III. Syntax
  B. Lexical Analysis



fnatom      =  1*<any CHAR, excluding CTLs, SPACE, and ":">

field-body  =  field-body-contents
               [CRLF LWSP-char field-body]

field-body-contents = <the TELNET ASCII characters making up the
               field-body, as defined in the following sections,
               and consisting of combinations of atom, quoted-
               string, and specials tokens, or else consisting of
               texts>

                                            ; (  Octal, Decimal.)
CHAR        =  <any TELNET ASCII character> ; (  0-177,  0.-127.)
ALPHA       =  <any TELNET ASCII alphabetic character>
                                            ; (101-132, 65.- 90.)
                                            ; (141-172, 97.-122.)
DIGIT       =  <any TELNET ASCII digit>     ; ( 60- 71, 48.- 57.)
CTL         =  <any TELNET ASCII control    ; (  0- 37,  0.- 31.)
                character and DEL>          ; (    177,     127.)
CR          =  <TELNET ASCII carriage return>;(     15,      13.)
LF          =  <TELNET ASCII linefeed>      ; (     12,      10.)
SPACE       =  <TELNET ASCII space>         ; (     40,      32.)
HTAB        =  <TELNET ASCII horizontal-tab>; (     11,       9.)
<">         =  <TELNET ASCII quote mark>    ; (     42,      34.)
CRLF        =  CR LF

LWSP-char   =  SPACE / HTAB                 ; semantics = SPACE
linear-white-space =  1*([CRLF] LWSP-char)  ; semantics = SPACE
                                            ; CRLF => folding

specials    =  "(" / ")" / "<" / ">" / "@"  ; To use in a word,
            /  "," / ";" / ":" / "\" / <">  ;  word must be a
                                            ;  quoted-string.

delimiters  =  specials / comment / linear-white-space

text        =  <any CHAR, including bare    ; => atoms, specials,
                CR and/or bare LF, but NOT  ;  comments and
                including CRLF>             ;  quoted-strings are
                                            ;  NOT interpreted.

atom        =  1*<any CHAR except specials and CTLs>

quoted-string = <"> *(qtext/quoted-pair) <">; Any number of qtext
                                            ;   chars or any
                                            ;   quoted char.

qtext       =  <any CHAR excepting <">      ; => may be folded
                and CR, and including
                linear-white-space>


Standard for the Format of Text Messages                       10
III. Syntax
  B. Lexical Analysis



comment     =  "(" *(ctext / comment / quoted-pair) ")"
ctext       =  <any CHAR excluding "(",     ; => may be folded
                ")" and CR, and including
                linear-white-space>

quoted-pair =  "\" CHAR


3.  Clarifications

a.  "White space"

    Remember that in field-names  and  structured  field  bodies,
    MULTIPLE  LINEAR  WHITE SPACE TELNET ASCII CHARACTERS (namely
    HTABs and SPACEs) ARE TREATED AS SINGLE SPACES AND MAY FREELY
    SURROUND ANY SYMBOL.  In all header fields, the only place in
    which at least one space is REQUIRED is at the  beginning  of
    continuation  lines  in a folded field.  When passing text to
    processes which do  not  interpret  text  according  to  this
    standard  (e.g.,  ARPANET FTP mail servers), then exactly one
    SPACE should be used in place of arbitrary linear-white-space
    and comment sequences.

    WHEREVER A MEMBER OF THE LIST  OF  <DELIMITER>S  IS  ALLOWED,
    LWSP-CHARS MAY ALSO OCCUR BEFORE AND/OR AFTER IT.

    Writers of mail-sending  (i.e.  header  generating)  programs
    should  realize  that  there is no Network-wide definition of
    the effect of horizontal-tab TELNET ASCII characters  on  the
    appearance  of  text  at another Network host; therefore, the
    use  of  tabs  in  message  headers,  though  permitted,   is
    discouraged.

    Note that  during  transmissions  across  the  ARPANET  using
    TELNET  NVT  connections,  data  must  conform  to TELNET NVT
    conventions (e.g., CR must be followed by either LF, making a
    CRLF, or <null>, if the CR is to stand alone).

b.  Comments

    Comments are detected as such  only  within  field-bodies  of
    structured  fields.   A  comment  is  a  set  of TELNET ASCII
    characters, which is not within a quoted-string and which  is
    enclosed  in  matching parentheses; parentheses nest, so that
    if an unquoted left parenthesis occurs in a  comment  string,
    there  must  also  be  a  matching right parenthesis.  When a
    comment is used to act as the delimiter between a sequence of
    two  lexical  symbols,  such  as  two  atoms, it is lexically
    equivalent with one SPACE, for the purposes  of  regenerating
    the  sequence,  such as when passing the sequence onto an FTP
    mail server.


Standard for the Format of Text Messages                       11
III. Syntax
  B. Lexical Analysis



    In particular comments are NOT passed to the FTP  server,  as
    part  of  a MAIL or MLFL command, since comments are not part
    of the "formal" address.

    If a comment is to be "folded" onto multiple lines, then  the
    syntax for folding must be adhered to.  (See items III.B.1.a,
    above,  and  III.B.3.f,  below.)   Note  that  the   official
    semantics therefore do not "see" any unquoted CRLFs which are
    in comments, although particular parsing programs may wish to
    note  their  presence.   For  these  programs,  it  would  be
    reasonable to interpret a "CRLF LWSP-char" as  being  a  CRLF
    which  is part of the comment; i.e., the CRLF is kept and the
    LWSP-char is discarded.   Quoted  CRLFs  (i.e.,  a  backslash
    followed  by a CR followed by a LF) still must be followed by
    at least one LWSP-char.

c.  Delimiting and quoting characters

    The quote character (backslash) and characters which  delimit
    syntactic units are not, generally, to be taken as data which
    are part  of  the  delimited  or  quoted  unit(s).   The  one
    exception is SPACE.  In particular, the quotation-marks which
    define  a  quoted-string,  the  parentheses  which  define  a
    comment  and the backslash which quotes a following character
    are  NOT  part  of  the  quoted-string,  comment  or   quoted
    character.   A  quotation-mark  which  is  to  be  part  of a
    quoted-string, a parenthesis which is to be part of a comment
    and  a  backslash  which is to be part of either must each be
    preceded by the quote-character backslash ("\").   Note  that
    the  syntax  allows  any  character  to  be  quoted  within a
    quoted-string or comment;  however  only  certain  characters
    MUST  be quoted to be included as data.  These characters are
    those which are not part of the alternate text  group  (i.e.,
    ctext or qtext).

    A single SPACE is assumed to exist between  contiguous  words
    in  a  phrase,  and this interpretation is independent of the
    actual number of LWSP-chars which the creator places  between
    the  words.  To include more than one SPACE, the creator must
    make the LWSP-chars be part of a quoted-string.

    Quotation marks which delimit a quoted string and backslashes
    which  quote the following character should NOT accompany the
    quoted-string when the string is used with processes that  do
    not  interpret  data  according  to this specification (e.g.,
    ARPANET FTP mail servers).


Standard for the Format of Text Messages                       12
III. Syntax
  B. Lexical Analysis



d.  Quoted-strings

    Where   permitted  (i.e.,  in  words  in  structured  fields)
    quoted-strings   are   treated   as  a  single  symbol  (i.e.
    equivalent to an atom, syntactically).  If a quoted-string is
    to  be  "folded"  onto  multiple  lines,  then the syntax for
    folding must be adhered to.  (See items III.B.1.a, above, and
    III.B.3.f,   below.)    Note   that  the  official  semantics
    therefore do not "see" any bare CRLFs which  are  in  quoted-
    strings,  although  particular  parsing  programs may wish to
    note  their  presence.   For  these  programs,  it  would  be
    reasonable  to  interpret  a "CRLF LWSP-char" as being a CRLF
    which is part of the quoted-string; i.e., the  CRLF  is  kept
    and  the  LWSP-char  is  discarded.   Quoted  CRLFs  (i.e., a
    backslash followed by a CR followed by a LF) are also subject
    to  rules  of  folding,  but  the  presence  of  the  quoting
    character (backslash) explicitly indicates that the  CRLF  is
    data to the quoted string.  Stripping off the first following
    LWSP-char is also appropriate when parsing quoted CRLFs.

e.  Bracketing characters

    There are three types of brackets which must be well nested:

        o  Parentheses are used to indicate comments.

        o  Angle brackets ("<" and ">") are  generally  used
           to indicate the presence of at least one machine-
           usable code (e.g., delimiting mailboxes).

        o  Colon/semi-colon  (":"  and  ";")  are   used  in
           address   specifications  to  indicate  that  the
           included list of addresses are to be treated as a
           group.

f.  Case independence of certain specials atoms

    Certain atoms, which are represented in the syntax as literal
    alphabetic  strings, can be represented in any combination of
    upper and lower case.  These are:

        -  field-name,
        -  "Include", "Postal" and equivalent atoms in a
           ":"<atom>":" address specification,
        -  "at", in a host-indicator,
        -  node,
        -  day-of-week,
        -  month, and
        -  zones.

    When matching an atom against one of these literals, case  is
    to  be ignored.  For example, the field-names "From", "FROM",

Standard for the Format of Text Messages                       13
III. Syntax
  B. Lexical Analysis



    "from", and even "FroM" should all  be  treated  identically.
    However,  the  case  shown in this specification is suggested
    for message-creating processes.  Note that, at the  level  of
    this  specification,  case  IS  relevant  to  other words and
    texts.  Also see Section IV.A.1.f, below.

g.  Folding long lines

    Each header item (field of the message) may be represented on
    exactly  one line consisting of the name of the field and its
    body; this is what the parser sees.  For readability,  it  is
    recommended  that the field-body portion of long header items
    be "folded" onto multiple lines of the actual header.  "Long"
    is  commonly  interpreted  to  mean  greater  than  65  or 72
    characters.  The former length is recommended as a limit, but
    it is not imposed by this standard.

h.  Backspace characters

    Backspace TELNET ASCII characters (ASCII BS, decimal 8.)  may
    be   included   in   texts   and   quoted-strings  to  effect
    overstriking; however, any use of backspaces which effects an
    overstrike  to  the  left  of  the  beginning  of the text or
    quoted-string is prohibited.



C.  GENERAL SYNTAX OF MESSAGES:




    NOTE:  Due to an artifact of the notational conventions,
           the  syntax indicates that, when present, "Date",
           "From", "Sender", and "Reply-To" fields  must  be
           in  a  particular order.  These header items must
           be unique (occur exactly once).   However  header
           fields, in fact, are NOT required to occur in any
           particular order, except that  the  message  body
           must  occur  AFTER  the headers.  For readability
           and ease of parsing  by  simple  systems,  it  is
           recommended  that  headers  be  sent in the order
           "Date", "From", "Subject", "Sender", "To",  "cc",
           etc.    This   specification   permits   multiple
           occurrences of  most  optional-fields.   However,
           their  interpretation  is not specified here, and
           their use is strongly discouraged.

The following syntax for the bodies of various fields  should  be
thought  of as describing each field body as a single long string
(or line).   The  section  on  Lexical  Analysis  (section  II.B)
indicates how such long strings can be represented on more than
one line in the actual transmitted message.


Standard for the Format of Text Messages                       14
III. Syntax
  C. Messages



message     =  fields *( CRLF *text )       ; Everything after
                                            ;  first null line
                                            ;  is message body

fields      =  date-field                   ; Creation time-stamp
               originator-fields            ;  & author id are
               *optional-field              ;  required: others
                                            ;  are all optional

originator-fields =
               (  "From"     ":" mailbox    ; Single author
                 ["Reply-To" ":" #address] )
            /  (  "From"     ":" 1#address  ; Multiple authors &
                  "Sender"   ":" mailbox    ;  may have non-mach-
                 ["Reply-To" ":" #address] );  ine addresses

date-field  =  "Date"       ":" date-time

optional-field  =
               "To"         ":" #address
            /  "cc"         ":" #address
            /  "bcc"        ":" #address    ; Blind carbon
            /  "Subject"    ":" *text
            /  "Comments"   ":" *text
            /  "Message-ID" ":" mach-id     ; Only one allowed
            /  "In-Reply-To"":" #(phrase / mach-id)
            /  "References" ":" #(phrase / mach-id)
            /  "Keywords"   ":" #phrase
            /  extension-field              ; To be defined in
                                            ;  supplemental
                                            ;  specifications
            /  user-defined-field           ; Must have unique
                                            ;  field-name & may
                                            ;  be pre-empted

extension-field = <Any field which is defined in a document
               published as a formal extension to this
               specification>

user-defined-field = <Any field which has not been defined in
               this specification or published as an extension to
               this specification; names for such fields must be
               unique and may be preempted by published
               extensions>




Standard for the Format of Text Messages                       15
III. Syntax
  D. Addressee Items



D.  SYNTAX OF GENERAL ADDRESSEE ITEMS


address     =  host-phrase                  ; Machine mailbox
            / ( [phrase] "<" #address ">")  ; Individual / List
            / ( [phrase] ":" #address ";")  ; Group
            /  quoted-string                ; Arbitrary text
            / (":" ( "Include"              ; File, w/ addr list
                   / "Postal"               ; (U.S.) Postal addr
                   /  atom )                ; Extended data type
               ":" address)

mailbox     =  host-phrase /  (phrase mach-id)

mach-id     =  "<" host-phrase ">"          ; Contents must never
                                            ;  be modified!



E.  SUPPORTING CONSTRUCTS




host-phrase =  phrase  host-indicator       ; Basic address

host-indicator =  1*( ("at" / "@") node )   ; Right-most node is
                                            ;  at top of network
                                            ;  hierarchy; left-
                                            ;  most must be host

node        =  word / 1*DIGIT               ; Official host or
                                            ;  network name or
                                            ;  decimal address


date-time   =  [ day-of-week "," ] date time

day-of-week =  "Monday"    / "Mon"  / "Tuesday"   / "Tue"
            /  "Wednesday" / "Wed"  / "Thursday"  / "Thu"
            /  "Friday"    / "Fri"  / "Saturday"  / "Sat"
            /  "Sunday"    / "Sun"

date        =  1*2DIGIT ["-"] month         ; day month year
               ["-"] (2DIGIT /4DIGIT)       ;  e.g. 20 Aug [19]77

month       =  "January"   / "Jan"  / "February"  / "Feb"
            /  "March"     / "Mar"  / "April"     / "Apr"
            /  "May"                / "June"      / "Jun"
            /  "July"      / "Jul"  / "August"    / "Aug"
            /  "September" / "Sep"  / "October"   / "Oct"
            /  "November"  / "Nov"  / "December"  / "Dec"


Standard for the Format of Text Messages                       16
III. Syntax
  E. Supporting Constructs



time        =  hour zone                    ; ANSI and Military
                                            ;  (seconds optional)

hour        =  2DIGIT [":"] 2DIGIT [ [":"] 2DIGIT ]
                                            ; 0000[00] - 2359[59]

zone        = ( ["-"] ( "GMT"               ; Relative to GMT:
                                            ; North American
                 /  "NST" /                 ;  Newfoundland:-3:30
                 /  "AST" / "ADT"           ;  Atlantic: - 4/ - 3
                 /  "EST" / "EDT"           ;  Eastern:  - 5/ - 4
                 /  "CST" / "CDT"           ;  Central:  - 6/ - 5
                 /  "MST" / "MDT"           ;  Mountain: - 7/ - 6
                 /  "PST" / "PDT"           ;  Pacific:  - 8/ - 7
                 /  "YST" / "YDT"           ;  Yukon:    - 9/ - 8
                 /  "HST" / "HDT"           ;  Haw/Ala   -10/ - 9
                 /  "BST" / "BDT"           ;  Bering:   -11/ -10
                    1ALPHA       ))         ; Military: Z = GMT;
                                            ;  A:-1; (J not used)
                                            ;  M:-12; N:+1; Y:+12
            / ( ("+" / "-") 4DIGIT )        ; Local differential
                                            ;  hours/min. (HHMM)

phrase      =  1*word                       ; Sequence of words.
                                            ;  Separation seman-
                                            ;  tically = SPACE

word        =  atom / quoted-string



Standard for the Format of Text Messages                       17
IV. Semantics
 A. Address Fields





                         IV.  SEMANTICS



A.  ADDRESS FIELDS


1.  General

a.  The phrase part of a host-phrase in an address  specification
    (i.e.,  the  host's name for the mailbox) is understood to be
    whatever the receiving FTP Server allows (for example,  TENEX
    systems  do  not  now understand addresses of the form "P. D.
    Q. Bach", but another system might).

    Note that a mailbox is a conceptual  entity  which  does  not
    necessarily pertain to file storage.  For example, some sites
    may choose to print mail on their line  printer  and  deliver
    the output to the addressee's desk.

    An individual may have  several  mailboxes  and  a  group  of
    individuals  may wish to receive mail as a single unit (i.e.,
    a distribution list).  The second and third  alternatives  of
    an  address  list  (#address)  allow  naming  a collection of
    subordinate  addresses  list(s).   Recipient  mailboxes   are
    specified  within the bracketed part ("<" - ">" or ":" - ";")
    of such named lists.  The use of angle-brackets ("<", ">") is
    intended for the cases of individuals with multiple mailboxes
    and of special mailbox lists; it is not expected to be nested
    more  than  one level, although the specification allows such
    nesting.  The use of colon/semi-colon (":", ";") is  intended
    for  the  case  of  groups.   Groups  can be expected to nest
    (i.e., to  contain  subgroups).   For  both  individuals  and
    groups,  a  copy  of the transmitted message is to be sent to
    EACH mailbox  listed.   For  the  case  of  a  special  list,
    treatment of addresses is defined in the relevant subsections
    of this section.

b.  The inclusion of bare quoted-strings as addresses (i.e.,  the
    fourth  address-form  alternative)  is allowed as a syntactic
    convenience, but no semantics  are  defined  for  their  use.
    However,  it is reasonable, when replicating an address list,
    to replicate ALL of its members, including quoted-strings.

c.  ":Include:" specifications are used to refer to one  or  more
    locations  containing  stored  address  lists (#address).  If
    more than one location is referenced, the address part of the
    Include  phrase  must  be  a  list  (#address)  surrounded by
    angle-brackets, as per the "Individual / List" alternative of
    <address>.   Constituent  addresses  must  resolve to a host-

Standard for the Format of Text Messages                       18
IV. Semantics
 A. Address Fields



    phrase; only they have any  meaning  within  this  construct.
    The phrase part of indicated host-phrases should contain text
    which the referenced  host  can  resolve  to  a  file.   This
    standard is not a protocol and so does not prescribe HOW data
    is to be retrieved from the  file.   However,  the  following
    requirements are made:

         o  The file must be accessible  through  the  local
            operating system interface (if it exists), given
            adequate user access rights; and

         o  If a host has an FTP server and a user  is  able
            to  retrieve  any files from the host using that
            server, then the file must be accessible through
            FTP,  using  DEFAULT  transfer  settings,  given
            adequate user access rights.

    It is intended that this mechanism allow programs to retrieve
    such lists automatically.

    The interpretation of such a file reference follows.  This is
    not  intended  to imply any particular implementation scheme,
    but is presented  to  aid  in  understanding  the  notion  of
    including  file  contents in address lists:

         o  Elements of the address list part are alternates
            and  the  contents of ONLY ONE of them are to be
            included in the resultant address list.

         o  The contents of the file indicated by  a  member
            host-phrase  are  treated as an address list and
            are inserted as an address  list  (#address)  in
            the  position  of  the  path item in the syntax.
            That is, the TELNET ASCII characters  specifying
            the  entire Include <address> is replaced by the
            contents of one of the files to which the  host-
            phrase(s),   of  the  address  list  (#address),
            refers.  Therefore, the contents of  each  file,
            indicated   by   an  Include  address,  must  be
            syntactically self-contained and must adhere  to
            the full syntax prescribed herein for an address
            list.

d.  ":Postal:" specifications are used to indicate (U.S.)  postal
    addresses,  but  can  be  treated  the  same as quoted-string
    addresses.  To reference a list of postal addresses, the list
    must  conform  to  the  "Individual  /  List"  alternative of
    <address>.  The ":Include:" alternative also is valid.

e.  The "':' atom ':'" syntax is intended as a general  mechanism
    for  indicating  specially  data-typed  addresses.   As  with
    extension-fields, the authors of this document will  regulate

Standard for the Format of Text Messages                       19
IV. Semantics
 A. Address Fields



    the  publishing  of  specifications  for these extended data-
    types.  In the absence of defined semantics,  any  occurrence
    of  an address in this form may be treated as a quoted-string
    address.

f.  A node name must be THE official name of a network or a host,
    or  else  a decimal number indicating the Network address for
    that network or host, at the time  the  message  is  created.
    The  USE  OF NUMBERS IS STRONGLY DISCOURAGED and is permitted
    only due to the occasional necessity of bypassing local  name
    tables.   For  the  ARPANET, official names are maintained by
    the Network Information Center at  SRI  International,  Menlo
    Park, California.

    Whenever a message might be transmitted or migrate to a  host
    on  another  network,  full  hierarchical  addresses  must be
    specified.   These  are  indicated  as  a  series  of  words,
    separated  by at-sign or "at" indications.  The communication
    environment is assumed to consist of a collection of networks
    organized  as  independent  "trees"  except  for  connections
    between the root nodes.  That is, only the roots can  act  as
    gateways  between  these  independent  networks.  While other
    actual connections may exist, it is believed  that  presuming
    this  type of organization will provide a reliable method for
    describing VALID, if not EFFICIENT, paths between  hosts.   A
    typical full mailbox specification might therefore look like:

         Friendly User @ hosta @ local-net1 @ major-netq

    In the simplest case, a mail-sending host should transmit the
    message  to the node which is mentioned last (farthest to the
    right), strip off that node reference from the specification,
    and then pass the remaining host-phrase to the recipient host
    (in  the  ARPANET,  its  FTP server) for it to process.  This
    treats the remaining portion of the host-indicator merely  as
    the terminating part of the phrase.

         NOTE:  When passing any portion of a host-indicator
                onto a process which does not interpret data
                according to this  standard  (e.g.,  ARPANET
                FTP  servers), "@" must be used and not "at"
                and it must not be preceded or  followed  by
                any  LWSP-chars.   Using  the above example,
                the following string would be passed to  the
                major-netq gateway:

                Friendly User@hosta@local-net1

    When the sending host  has  more  knowledge  of  the  network
    environment,  then  it  should  send the message along a more
    efficient path, making appropriate changes to the form of the
    host-phrase which it gives to the recipient host.

Standard for the Format of Text Messages                       20
IV. Semantics
 A. Address Fields



    To use the above specification as an example:  If  a  sending
    hostb  also were part of local-net1, then it could  send  the
    message  directly  to  hosta  and  would give only the phrase
    "Friendly User" to hosta's mail-receiving program.  If  hostb
    were  part  of  local-net2, along with hostc, and happened to
    know that hosta and hostc were  part  of  another  local-net,
    then  hostb  could  send  the message to hostc to the address
    "Friendly User@hosta".

    The phrase in a host-phrase is intended to be meaningful only
    to  the  indicated  receiving  host.  To all other hosts, the
    phrase is to be treated as an uninterpreted string.  No  case
    transformations  should  be  (automatically) performed on the
    phrase.  The phrase  is  passed  to  the  local  host's  mail
    sending  program; it is the responsibility of the destination
    host's mail receiving (distribution) program to perform  case
    mapping on this phrase, if required, to deliver the mail.


2.  Originator Fields

    WARNING:  The standard  allows  only  a  subset  of  the
              combinations  possible  with the From, Sender,
              and  Reply-To  fields.   The   limitation   is
              intentional.

a.  From

    This field contains the identity of the person(s) who  wished
    this message to be sent.  The message-creation process should
    default this field to be a single machine address, indicating
    the AGENT (person or process) entering the message.  If  this
    is  NOT  done, the "Sender" field MUST be present; if this IS
    done, the "Sender" field is optional.

b.  Sender

    This field contains  the  identity  of  the  AGENT (person or
    process) who  sends the message.  It is intended for use when
    the sender is not the author of the message, or  to  indicate
    who  among  a group of authors actually sent the message.  If
    the contents  of  the  "Sender"  field  would  be  completely
    redundant with the "From" field, then the "Sender" field need
    not be present and  its  use  is  discouraged  (though  still
    legal);  in  particular,  the  "Sender" field MUST be present
    if it is NOT the same as the "From" Field.

    The  Sender  host-phrase  includes  a   phrase   which   must
    correspond  to  a  specific  agent  (i.e.,  a human user or a
    computer program)  rather  than  a  standard  address.   This
    indicates  the  expectation  that the field will identify the
    single AGENT (person or process) responsible for sending  the

Standard for the Format of Text Messages                       21
IV. Semantics
 A. Address Fields



    mail  and not simply include the name of a mailbox from which
    the mail was sent.  For example in the case of a shared login
    name, the name, by itself, would not be adequate.  The phrase
    part of the host-phrase,  which  refers  to  this  agent,  is
    expected  to be a computer system term, and not (for example)
    a generalized person reference which can be used outside  the
    network text message context.

    Since the critical function served by the "Sender"  field  is
    the  identification of the agent responsible for sending mail
    and since computer programs cannot be  held  accountable  for
    their  behavior, is strongly recommended that when a computer
    program generates a message, the HUMAN who is responsible for
    that  program  be  referenced  as  part of the "Sender" field
    host-phrase.

c.  Reply-To

    This field provides a general mechanism  for  indicating  any
    mailbox(es) to which responses are to be sent.  Three typical
    uses for this feature can be  distinguished.   In  the  first
    case,  the  author(s)  may  not  have  regular  machine-based
    mailboxes and therefore wish(es)  to  indicate  an  alternate
    machine  address.   In  the  second  case, an author may wish
    additional persons to be made aware of, or  responsible  for,
    responses;  responders  should  send  their  replies  to  the
    "Reply-To" mailbox(es) listed in  the  original  message.   A
    somewhat  different  use may be of some help to "text message
    teleconferencing" groups equipped with automatic distribution
    services:   include  the  address  of  that  service  in  the
    "Reply-To"  field  of   all   messages   submitted   to   the
    teleconference;  then  participants can "reply" to conference
    submissions to guarantee  the  correct  distribution  of  any
    submission of their own.

    Reply-To fields are  NOT  required  to  contain  any  machine
    addresses  (i.e., host-phrases).   Note,  however,  that  the
    absence  of even one  valid  network  address  will  tend  to
    prevent  software  systems from automatically assisting users
    in conveniently responding to mail.

NOTE:  For systems which automatically generate address lists for
replies to messages, the following recommendations are made:

     o  The receiver, when replying  to  a  message,  should
        NEVER automatically include the "Sender" host-phrase
        in the reply's address list;

     o  If the  "Reply-To"  field  exists,  then  the  reply
        should  go  ONLY  to the addresses indicated in that
        field and not to  the  addresses  indicated  in  the
        "From" field.

Standard for the Format of Text Messages                       22
IV. Semantics
 A. Address Fields



(Extensive    examples  are  provided  in   Section   V.)    This
recommendation  is intended only for originator-fields and is not
intended to suggest that replies should not also be sent  to  the
other  recipients  of  this  message.  It is up to the respective
mail handling programs to decide what additional facilities  will
be provided.


3.  Receiver Fields

a.  To

    This field contains the identity of the primary recipients of
    the message.

b.  cc

    This field contains the identity of the secondary  recipients
    of the message.

b.  Bcc

    This field contains the identity of additional recipients  of
    the  message.  The contents of this field are not included in
    copies of the message  sent  to  the  primary  and  secondary
    recipients.   Some  systems may choose to include the text of
    the "Bcc" field only in the author(s)'s  copy,  while  others
    may  also  include it in the text sent to all those indicated
    in the "Bcc" list.



B.  REFERENCE SPECIFICATION FIELDS


1.  Message-ID

This field contains a unique identifier (the phrase) which refers
to  THIS  version of THIS message.  The uniqueness of the message
identifier is guaranteed by the host which  generates  it.   This
identifier is intended to be machine readable and not necessarily
meaningful to humans.  A message identifier pertains  to  exactly
one  instantiation  of a particular message; subsequent revisions
to the message should each receive a new message identifier.


2.  In-Reply-To

The contents of this field identify previous correspondence which
this  message answers.  Note that if message identifiers are used
in this field, they must use the mach-id specification format.


Standard for the Format of Text Messages                       23
IV. Semantics
 B. Reference Specification Fields



3.  References

The contents of this field identify  other  correspondence  which
this  message  references.   Note  that  if  message  identifiers
are used, they  must  use  the  mach-id  specification format.


4.  Keywords

This field contains keywords or phrases, separated by commas.



C.  OTHER FIELDS AND SYNTACTIC ITEMS


1.  Subject

The "Subject" field is intended to provide as much information as
necessary  to  adequately summarize or indicate the nature of the
message.


2.  Comments

Permits adding text comments onto the message without  disturbing
the contents of the message's body.


3.  Extension-field

A relatively limited number of common fields have been defined in
this  document.  As network mail requirements dictate, additional
fields may be standardized.  The authors of  this  document  will
regulate  the publishing of such definitions as extensions to the
basic specification.


4.  User-defined-field

Individual users of network mail  are  free  to  define  and  use
additional  header fields.  Such fields must have names which are
not  already  used  in  the  current  specification  or  in   any
definitions  of extension-fields, and the overall syntax of these
user-defined-fields must conform to  this  specification's  rules
for  delimiting and  folding  fields.  Due to the extension-field
publishing process, the name of a user-defined-field may be  pre-
empted.




Standard for the Format of Text Messages                       24
IV. Semantics
 D. Dates



D.  DATES AND TIMES

If included, day-of-week must be the  day  implied  by  the  date
specification.

Time zone  may  be  indicated  in  several  ways.   The  military
standard   uses  a  single  character  for  each  zone.   "Z"  is
Greenwhich Mean Time; "A" indicates one  hour  earlier,  and  "M"
indicates  12 hours earlier; "N" is one hour later, and "Y" is 12
hours later.  The letter "J" is not used.   The  other  remaining
two  forms  are  taken from ANSI standard X3.51-1975.  One allows
explicit indication of the amount of offset from GMT;  the  other
uses  common  3-character  strings  for  indicating time zones in
North America.


Standard for the Format of Text Messages                       25
V. Examples
A. Addresses





                          V.  EXAMPLES




A.  ADDRESSES


1.  Alfred E. Neuman <Neuman at BBN-TENEXA>

2.  Neuman@BBN-TENEXA

These two "Alfred E. Neuman" examples have  identical  semantics,
as  far  as  the  operation  of  the  local  host's  mail sending
(distribution) program (also sometimes called its  "mailer")  and
the  remote  host's  FTP  server  are  concerned.   In  the first
example, the "Alfred E. Neuman" is  ignored  by  the  mailer,  as
"Neuman  at  BBN-TENEXA" completely specifies the recipient.  The
second example contains no superfluous information,  and,  again,
"Neuman@BBN-TENEXA" is the intended recipient.


3.  Al Neuman at BBN-TENEXA

This is identical to "Al Neuman <Al Neuman at BBN-TENEXA>".  That
is,  the  full  phrase, "Al Neuman", is passed to the FTP server.
Note that not all FTP servers accept multi-word identifiers;  and
some  that  do  accept  them  will treat each word as a different
addressee (in this case, attempting to send a copy of the message
to "Al" and a copy to "Neuman").


4.  "George Lovell, Ted Hackle" <Shared-Mailbox at Office-1>

This form might be used to indicate  that  a  single  mailbox  is
shared  by  several  users.   The quoted string is ignored by the
originating  host's  mailer,  as  "Shared-Mailbox  at   Office-1"
completely specifies the destination mailbox.


4.  Wilt (the Stilt) Chamberlain at NBA

The "(the Stilt)" is a comment, which  is  NOT  included  in  the
destination  mailbox  address  handed to the originating system's
mailer.  The  address  is  the  string  "Wilt Chamberlain",  with
exactly  one  space  between  the  first  and second words.  (The
quotation marks are not included.)




Standard for the Format of Text Messages                       26
V. Examples
B. Address Lists



B.  ADDRESS LISTS

    Gourmets:  Pompous Person <WhoZiWhatZit at Cordon-Bleu>,
               Cooks:  Childs at WGBH, Galloping Gourmet at
                       ANT (Australian National Television);,
               Wine Lovers:  Cheapie at Discount-Liquors,
                             Port at Portugal;;,
    Jones at SEA

This group list example points  out  the  use  of  comments,  the
nesting  of groups, and the mixing of addresses and groups.  Note
that the two consecutive semi-colons  preceding  "Jones  at  SEA"
mean that Jones is NOT a member of the Gourmets group.


C.  ORIGINATOR ITEMS


1.  Author-sent

George Jones logs into  his  Host  as  "Jones".   He  sends  mail
himself.

    From:  Jones at Host
or
    From:  George Jones <Jones at Host>


2.  Secretary-sent

George Jones logs in as Jones on his Host.   His  secretary,  who
logs in as Secy on Shost sends mail for him.  Replies to the mail
should go to George, of course.

    From:    George Jones <Jones at Host>
    Sender:  Secy at SHost


3.  Shared directory or unrepresentative directory-name

George Jones logs in as Group at Host.  He  sends  mail  himself;
replies should go to the Group mailbox.

    From:  George Jones <Group at Host>



Standard for the Format of Text Messages                       27
V. Examples
C. Originator Items



4.  Secretary-sent, for user of shared directory

George Jones' secretary sends mail for George in his capacity  as
a  member  of  Group  while  logged  in as Secy at Host.  Replies
should go to Group.

    From:   George Jones<Group at Host>
    Sender: Secy at Host

Note that there need not be a space between "Jones" and the  "<",
but  adding a space enhances readability (as is the case in other
examples).


5.  Secretary acting as full agent of author

George Jones asks his secretary (Secy at Host) to send a  message
for  him  in  his  capacity  as Group.  He wants his secretary to
handle all replies.

    From:     George Jones <Group at Host>
    Sender:   Secy at Host
    Reply-To: Secy at Host


6.  Agent for user without online mailbox

A  non-ARPANET  user  friend  of  George's,  Sarah,  is  visting.
George's  secretary  sends  some  mail  to  a  friend of Sarah in
computer-land.  Replies should go to  George,  whose  mailbox  is
Jones at Host.

    From:     Sarah Friendly
    Sender:   Secy at Host
    Reply-To: Jones at Host


7.  Sent by member of a committee

George is a member of a committee.  He wishes to have any replies
to his message go to all committee members.

    From:     George Jones
    Sender:   Jones at Host
    Reply-To: Big-committee: Jones at Host,
                             Smith at Other-Host,
                             Doe at Somewhere-Else;

Note that if George had not included himself in  the  enumeration
of Big-committee, he would not have gotten an implicit reply; the
presence  of  the  "Reply-to"  field  SUPERSEDES the sending of a
reply to the person named in the "From" field.

Standard for the Format of Text Messages                       28
V. Examples
C. Originator Items



8.  Example of INCORRECT use



George desires a reply to go  to  his  secretary;  therefore  his
secretary  leaves  his  mailbox  address  off  the  "From" field,
leaving only his name, which is not, itself, a mailbox address.

         From:   George Jones
         Sender: Secy at SHost

THIS IS NOT PERMITTED.  Replies are NEVER implicitly sent to  the
"Sender";  George's  secretary  should  have  used the "Reply-To"
field, or the  mail  creating  program  should  have  forced  the
secretary to.

9.  Agent for member of a committee



George's secretary sends out a message which was authored jointly
by all the members of the "Big-committee".

         From:   Big-committee: Jones at Host,
                                Smith at Other-Host,
                                Doe at Somewhere-Else;
         Sender: Secy at SHost



D.  COMPLETE HEADERS


1.  Minimum required:

       Date:  26 August 1976 1429-EDT
       From:  Jones at Host


2.  Using some of the additional fields:

       Date: 26 August 1976 1430-EDT
       From:George Jones<Group at Host>
       Sender:Secy at SHOST
       To:Al Neuman at Mad-Host,
                Sam Irving at Other-Host
       Message-ID:  <some string at SHOST>



Standard for the Format of Text Messages                       29
V. Examples
D.  Complete Headers



3.  About as complex as you're going to get:

       Date     :  27 Aug 1976 0932-PDT
       From     :  Ken Davis <KDavis at Other-Host>
       Subject  :  Re: The Syntax in the RFC
       Sender   :  KSecy at Other-Host
       Reply-To :  Sam Irving at Other-Host
       To       :  George Jones <Group at Host>,
                   Al Neuman at Mad-Host
       cc       :  Important folk:
                   Tom Softwood <Balsa at Another-Host>,
                   Sam Irving at Other-Host;,
                   Standard Distribution::Include:
                    </main/davis/people/standard at Other-Host,
                     "<Jones>standard.dist.3" at Tops-20-Host>,
                   (The following Included Postal list is part
                   of Standard Distribution.)
                   :Postal::Include: Non-net-addrs@Other-host;,
                   :Postal: "Sam Irving, P.O. Box 001, Las Vegas,
                             Nevada"  (So that he can stay
                             apprised of the situation)
       Comment  :  Sam is away on business. He asked me to handle
                   his mail for him.  He'll be able to provide  a
                   more  accurate  explanation  when  he  returns
                   next week.
       In-Reply-To: <some string at SHOST>
       Special (action):  This is a sample of multi-word field-
                   names, using a range of characters.  There
                   could also be a field-name "Special (info)".
       Message-ID: <4231.629.XYzi-What at Other-Host>


Standard for the Format of Text Messages                       31
Appendix
A. Alphabetical Listing of Syntax Rules





                        APPENDIX



A.  ALPHABETICAL LISTING OF SYNTAX RULES


address     =  host-phrase / quoted-string
            / (*phrase "<" #address ">" )
            / (*phrase ":" #address ";" )
            / (":" ("Include" / "Postal" / atom) ":" address)
ALPHA       =  <any TELNET ASCII alphabetic character>
atom        =  1*<any CHAR except specials and CTLs>

CHAR        =  <any TELNET ASCII character>
comment     =  "(" *(ctext / comment / quoted-pair) ")"
CR          =  <TELNET ASCII carriage return>
CRLF        =  CR LF
ctext       =  <any CHAR excluding "(", ")", CR, LF and
               including linear-white-space>
CTL         =  <any TELNET ASCII control character and DEL>

date        =  1*2DIGIT ["-"] month ["-"] (2DIGIT /4DIGIT)
date-field  =  "Date"       ":" date-time
date-time   =  [ day-of-week "," ] date time
day-of-week =  "Monday"    / "Mon"  / "Tuesday"   / "Tue"
            /  "Wednesday" / "Wed"  / "Thursday"  / "Thu"
            /  "Friday"    / "Fri"  / "Saturday"  / "Sat"
            /  "Sunday"    / "Sun"
delimiters  =  specials / comment / linear-white-space
DIGIT       =  <any TELNET ASCII digit>

extension-field = <Any field which is defined in a document
               published as a formal extension to this
               specification>

field       =  field-name ":" [ field-body ] CRLF

fields      =  date-field  originator-fields  *optional-field
field-body  =  field-body-contents
               [CRLF LWSP-char field-body]
field-body-contents = <the TELNET ASCII characters making up the
               field-body, as defined in the following sections,
               and consisting of combinations of atom, quoted-
               string, and specials tokens, or else consisting of
               texts>
field-name  =  fnatom *(LWSP-char [fnatom])
fnatom      =  1*<any CHAR, excluding CTLs, SPACE, and ":">


Standard for the Format of Text Messages                       32
Appendix
A. Alphabetical Listing of Syntax Rules



host-indicator =  1*( ("at" / "@") node )
host-phrase =  phrase  host-indicator
hour        =  2DIGIT [":"] 2DIGIT [ [":"] 2DIGIT ]
HTAB        =  <TELNET ASCII horizontal-tab>

LF          =  <TELNET ASCII linefeed>
linear-white-space =  1*([CRLF] LWSP-char)
LWSP-char   = SPACE / HTAB

mach-id     =  "<" host-phrase ">"
mailbox     =  host-phrase /  (phrase mach-id)
message     =  fields *(CRLF *text)
month       =  "January"   / "Jan"  / "February"  / "Feb"
            /  "March"     / "Mar"  / "April"     / "Apr"
            /  "May"                / "June"      / "Jun"
            /  "July"      / "Jul"  / "August"    / "Aug"
            /  "September" / "Sep"  / "October"   / "Oct"
            /  "November"  / "Nov"  / "December"  / "Dec"

node        =  word / 1*DIGIT

optional-field  =
               "To"         ":" #address
            /  "cc"         ":" #address
            /  "bcc"        ":" #address
            /  "Subject"    ":" *text
            /  "Comments"   ":" *text
            /  "Message-ID" ":" mach-id
            /  "In-Reply-To"":" #(phrase / mach-id)
            /  "References" ":" #(phrase / mach-id)
            /  "Keywords"   ":" #phrase
            /  extension-field
            /  user-defined-field
originator-fields =
               (  "From"     ":" mailbox
                 ["Reply-To" ":" #address] )
            /  (  "From"     ":" 1#address
                  "Sender"   ":" mailbox
                 ["Reply-To" ":" #address] )

phrase      =  1*word

quoted-pair =  "\" CHAR
quoted-string =  <">  *(qtext / quoted-pair)  <">
qtext       =  <any CHAR except <">, CR, or LF and including
               linear-white-space>
SPACE       =  <TELNET ASCII space>
specials    =  "(" / ")" / "<" / ">" / "@"/ "," / ";" / ":"
            /  "\" / <">

text        =  <any CHAR, including bare CR and/or bare LF, but
               NOT including CRLF>

Standard for the Format of Text Messages                       33
Appendix
A. Alphabetical Listing of Syntax Rules



time        =  hour zone

user-defined-field = <Any field which has not been defined in
               this specification or published as an extension to
               this specification; names for such fields must be
               unique and may be preempted by putlished
               extensions>

word        =  atom / quoted-string

zone        = ( ("+" / "-") 4DIGIT )
            / ( ["-"] (1ALPHA
              / "GMT" / "NST"  / "AST" / "ADT" / "EST" / "EDT"
              / "CST" / "CDT"  / "MST" / "MDT" / "PST" / "PDT"
              / "YST" / "YDT"  / "HST" / "HDT" / "BST" / "BDT" ))

<">         =  <TELNET ASCII quote mark>


Standard for the Format of Text Messages                       35
Appendix
B. Simple Parsing




B.  SIMPLE PARSING


     Some mail-reading software systems may wish to perform  only
minimal  processing,  ignoring  the internal syntax of structured
field-bodies and treating them the  same  as  unstructured-field-
bodies.  Such software will need only to distinguish:

     -  Header fields from the message body,
     -  Beginnings of fields from lines which continue fields,
     -  Field-names from field-contents.

     The abbreviated set of syntactic rules  which  follows  will
suffice  for  this  purpose.   They  describe  a  limited view of
messages and are a subset of the syntactic rules provided in  the
main part of this specification.  One small exception is that the
contents of field-bodies consist only of text:


SYNTAX:

message         =  *field *(CRLF *text)

field           =  field-name ":" [field-body] CRLF

field-name      =  fnatom *( LWSP-char [fnatom] )

fnatom          =  1*<any CHAR, excluding CTLs, SPACE, and ":">


field-body      =  *text [CRLF LWSP-char field-body]


SEMANTICS:

     Headers occur before the message body and are terminated  by
a null line (i.e., two contiguous CRLFs).

     A line which continues a header field begins with a SPACE or
HTAB  character,  while  a  line  beginning a field starts with a
printable character which is not a colon.

     A field-name consists of one or  more  printable  characters
(excluding colon), each separated by one or more SPACES or HTABS.
A field-name MUST be contained on one line.  Upper and lower case
are not distinguished when comparing field-names.


Standard for the Format of Text Messages                       37
Bibliography






                          BIBLIOGRAPHY


ANSI.   Representations   of   universal   time,    local    time
   differentials,  and  United  States  time  zone references for
   information interchange.  ANSI X3.51-1975;  American  National
   Standards Institute:  New York, 1975.

Bhushan, A.K.  The File Transfer Protocol.  ARPANET  Request  for
   Comments,  No.   354,  Network  Information Center No.  10596;
   Augmentation Research  Center,  Stanford  Research  Institute:
   Menlo Park, July 1972.

Bhushan, A.K.  Comments on the File Transfer  Protocol.   ARPANET
   Request for Comments, No.  385, Network Information Center No.
   11357;  Augmentation  Research   Center,   Stanford   Research
   Institute:  Menlo Park, August 1972.

Bhushan, A.K., Pogran, K.T., Tomlinson,  R.S.,  and  White,  J.E.
   Standardizing  Network  Mail  Headers.   ARPANET  Request  for
   Comments, No.  561,  Network  Information  Center  No.  18516;
   Augmentation  Research  Center,  Stanford  Research Institute:
   Menlo Park, September 1973.

Feinler,  E.J.  and  Postel,  J.B.   ARPANET  Protocol  Handbook.
   Network  Information  Center  No.  7104; Augmentation Research
   Center, Stanford Research Institute:  Menlo Park,  April  1976.
   (NTIS AD A003890).

McKenzie,  A.   File  Transfer  Protocol.   ARPANET  Request  for
   Comments,  No.  454,  Network  Information  Center  No. 14333;
   Augmentation Research  Center,  Stanford  Research  Institute:
   Menlo Park, February 1973.

McKenzie,  A. TELNET Protocol Specification.  Network Information
   Center  No.   18639;  Augmentation  Research  Center, Stanford
   Research Institute:  Menlo Park, August 1973.

Myer, T.H. and Henderson, D.A.   Message  Transmission  Protocol.
   ARPANET  Request  for  Comments,  No. 680, Network Information
   Center  No.  32116;  Augmentation  Research  Center,  Stanford
   Research Institute:  Menlo Park, 1975.

Neigus,  N.   File  Transfer  Protocol.   ARPANET   Request   for
   Comments,  No.  542,  Network  Information  Center  No. 17759;
   Augmentation Research  Center,  Stanford  Research  Institute:
   Menlo Park, July 1973.

Pogran, K., Vittal, J., Crocker, D. and Henderson,  A.   Proposed
   official  standard  for  the  format of ARPA network messages.

Standard for the Format of Text Messages                       38
Bibliography




   ARPANET Request for Comments,  No.  724,  Network  Information
   Center  No.  37435;  Augmentation  Research  Center,  Stanford
   Research Institute:  Menlo Park, May 1977.

Postel, J.B.  Revised  FTP  Reply  Codes.   ARPANET  Request  for
   Comments,  No.  640,  Network  Information  Center  No. 30843;
   Augmentation Research  Center,  Stanford  Research  Institute:
   Menlo Park, June 1974.