[TUHS] Curly braces: An evolution of UNIX and C

[segaloco via TUHS]

On Wednesday, May 20th, 2026 at 16:07, G. Branden Robinson via TUHS <tuhs at tuhs.org> wrote:

> But it should have been solidly within USG/
> "Program Generic"'s wheelhouse.  And eventually it was, but took
> something like 7 years to get there.
> 
> What gives?

So one of the earlier attempts at expanding the terminal support at USG was addition of line disciplines, I *think* in an evolutionary dead-end fashion, but I don't know enough about terminal drivers in AT&T offerings to be sure whether it died in the 70s or was incorporated in later efforts.

In any case, the tty(IV) page of the USG Program Generic Issue 3[1] manual is informative.  My understanding is that some of the work was done to specifically support half-duplex terminals[2], among them the Dataspeed/TTY 40/1.  The USG Issue 3 manual explicitly adds bits to How to Get Started, the tty driver page, and a few others describing the new TTY 40 compatibility.

Indeed in much AT&T promotional photography and the 3B20S UNIX 4.1 User's Manual[4], the terminal in use at the turn of the 80s is often pictured as a TTY 40 of some kind.  There were 4 models, with the 40/1 being half-duplex, the 40/2 supporting both half and full, and the 40/4 being synchronous instead of async (I don't recall what set the 40/3 apart).  As an aside Teletype also produced hardened models for the USDOD Tempest program, some advertisements for these are floating around the net.

A nice design factor in the Teletype 40 is it shares a lot in common with the Dataphone modems of the time (black on aluminum, silver outline accents on bezels) that then showed up briefly in some post-divestiture commercial products such as the Sceptre videotex terminal.  Their hardware design language at the time appeared to be blue metal paneling (ala 3B20S) for large compute hardware and then this silver-on-black-on-aluminum facade for peripherals and user devices a welcome change from the "everything is grey and beige" look of most telco offices of the time.  Granted these are cursory observations, running down "fact" on these design languages and decisions is proving nigh impossible.  I wonder if any of those design specs will ever surface....

But yeah in conclusion, the mid 70's Dataspeed 40 line is one I rarely see mentioned or talked about in terminal discussions but so much leads me to believe Ma Bell was trying to push the 40/1 and 40/2 as the "new standard" in UNIX terminals, at least internally, in the late 70s and early 80s.

- Matt G.

P.S. Where Thalia you are on the hunt for a TTY 37, I'm looking for its younger cousin the 40/2.  Should your own search yield any TTY 40 stuff, could you shoot me a headsup?  Will do the same if I spot a 37, I comb around internet auctions for TTY stuff from time to time.  Many 40 parts on eBay right now...but just some boards, power supply, and keyboard.

[1] - https://www.tuhs.org/Archive/Documentation/Manuals/Program_Generic_Issue_3/UNIX_Programmers_Manual_Program_Generic_Issue_3.pdf
[2] - https://www.tuhs.org/Archive/Documentation/TechReports/USG_Library/1077_UNIX_DH_11_Driver_to_Support_Both_TTY_and_Dataspeed_40_Terminals.pdf
[3] - https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcQH68WMstdCtpSbj0_kIqhEtfiJK-iKKpnEsxxxEk1sfjwd8C6x_WP3Sw-i&s=10
[4] - https://upload.wikimedia.org/wikipedia/commons/thumb/7/74/UNIX4.1UsersManualCover.png/960px-UNIX4.1UsersManualCover.png