RFC 365
Letter to All TIP Users
A Letter to All TIP Users π
Network Working Group Β· David Walden Β· Bolt Beranek and Newman Inc. Β· NIC # 10607 Β· July 11, 1972
RFC 365 Β· Categories: (none listed) Β· Updates: (none listed) Β· Obsoletes: (none listed) π
Dear TIP Users, π«Ά
Okay bestie, a revised version of the "TIP User's Guide" is about to land in your inbox β¨. It has a loose leaf format now, and the big news is that a bunch of new command descriptions were added. Appendix B will be an alphabetical list of every TIP command. The new commands are:
@BINARY INPUT START
@BINARY INPUT END
@BINARY OUTPUT START
@BINARY OUTPUT END
@CLEAR DEVICE WILD
@CLEAR INSERT LINEFEED
@INSERT LINEFEED
@SEND COMMAND
@RECEIVE FROM WILD
@SEND TO WILD
@SET DEVICE WILD
@MAG ABORT
@MAG BACKSPACE FILE
@MAG BACKSPACE RECORD
@MAG READ FILE
@MAG READ RECORD
@MAG SETUP COPY
@MAG SPACE FILE
@MAG SPACE RECORD
@MAG UNLOAD
@MAG WRITE EOF
@MAG WRITE TAPE
The MAG commands are, of course, not relevant if your TIP doesn't have the magnetic tape option π.
A TIP system with all of the above commands has been running since late June, and we think it's a substantial upgrade from the older versions π .
In case you haven't been keeping track, there are now eleven TIPs on the ARPA Network: ETAC, GWC, AMES, ARPA, MITRE, NBS, BBN, USC, NOAA, ROME, and SAAC. And a TIP is coming soon to CCA π.
Now let me spill about a few topics I think you'll find interesting π.
Getting TIP status information β¨
As we build new features and squash bugs, we keep releasing new versions of the TIP program. We do this by reloading your TIP when you're not looking, meaning when nobody is using it π₯·. Going forward, we'll let you know whenever a new version gets loaded into your TIP by adding a tiny "status message" to the "HELLO" you see when you "log onto" the TIP π. That status message will usually just be the TIP's version number. Sometimes, though, it'll tell you (by typing "NEWS") that there's news about the TIP's status that you should read before continuing your session. You can pull up the NEWS by typing the command @NEWS, which will ICP to a special socket at BBN's TENEX or PDP-1D and print the news π°. Of course, either of those systems might be down sometimes, but we won't stress about that until we see how serious it gets π.
To whom to complain or make suggestions π
Lots of you have complained about how your TIP runs, or suggested ways to make it better, and honestly, way too often those complaints went to the Host system you were using from the TIP instead of to us π. BBN runs a Network Control Center that is always manned. Its telephone number is 617-661-0100. Any TIP problems should be reported right away to the Network Control Center π. If you've got an idea for a new command, or you think you found a sneaky bug in the TIP program, ask to talk to Dave Walden or Bernie Cosell when you call the Network Control Center π¦.
Device buffers π§
Generally, each TIP MLC port has its own buffer size, and that buffer belongs to whatever device is connected to the port. Faster devices need bigger buffers, especially for output. If you don't have enough buffering (you'll usually notice it when output comes in short bursts π©), try to get whoever is locally responsible for your TIP to plug you into a port with a bigger buffer allocation. If that's not enough, that person can ask us to specially tailor the buffer allocation for your TIP π.
Here's the tea π«: there are only about 2500 characters of buffers available for the TIP's terminal devices. Every device gets one input buffer and two output buffers (output is double buffered), so each of the 63 devices has an average of about 13 characters for input and 13 characters for each output buffer. New features keep getting piled on, and old features keep getting asked to work more conveniently, so the space for device buffers keeps shrinking π. There are basically two ways to get more space: get more core memory, or modularize the program so unneeded features can be removed from particular sites. For example, the 2741 code and character conversion tables could be removed from a site that never uses them βοΈ. You always have the option of getting more core for your TIP. We're currently working on the second method and should have it ready in a few months. In the meantime, the space for device buffers will probably keep slowly diminishing π₯.
The nominal (untailored) buffer sizes are currently about this:
device input (characters) output (characters/buffer)
1-3 60 56
4-7 28 27
8-15 20 20
16-32 12 13
33-63 6 6
Echoing π₯
The TIP's echoing capability has been a hot topic (and not in a good way) in the past. We recently fixed it up a bit, and we think you'll like it a lot better if you try it now. Unfortunately, "@@" still isn't echoed correctly very often. That's a bug, and we're going to fix it, but it's hairy to fix π΅βπ«.
We'll soon be removing three of the ECHO commands (ECHO ALL, ECHO HALFDUPLEX, and ECHO NONE) and slightly changing what the other two ECHO commands (ECHO REMOTE and ECHO LOCAL) mean. We're also adding two new ECHO commands (PHYSICAL HALFDUPLEX and PHYSICAL FULLDUPLEX). So the new set of ECHO commands will be:
@PHYSICAL HALFDUPLEX
@PHYSICAL FULLDUPLEX
@ECHO LOCAL
@ECHO REMOTE
ECHO LOCAL and ECHO REMOTE will probably be ignored for physical halfduplex devices. So for physical fullduplex devices, ECHO LOCAL and ECHO REMOTE will declare the device to be virtual halfduplex or virtual fullduplex. Devices known to be physical halfduplex (like the IBM 2741's) will come up in physical halfduplex mode π¨οΈ. Every other device will come up in physical fullduplex mode and local echo mode (aka virtual halfduplex mode). Whenever a connection opens, the TIP automatically sends the current virtual echo mode to the server π€. Whenever an ECHO LOCAL or ECHO REMOTE command is given, the virtual mode gets updated and forwarded to the server. When the TIP receives the Telnet ECHO (code 132) (=virtual fullduplex) or Telnet NO ECHO (code 131) (=virtual halfduplex) characters, it follows these rules:
ECHO received and
TIP is physical halfduplex -- NO ECHO sent to server
TIP is virtual halfduplex -- ECHO sent to server and
mode changed to virtual
fullduplex
TIP is virtual fullduplex -- nothing
NO ECHO received and
TIP is physical halfduplex -- nothing
TIP is virtual halfduplex -- nothing
TIP is virtual fullduplex -- NO ECHO sent to server and
mode changed to virtual
halfduplex
When a connection breaks, devices in physical fullduplex mode get reset to virtual halfduplex mode π.
Some other command changes we're planning ποΈ
- PROTOCOL TO LOGIN is getting removed. It's been replaced by LOGIN.
- LOGIN will take a parameter, the Host number, so you won't need to give both HOST and LOGIN commands. Main character energy π.
Some things we're doing right now π οΈ
- When a terminal hangs up, any captured devices get given back.
- Output to high speed devices will be possible π.
- Data sets will automatically hang up when the caller disconnects, even if the caller never "logged in".
- 202C modems will work π.
- Major improvements to the magnetic tape option πΌ.
- Trying to find the "R half" of the "R T CLOSED" message π΅οΈ.
- We've gotten lots of complaints that the TIP once in a while loses an allocate. We'll fix it, or else prove it's not the TIP's fault π ββοΈ.
Some things we're thinking about π€
- Adding lower case capability for Model 33 Teletypes π€.
- Adding a mechanism so you can get the TIP's status, or a device's status (for example, its device number) π.
Each of the changes above will be announced through the NEWS first, and we plan to put out frequent updates to the TIP User's Guide from now on π.
We're really working hard to make the TIP better. If you've got a suggestion (especially one that doesn't take any memory), let me hear from you π«Άπ.
Regards,
< signed "Dave" >
David C. Walden Bolt Beranek and Newman Inc.
July 11, 1972
[ This RFC was put into machine readable form for entry ]
[ into the online RFC archives by BBN Corp. under the ]
[ direction of Alex McKenzie. 1/97 ]
π¦ where this page comes from
- ietf: rfc365 1972-07
- π girlified by claude-haiku-5-5 on : a language model's rewrite of the original, under its terms