RFC 703 UNKNOWN TELNET · SURVEY

July, 1975, survey of New-Protocol Telnet Servers

A July 1975 headcount of which ARPANET hosts had adopted the new Telnet protocol.

On rfc-editor.org ↗
PUBLISHED
July 1975
AUTHORS
D.W. Dodds
STREAM
Legacy
BUILDS ON
RFC 679 — February, 1975, survey of New-Protocol Telnet servers
SERVER HOSTS SURVEYED
43
WITH NEW-PROTOCOL SERVER
25 (58%)
WITHOUT
16 (37%)
SURVEY DATE
10 July 1975
DOI
10.17487/RFC0703
ERRATA
Reported errata

RFC text

Original text · RFC Editor View on rfc-editor.org ↗
Network Working Group                                      D.W. Dodds
RFC # 703                                                  BBN-TENEXA
NIC # 32988                                                11 July, 1975


           July, 1975, Survey of New-Protocol TELNET Servers


In the nearly five months since the last report several new server sites
have joined the network and there has been some significant progress in
New-Protocol implementation.  Perhaps the New-Protocol implementation on
the TIPs scheduled for the end of the year will encourage further
progress in the near future.

There have been numerous changes since the last list (numbers in octal):

SRI-ARC (2) is no longer a server host;
MIT-AI (206), MIT-ML (306), LONDON (52), and OFFICE-1 (53) have added
    New-Protocol servers;
LLL-RISOS (25), SDAC-44 (32), and ARPA-DMS (34) have joined the network;
SCI-TENEX (55), Rutgers-10 (56), USC-ISIC (364), and SUMEX-AIM (70) have
    joined the network with New-Protocol servers.

The following is the latest version* of the summary and tabulation of
server-host TELNET servers.
        total server hosts             43      100%
        no New-Prot server             16       37%
        unknown status (new host)       2        5%
        total New-Prot implem.         25       58%
            New-Prot on 1 and 27,
                Old on socket 1 (2)    13       30%
            New-Prot on 1 and 27 (3)    8       19%
            New-Prot on 1 only (3)      4        9%

Notes:

*  All data in this report were gathered via a surveying program run at
    various times, plus a few manual checks to fill out the list.  What
    is reported here is the way the various servers work as seen by the
    new-Protocol User Telnet at BBNA, as of 10 July 1975.

(2) These are the sites whose operation is 100% correct according to all
    protocols and conventions, as I understand them.

(3) We realize that some of the servers that appear here as New-Protocol
    servers on socket 1 are actually servers which attempt to
    communicate with both Old- and New-Protocol User TELNETs according
    to what control sequences are received.

           Tabulation of server status for all server sites:

 Host  Number   Host            Socket  Socket   New-Prot, Options
(Oct)   (Dec)   Name              1       27     Implementated (if any)

101     65      UCLA-CCN        Old     X
201     129     UCLA-CCBS       Old     X
102     66      SRI-AI          Old     New      I1,3,6; O3
3       3       UCSB-MOD75      Old     X
4       4       UTAH-10         Old     X
305     197     BBN-TENEXA      Old     New      I1,3,6; O3
106     70      MIT-DMS         New     New      I1,3; O3
206     134     MIT-AI          New     X        I1,3; O3
306     198     MIT-ML          New     X        I1,3; O3
7       7       RAND-RCC        Old     X
107     71      RAND-ISD        ?       ?
10      8       SDC-LAB         Old     X
11      9       HARV-10         New     X        I1,3; O3
12      10      LL-67           New     X        None
112     74      LL-TX-2         Old     X
13      11      SU-AI           New*    New*     I1,3
16      14      CMU-10B         New     New      I1,3; O3
116     78      CMU-10A         New     New      I1,3; O3
17      15      I4-TENEX        Old     X
217     143     I4B-TENEX       Old     X
20      16      AMES-67         New     New      None
25      21      LLL-RISOS       Old     Old
126     86      USC-ISI         Old     New      I1,3,6; O3
226     150     USC-ISIB        Old     New      I1,3,6; O3
27      23      USC-44          Old     X
327     215     USC-ECL         Old     X
32      26      SDAC-44         Old     X
34      28      ARPA-DMS        Old     X
37      31      CCA-TENEX       Old     X
40      32      PARC-MAXC       Old     New      I1,3,6; O3
43      35      UCSD-CC         Old     New      I0(!),3; O0,3
344     228     HAWAII-500      Old     X
52      42      LONDON          New     New      None
53      43      OFFICE-1        Old     New      I1,3,6; O3
54      44      MIT-MULTICS     New     New      None
55      45      SCI-TENEX       Old     New      I1,3,6; O3
56      46      RUTGERS-10      New     New      I1,3; O3
61      49      BBN-TENEXB      Old     New      I1,3,6; O3
361     241     BBN-TENEX       Old     New      I1,3,6; O3
162     114     BBN-TENEXD      Old     New      I1,3,6; O3
364     244     USC-ISIC        Old     New      I1,3,6; O3
67      55      ANL             ?       ?
70      56      SUMEX-AIM       Old     New      I1,3,6; O3

Key:    X  No server at this socket
        ?  Status not ascertained-unable to connect to host
        I# Option # implemented incoming to user (Server says "Will #")
        O# Option # implemented outgoing from user (Server says "Do #")
           (# is option number in new Protocol, Options in tabulation:
                0 Transmit-Binary       1  Echo
                3 Suppress-Go-Ahead     6  Timing-Mark)

Note: * These servers return improper responses to some TELNET option
requests.










       [ This RFC was put into machine readable form for entry ]
       [ into the online RFC archives by Alex McKenzie with    ]
       [ support from GTE, formerly BBN Corp.            11/99 ]

AI ELI5

On this page
PROBLEMThe network was moving from the old Telnet protocol to a new one, and implementers needed to know which server hosts had made the switch.
MECHANISMA surveying program, plus a few manual checks, probed each server host as seen by the New-Protocol User Telnet at BBNA, as of 10 July 1975, and the results were tabulated by socket and option.
WHAT IT FOUNDOf 43 server hosts, 25 (58%) had New-Protocol implementations and 16 (37%) did not; two were of unknown status.

§1What Changed Since the Last Survey

The report comes nearly five months after the previous one (RFC 679). Several new server sites joined and there was significant progress with the New-Protocol, and the author hopes the TIP implementation scheduled for the end of the year will encourage more.

SRI-ARC is no longer a server host. MIT-AI, MIT-ML, LONDON and OFFICE-1 added New-Protocol servers. LLL-RISOS, SDAC-44 and ARPA-DMS joined the network, and SCI-TENEX, Rutgers-10, USC-ISIC and SUMEX-AIM joined with New-Protocol servers. Host numbers in the text are octal.

Since the last check, some computers left the network, some joined, and several learned the new language.

§2The Tally

Of 43 server hosts, 16 (37%) had no New-Protocol server, 2 (5%) were new hosts of unknown status, and 25 (58%) had a New-Protocol implementation. Within those 25, 13 (30%) ran the New-Protocol on sockets 1 and 27 while keeping the Old protocol on socket 1, 8 (19%) ran it on 1 and 27, and 4 (9%) on socket 1 only.

The hosts in the first group are the ones the author understands to be 100% correct under all protocols and conventions. Some servers listed as New-Protocol on socket 1 actually try to talk to both Old and New user Telnets, depending on the control sequences received.

More than half the computers had the new language. A few could speak both, depending on who called.

TELNET SERVER SURVEY SUMMARY, 10 JULY 1975
CategoryHostsShare
Total server hosts43100%
No New-Prot server1637%
Unknown status (new host)25%
Total New-Prot implementations THIS RFC2558%
New-Prot on 1 and 27, Old on socket 11330%
New-Prot on 1 and 27819%
New-Prot on 1 only49%

§3The Per-Host Table and Options

A table lists each server host with its number, name, the state of socket 1 and socket 27 (Old, New, X for no server, or ? when it could not be reached), and the New-Protocol options implemented. Option codes: I# means option # is implemented incoming to the user (the server says "Will #"), and O# means outgoing (the server says "Do #"). The options tabulated are 0 Transmit-Binary, 1 Echo, 3 Suppress-Go-Ahead and 6 Timing-Mark.

Two hosts, RAND-ISD and ANL, could not be reached. Servers marked with an asterisk, such as SU-AI, return improper responses to some TELNET option requests. All data came from the surveying program plus a few manual checks.

The table is a report card: for each computer it shows whether it answers the old door, the new door, or neither, and which extra tricks it knows.

§4References