Network Working Group Peggy Karp
Request for Comments: #247 MITRE
NIC 7688 12 October 1971
Categories: Policy, Telnet
Related: #226, 236, 239, 233, 237
Proferred Set of Standard Host Names
In RFC #226, BBN's TENEX list of Host names was set up as a strawman
set of standard Host names. Comments received since then (an RFC
actually generated comments!!!) have influenced me to propose the
following general rules for forming Host names.
The Host names will be 8 characters in length. The general form is
<site> '-' <machine>
<site> will be at most 4 characters, formed as follows:
(a) Use the keyword in the site name, if not more than
four characters, e.g., NASA Ames, Case Western
Reserve. ---- ----
(b) Use the standard acronym, if not more than four
characters, e.g., UCLA, RADC, NBS.
(c) If a standard abbreviation exists, use it, e.g., Ill.
(d) If none of the above apply, use the first four letters
in the site name, e.g., Burr, Mitr, Harv.
(e) If none of the above is acceptable to the site, the
technical liaison should select the site mnemonic.
<machine> will be at most 4 characters of the form <mfg. #>
Examples of mfg. # are:
IBM 360 2 digit model number
IBM 370 3 digit model number
PDP 1 - 2 digit model number
Burroughs 4 digits
CDC 4 digits
<designator> will be used when more than one machine of the same
type is located at a site (e.g., 2 PDP-10s at MIT, at SRI, and
Limiting <machine> to 4 characters does not permit distinctions
to be made between machines with 4 digit mfg. #s. I expect
the situation will be handled in an ad hoc manner by the NIC if
TIPs are identified as 'TIP' rather than by '316'. If a Host
is not to be permanently addressable, the machine is identified
A list of Host names, formed according to these rules, is
attached. Alternate Host names should be provided, as
suggested by Jon Postel (RFC #236). RFC's 206, 233, and
236 present lists with 4-character alternate names. The
Technical Liaison should select the alternate name for his
site and communicate the selection to the NIC.
The preceding rules and the attached list of Host names are
subject to the approval of the NWG. Hereafter, the list will
be generated and maintained by the NIC in cooperation with
the Technical Liaison at each site, as suggested in RFC #237.
Comments should be addressed to Dick Watson.
[ This RFC was put into machine readable form for entry ]
[ into the online RFC archives by BBN Corp. under the ]
[ direction of Alex McKenzie. 12/96 ]
NETWORK ADDRESS STANDARD NAME
An Implementation Scheme
If the standard Host names are formed according to the proposed
rules, the following implementation scheme, suggested by Steve
Crocker, can be used.
Map <site> into an 8-bit number, S and
map <machine> into an 8-bit number, M,
S + M = Network Address.
S and M can be selected such that specification of <site>
alone could cause a default to the "primary" Host at
the site. Note that this scheme depends on a unique
<site> designator for each IMP.
If the "primary" Host at UCLA is the 91, let
UCLA -> S = X'41'
7 -> M = X'40'
91 -> M = X'00'
UCLA-7, S + M = X'01' = 1 base 10
UCLA-91,S + M = X'41' = 65 base 10
UCLA alone = X'41' = 65 base 10
If the primary Host at BBN is TENEX System A, let
BBN -> S = X'45'
516 -> M = X'40'
10A -> M = X'00'
10B -> M = X'C0'
BBN-516, S + M = X'05' = 5 base 10
BBN-10A, S + M = X'45' = 69 base 10
BBN-10B, S + M = X'85' = 133 base 10
BBN alone = X'45' = 69 base 10
The primary Host for each IMP would be designated by the
site and such information disseminated by the NIC.
Back to RFC index