Talk:TECO (text editor)

TECO hype

From the "Standard TECO <nl> TEXT EDITOR AND CORRECTOR FOR THE <nl> VAX-11, PDP-11, PDP-10, and PDP-8 <nl> USER'S GUIDE AND LANGUAGE REFERENCE MANUAL" May 1985 Edition:

"... a novice TECO user can begin creating and editing text files after only a few hours of instruction."

Which only goes to show how much things have improved. --Kop 17:39, 4 Sep 2004 (UTC)

TECO description on Emacs

Could a TECO user please reivew the description of TECO, circa 1976, as written at the top of Emacs#Emacs history? I've rewritten the paragaraph but have only looked at the 1985 manual and never used TECO. Thanks. --kop 00:22, 5 Sep 2004 (UTC)

Ye gods, I used TECO on the university computer, in Frankfurt, in 1980, to write my dissertation. (That was before anyone had heard of something called a 'personal computer.') Jeepers, 25 years ago.... I think I still have all the macros from back then - I mean: I wrote my own goddamn text editor... (that's for the nostalgia corner)

"Write-only": POV or not?

The Trivia section's description of TECO as a "Write-only" language strikes me as a tad biased. It should be NPOV-ified, or at least referenced. --Maru (talk) 20:19, 16 September 2005 (UTC)

You've apparently never tried to read a TECO program. It's entirely a fact that most can't be read without a huge investment of time and effort. (And no, I'm not the editor who inserted that statament :-) ! )

Atlant 13:07, 20 September 2005 (UTC)

Well, if it's really that obvious, then finding a reference for that desciption of TECO should be pretty trivial. --Maru (talk) 17:56, 20 September 2005 (UTC)

Sure: Here's the reference from FOLDOC: [1]. And that took me all of five seconds with Google, so maybe next time you can even look it up yourself!

But I'm sure you would have had no trouble understanding the following perfectly-clear TECO program:

   [1 J^P$L$$
   J <.-Z; .,(S,$ -D .)FX1 @F^B $K :L I $ G1 L>$$

Atlant 14:16, 21 September 2005 (UTC)

Pffth! Why don't you try something a little more challenging? It's obvious that that implements an interpreter and bootstrap compiler. --Maru (talk) 19:42, 21 September 2005 (UTC)

Whose Building 26 are we talking about?

The current History paragraph refers to two MIT PDP-1 computers housed in DEC's Building 26. Isn't this in error? To my knowledge, DEC buildings were not numbered. RLE had a PDP-1 computer, and perhaps TMRC had one also. Most early PDP-1 hackers were TMRC members. The RLE machine was in MIT's Building 26 (or possibly 24) on the second floor, next door to the TX-0, to which it had a kludgy data connection. TMRC was in Building 20, and it's thus likely that its PDP-1 would have been there also, but I don't remember seeing one there. I'm disinclined to make any correction to this aspect of the history until someone from MIT, preferably an old TMRC-er from RLE, steps forward.

Side note - FWIW, In a practical sense, TECO is generally write-only. By the time you have decoded someone else's TECO command string you could easily write your own version to do the same thing, probably better. I speak as someone who used TECO as recently as 1995, on a VAX no less, because it was too much trouble to do some random but fancy editing task any other way. Snezzy 05:08, 12 June 2006 (UTC)

Buildings in DEC's Maynard Mill complex were numbered (and I spent time in Building 3 and Building 5), but IIRC, the numbering didn't go as high as 26, and even if there was a 26, it would have been an outpost like the Security building rather than any of the "real" buildings.

So your supposition that this is an MIT building sounds a lot more like reality to me.

Atlant 13:08, 12 June 2006 (UTC)

It was in building 26 of MIT, on the first floor, housed in the room next to the room with the TX-0 (reputed to be the first transistor computer), where I encountered the PDP-1 referred to here. I don't remember TECO from the PDP-1. When I used the PDP-1 (actually, PDP-1x, as it had paging modifications and twos complement arithmetic), ET (Expensive Typewriter) seemed to be the editor of choice, while at the same time as TECO was used on the MIT AI Lab PDP-6 and PDP-10 (both running ITS). This was slightly after the peak time for the PDP-1x and the PDP-6 at MIT, being around the fall of 1971 and a little earlier. foxkid (talk) 18:42, 28 October 2009 (UTC)

TECO as an editor

While I agree that TECO programs can be incomprehensible, TECO used as an editor wasn't that bad. I think it's comparable to editors like ed. I've added a section showing a simple TECO editing session to show this side of TECO. Alas, it's been over twenty years since I last used it for real, and can't find a copy to run right now, so I'm not 100% sure of the example. Of course, there are any number of ways to make any given change and this is only one of them. Jordan Brown 07:11, 11 September 2006 (UTC)

Got a copy of Tom Almy's TECO for Windows and tested that edit session; it's fine. Boy, that's a blast from the past. Jordan Brown 07:45, 19 December 2006 (UTC)

calculate pi with TECO

No discussion of TECO is complete without this program. I'm not sure it's really encyclopedic, but it is insane.

+0UN QN"E 20UN ' BUH BUV HK
QN< J BUQ QN*10/3UI
QI< \+2*10+(QQ*QI)UA B L K QI*2-1UJ QA/QJUQ
QA-(QQ*QJ)-2\ 10@I// -1%I >
QQ/10UT QH+QT+48UW QW-58"E 48UW %V ' QV"N QV^T ' QWUV QQ-(QT*10)UH >
QV^T @^A/
/

I have no idea how it works, but it does. It defaults to calculating 20 digits, but you can ask for a different number by typing a number before the program, e.g. "100+0UN ...". (I think it's intended to be executed out of a Q-register, so that the number of digits is a straightforward parameter to the macro call. Unfortunately I don't remember how to put it into a Q-register or execute it from one.) Here's a few tidbits of how it works...

+0UN	Add zero to the current value and store it in register N.
QN	Retrieve the value of register N
"E ... '	Do this stuff if the value is equal to zero
20UN	Store 20 into register N
B	Synonym for zero (usually used to refer to the start of the buffer)
UH	Store it in register H
BUV	Store zero in register V
H	Equivalent to B,Z: the entire buffer
K	Kill (so HK empties the buffer)
QN< ... >	Do the enclosed stuff as many times as register N says
... The rest is left as an exercise for the reader ...

I'm pretty sure this program is at least 30 years old, since I saw it (or some other program that calculated pi in TECO :-) back when I was using TECO in ~1976.

I grabbed this from [2]. Nearby they have a PDP-10 simulator running TOPS-20...

Jordan Brown 08:27, 19 December 2006 (UTC)

Here's more detail on PI.TEC: [3]. Among other things it mentions the author (Stan Rabinowitz) and even an academic journal article reference ([4]) on the subject. Paul Koning 16:02, 23 March 2007 (UTC)

While I was a student at York in the late 1970s some maniac wrote a program in Teco to solve the Mastermind game. Now, how many of you ever typed 0YY? Number774 (talk) 20:24, 1 October 2012 (UTC)

Rename article to TECO (text editor)?

Given that the editor in question is generally refered to just as "TECO" I would like to propose to rename the article from Text Editor and Corrector to TECO (text editor). This is probably somewhat controversial as expansion of abbreviations is apparently the prefered method for disambiguation of article titles. However, I was just browsing through Category:Text editors and was confused for a while because I did not immediately realise that the name "Text Editor and Corrector" refered to TECO. It also would match the usage in the article text itself. (The name "TECO (text editor)" would also allow to use the pipe trick when linking to this article, but this is only relevant to wiki editors, not to wiki readers, and therefore not really relevant.) —Tobias Bergemann 10:26, 22 October 2007 (UTC)

A redirect would serve the same purpose Tedickey 10:36, 22 October 2007 (UTC)

No, it wouldn't, as the article would still be listed under the title "Text Editor and Corrector" in its categories. —Tobias Bergemann 11:11, 22 October 2007 (UTC)

Perhaps - but googling on "TECO", I see that it has relatively low name-recognition (perhaps 10% of the hits - the other 90% are unrelated commercial abbreviations) Tedickey 11:40, 22 October 2007 (UTC)

And reading further, I see for example a hit on wikipedia's dab page for "TECO". While popular long ago, it's pretty apparent that TECO-the-editor hasn't enough name-recognition to displace other topics. Of course, changing wikipedia around would increase it.... Tedickey 11:45, 22 October 2007 (UTC)

pronunciation of "TECO"

"tikou" is a variant I've never heard. Perhaps a reliable source is needed here. Tedickey (talk) 17:30, 11 December 2008 (UTC)

/'tikoʊ/ is the only version I have heard (well, maybe /'tiːkoʊ/), and I was at MIT Project MAC from 1971-1977, when TECO was in active use. Pronunciation on this page was previously given as tee'koh, which presumably means the same thing, using the World Book Encyclopedia's pronunciation system. But IPA is the Wikipedia standard. --macrakis (talk) 00:10, 12 December 2008 (UTC)

It seems that I was misled by the fonts at hand - on a different machine, I see the "ʊ". So... that's "iː bead, peat, feet" and "oʊ beau, hoe, poke". Sounds okay...Tedickey (talk) 00:31, 12 December 2008 (UTC)

Not all character strings are TECO programs!

The page previously to my edit stated that "thus any character string is a TECO program", which is simply untrue. TECO has a well-defined syntax, and will reject ill-formed command strings just as any other programming language interpreter will. As examples of syntax restrictions: only single alphabetic characters or single digit characters are valid as Q-register names; parentheses and angle brackets must occur in balanced pairs; and so on. I could easily add further examples.

The statement to which I've objected may well be reminiscently humorous and rhetorically pointed, but falsehoods shouldn't be in an encyclopaedia.

As always, feel free to disagree with my changes. Kay Dekker (talk) 20:18, 13 June 2009 (UTC)

I agree completely. If *every* string were a program, I could never have found space to add the "." series of Q-register names, nor the <cntl>-closebracket commands. Of course, the name space was pretty full, except for big gaps in the multi-letter commands (F_ and E_, for instance). Off hand, I don't remember any other free characters.

The worst was the change in keyboard standards to place the underscore "_" character on the same keycap as the minus "-" character. I ran into it first on some Memorex(?) chain printer terminals. Of course, minus simply means negation, but underscore meant "search through the file, dumping buffers as necessary, until you find the string which follows." On a printing terminal, it was common to use "-TT" after an edit to print the current line. If that was "_TT" instead, you would usually end up with your edits dumped, and facing an empty buffer. foxkid (talk) 18:52, 28 October 2009 (UTC)

Wiki as a source...

A wiki seems to lack the requirements for a reliable source Tedickey (talk) 21:32, 27 October 2009 (UTC)

I'm trying to contribute properly here, and understand what should be attributed. Mostly, I write from personal experiences at the time, since this is computer history. How should such material be cited, or is it improperly placed here? Actually, I was pleased to find *any* reference to some of that history.

I understand that a wiki isn't as reliable as a published book. What would you suggest? foxkid (talk) 22:00, 27 October 2009 (UTC)

well, the main issue with a wiki is that it's not possible to source the statements to a single individual (or group) who're known for their expertise. Perhaps by using the material in the wiki you could find some website where the material is discussed in that manner. Tedickey (talk) 22:10, 27 October 2009 (UTC)

our

The use of "our" as in "...program source on our PDP-1..." is a bit idiosyncratic :) — Preceding unsigned comment added by 69.228.170.193 (talk) 23:34, 27 February 2018 (UTC)