User/Server Site Protocol Network Host Questionnaire
A 1971 questionnaire asking every ARPANET host how its terminals and character handling work, to inform Telnet.
O'Sullivan, T.C., "User/Server Site Protocol Network Host Questionnaire", RFC 106, DOI 10.17487/RFC0106, March 1971, <https://www.rfc-editor.org/info/rfc106>.
@misc{rfc106,
series = {Request for Comments},
number = 106,
howpublished = {RFC 106},
publisher = {RFC Editor},
doi = {10.17487/RFC0106},
url = {https://www.rfc-editor.org/info/rfc106},
author = {T.C. O'Sullivan},
title = {{User/Server Site Protocol Network Host Questionnaire}},
pagetotal = 5,
year = 1971,
month = mar,
} RFC text
NETWORK WORKING GROUP NIC 5776
Request for Comments #106 Thomas O'Sullivan
RDS
3 March 1971
USER/SERVER SITE PROTOCOL
NETWORK HOST QUESTIONNAIRE
The answer to a number of questions about the characteristics
and operation of Network HOSTS will be useful to implementers of some
early versions of TELNET, and/or to the USER/SERVER SITE PROTOCOL
Committee in their consideration of a full TELNET PROTOCOL
specification.
Questions of current interest are covered on the attached
questionnaire. The Technical Liaison contact at each HOST site is asked
to review these questions and fill in the answers as soon as convenient
after receipt.
One of the members of the committee listed below will contact
each HOST site by telephone in the next few days to record the answers.
Committee members include:
Bob Long SDC
John Melvin SRI/ARC
Ed Meyer MIT/MAC
Tom O'Sullivan RDS
Results of the questionnaire will be tabulated and distributed
as an RFC.
NETWORK HOST QUESTIONNAIRE*
For TELNET Consideration
Installation __________________________ Prepared By _______________
System** _____________________________ Date _____________________
1.) Does system operate on keyboard terminal input on one
character at a time (e.g. searching for a special
character other than end of line), or does it accumulate
a line at a time and operate on the line when a
special character is entered, e.g. Carriage Return (CR)?
[ ] Line at time
[ ] Character at time
[ ] Both (Explain) __________________________________
_____________________________________________________________
_____________________________________________________________
_____________________________________________________________
2.) What maximum buffer size does your system provide for
input of a physical line from a keyboard terminal, and
for output to a terminal, printer, or screen?
Input______Char
Output_____Char
Remarks: ____________________________________________________
_____________________________________________________________
_____________________________________________________________
_____________________________________________________________
=================
* If there is insufficient space on the questionnaire, continue
answers on back of page.
** Fill out separate questionnaire for each HOST to be directly
interfaced to an IMP.
Network HOST Questionnaire (Continued)
3.) Describe the effect in your system of use of the following
keying conventions. Indicate how implemented, i.e. is there
a hardware interrupt? Does the terminal respond? How?
What is sent into system as data, what is echoed, etc.
If the key(s) are not available, so indicate (as may be
the case for Newline for some systems, line feed for 2741's,
etc.). If available but has no meaning, so indicate.
Key Strokes Implementation System Action
CR __________________________ __________________
LF __________________________ __________________
NL __________________________ __________________
CR,LF __________________________ __________________
Terminal Type = __________________________
If appropriate, enter data for other terminal types on the
back of this page.
4.) What special character or characters are used in your
system to awaken some process during the entry of data
from a terminal (e.g. control C for some systems, break on
others, etc.)?
Character(s) Operation
___________ ___________________________________________
___________ ___________________________________________
___________ ___________________________________________
___________ ___________________________________________
Network HOST Questionnaire (Continued)
5.) List the types of keyboard terminals and codes supported
by your system.
__________________________________________________________
__________________________________________________________
__________________________________________________________
6.) What internal code(s) is used in your system to represent
character sets? If not ASCII nor EBCDIC, attach a copy of
the code.
__________________________________________________________
__________________________________________________________
__________________________________________________________
7.) For each terminal type do you support it in:
Half Duplex (Pure) _______________________________________
Half Duplex (with Break or Attention) ____________________
Half Duplex (Break & Reverse Break) ______________________
Full Duplex (without echo) _______________________________
Full Duplex (with echo) __________________________________
8.) For optional echo systems, at initial connect time
is the terminal assumed to require an echo?
[ ] Yes
[ ] No
Network HOST Questionnaire (Continued)
9.) Does your system perform error detection (e.g. parity
check) on terminal input and, if so, is this a hardware
or software check?
_____________________________________________________________
_____________________________________________________________
_____________________________________________________________
10.) Does your system support punched paper tape input? If so,
what sort of end of line sequence is used?
_____________________________________________________________
_____________________________________________________________
_____________________________________________________________
[ This RFC was put into machine readable form for entry ]
[ into the online RFC archives by S.M.H. 5/97 ]
AI ELI5
On this page
§1Why the Survey
RFC 106 is dated 3 March 1971 and was written by Thomas O'Sullivan of RDS. It says that answers about the characteristics and operation of Network HOSTs will help implementers of some early versions of TELNET and the User/Server Site Protocol Committee as it considers a full TELNET specification.
The Technical Liaison contact at each HOST site is asked to review the questions. A committee member will phone each site within a few days to record the answers. The committee is Bob Long (SDC), John Melvin (SRI/ARC), Ed Meyer (MIT/MAC) and Tom O'Sullivan (RDS).
A few people building a way for far-away terminals to talk to other computers first want to learn how each computer treats keyboards. So they phone every site with questions.
§2What the Questionnaire Asks
The form has ten numbered questions, with a separate form for each HOST directly interfaced to an IMP. It asks, for example, whether the system works on input one character at a time or one line at a time, and what maximum buffer size it gives for input and output of a physical line.
Other questions cover the effect of the CR, LF, NL and CR,LF keystrokes, the special characters used to awaken a process during data entry (such as control C or break), the keyboard terminal types and codes supported, and the internal character code (if not ASCII or EBCDIC, a copy of the code is to be attached).
The rest ask which duplex modes each terminal type supports, whether an optional-echo terminal is assumed to need echo at connect time, whether the system detects errors such as parity on terminal input, and what end-of-line sequence is used for punched paper tape.
The questions are things like: do you read one letter at a time or a whole line, and what happens when someone presses Return?
§3What Happens to the Answers
The RFC says results of the questionnaire will be tabulated and distributed as an RFC. The text printed here is the blank form, so it carries no answers itself.
This note only has the empty questions. The filled-in answers were to be shared later.