8 Jul
2016
8 Jul
'16
2:52 p.m.
On Fri, Jul 8, 2016 at 7:09 AM, Steffen Nurpmeso <steffen(a)sdaoden.eu> wrote:
...and that actually makes me wonder why the engineers
that
created what became POSIX preferred slash instead
I can not speak for anyone else. But at the time when I was a part of
the /usr/group UNIX standards** mtgs I personally do not believe I had
ever heard of the term "solidus." Such a term maybe had been used in my
first form Latin classes from the 1960s, but by the 1980s I had long ago
forgotten any/all of my Latin. I certainly did not try to remember it as a
computer professional.
In those days many of us, including me, did (and still do) refer to the
asterisk as "splat" and the exclamation point as "bang" from the
sound
made by them when they printed yellow oiled paper @ 10 cps from the console
TTY. But slash was what we called the character that is now next to the
shift key on modern keyboards. I do not remember ever using, much less
needed to refer to, the character "back slash" until the unfortunate crap
that the folks in Redmond forced on the industry. Although interestingly
enough, the vertical bar or UNIX "pipe" symbol was used and discussed
freely in those days. I find it interesting that Redmond-ism became the
unshifted character, not the vertical bar by the shear force of economics
of the PC.
Clem
** For those that do not know (my apologies to those that do) the 1985
/usr/group standards committee was the forerunner to IEEE P1003. Which we
published as the first "official UNIX API standard agreed by the community"
(I still have a hardcopy). But neither /usr/group nor USENIX had the
political authority to bring an official standard to FIPS, ANSI, ECMA, ISO
or like, while IEEE did. So a few months before the last meeting, Jim
Issak petitioned IEEE for standards status, and the last meeting of the
/usr/group UNIX standards meeting was very short -- about 10 minutes. We
voted to disband and then everyone in the room officially reformed a few
minutes later all signing in as IEEE P1003, later to be called POSIX. For
further historical note, I was a "founding member" of both groups and the
editor of a number of early drafts (numbers 5-11 IIRC), as well as the
primary author of the Tape Format and Terminal I/O sections of P1003.1.
With Keith Bostic, I would later be part of the P1003.2 and pen the
original PAX compromise. After that whole mess I was so disgusted with the
politics of the effort, I stopping going to the POSIX mtgs.
PPS While I did not work for them at the time, you can blame DEC for the
mess with the case/character sets in the POSIX & FIPS standards. A number
of the compromises in the standard documents were forced by VMS, 7-bit
(case insensitivity) being the prime one. While we did get in the
rational section of document that it was suggested/advised that systems
implementations and applications code be case insensitive and 8 bit clean
so that other character sets could be supported. However the DEC folks
were firmly against anything more than 7-bit ASCII and supporting anything
in that character set. My memory is that the IBM folks were silent at the
time and just let the DEC guys carry the torch for 1960's 7-bit US English.
8 Jul
8 Jul
2:59 p.m.
New subject: Slashes (was: MS-DOS)
On Fri, Jul 8, 2016 at 10:52 AM, Clem Cole <clemc(a)ccc.com> wrote:
We voted to disband and then everyone in the room
officially reformed a
few minutes later all signing in as IEEE P1003, later to be called POSIX.
I should have added, the /usr/group UNIX standard document became "draft
0" of the IEEE P1003.1 after being put into "official" IEEE format, with
some magic macro work by Jim, John Quarterman, and myself. Ah the wonders
of troff. And it was originally formatted on a Masscomp MC-500 which I
would bring to the IEEE meetings, which was considered a wonderment at that
time. We could actually edit the document as we discussed it!!! There
was no such thing as a laptop and the even the "Compaq Luggable" PC was
still a few years out.
3:47 p.m.
New subject: Slashes (was: MS-DOS)
On 8 July 2016 at 10:52, Clem Cole <clemc(a)ccc.com> wrote (in part):
I do not remember ever using, much less needed to
refer to, the character "back slash"
printf("I do not remember ever using, much less needed to\n\
refer to, the character \"back slash\"\n");
And thank you for the very interesting historical notes.
N.
4:31 p.m.
New subject: Slashes (was: MS-DOS)
Nemo <cym224(a)gmail.com> wrote:
|On 8 July 2016 at 10:52, Clem Cole <clemc(a)ccc.com> wrote (in part):
|> I do not remember ever using, much less needed to
|> refer to, the character "back slash"
|
|printf("I do not remember ever using, much less needed to\n\
| refer to, the character \"back slash\"\n");
|
|And thank you for the very interesting historical notes.
Fish.
--steffen
8:29 p.m.
New subject: Slashes (was: MS-DOS)
Good point.... shows how good my memory is... sigh...
On Fri, Jul 8, 2016 at 11:47 AM, Nemo <cym224(a)gmail.com> wrote:
On 8 July 2016 at 10:52, Clem Cole
<clemc(a)ccc.com> wrote (in part):
I do not remember ever using, much less needed
to
refer to, the character "back slash"
printf("I do not remember ever using, much less needed to\n\
refer to, the character \"back slash\"\n");
And thank you for the very interesting historical notes.
N.
4:27 p.m.
New subject: Slashes (was: MS-DOS)
Hello.
Clem Cole <clemc(a)ccc.com> wrote:
|On Fri, Jul 8, 2016 at 7:09 AM, Steffen Nurpmeso <[1]steffen(a)sdaoden.\
|eu[/1]> wrote:
|
|...and that actually makes me wonder why the engineers that
|created what became POSIX preferred slash instead
|
|I can not speak for anyone else. But at the time when I was a part \
|of the /usr/group UNIX standards** mtgs I personally do not believe \
|I had ever
|heard of the term "solidus." Such a term maybe had been used in my fi\
|rst form Latin classes from the 1960s, but by the 1980s I had long a\
|go forgotten
|any/all of my Latin. I certainly did not try to remember it as a com\
|puter professional.
Of course. It doesn't even sound friendlier with its origins in
the Roman culture, as a name for money. I wasn't consciously
aware of this once i've responded.
|In those days many of us, including me, did (and still do) refer to t\
|he asterisk as "splat" and the exclamation point as "bang" from the
\
|sound made by
|them when they printed yellow oiled paper @ 10 cps from the console T\
|TY. But slash was what we called the character that is now next to t\
That is something to remind. The standard has two occurrences of
bang, in the shell syntax and for the Mail variable of the same name.
And yes, young, brilliant and highly educated men and women
explore new worlds. So splat makes a lot of sense.
If seen from this angle it is even less enjoyable that Linguists
put more prominence on the correlation of slash and solidus.
|he shift key
|on modern keyboards. I do not remember ever using, much less needed\
| to refer to, the character "back slash" until the unfortunate crap t\
|hat the folks
|in Redmond forced on the industry. Although interestingly enough, t\
You mean the word? We don't have a word for this character in
German, as far as i know. (But you surely used it for \n \t etc.)
|he vertical bar or UNIX "pipe" symbol was used and discussed freely i\
|n those days.
| I find it interesting that Redmond-ism became the unshifted charact\
|er, not the vertical bar by the shear force of economics of the PC.
This sentence prevented work and caused an interesting journey via
Wikipedia. I am looking at a DEC VT52, from before Microsoft.
(I really like my current Apple (US) keyboard, after say twenty
years of sorrow in which i couldn't forget the robust metal
i think IBM keyboards that the administrators of the insurance my
father worked at used in their two-skyscraper-floors bunker.)
--steffen
6 p.m.
New subject: Slashes (was: MS-DOS)
On Fri, 8 Jul 2016, Clem Cole wrote:
I find it interesting that Redmond-ism became the
unshifted character,
not the vertical bar by the shear force of economics of the PC.
Beta documents suggest M$ wanted to switch the...switch character to dash
(-) for 2.0 and use / for the path separator, but *IBM* insisted on
keeping / and using \ for path separator...and what IBM wanted, IBM got.
That said I think one or two OEMs did in fact set the SWITCHAR to - and
use / for a path separator anyway.
-uso.
6:23 p.m.
New subject: Slashes (was: MS-DOS)
On Fri, Jul 8, 2016, at 10:52, Clem Cole wrote:
I can not speak for anyone else. But at the time
when I was a part
of the /usr/group UNIX standards** mtgs I personally do not believe I
had ever heard of the term "solidus." Such a term maybe had been used
in my first form Latin classes from the 1960s, but by the 1980s I had
long ago forgotten any/all of my Latin. I certainly did not try to
remember it as a computer professional.
In those days many of us, including me, did (and still do) refer to
the asterisk as "splat" and the exclamation point as "bang" from
the
sound made by them when they printed yellow oiled paper @ 10 cps from
the console TTY. But slash was what we called the character that is
now next to the shift key on modern keyboards. I do not remember
ever using, much less needed to refer to, the character "back slash"
until the unfortunate crap that the folks in Redmond forced on the
industry.
You never had to use it for escaping in C/Regex/Troff/etc?
Although interestingly enough, the vertical bar or
UNIX "pipe" symbol
was used and discussed freely in those days. I find it interesting
that Redmond-ism became the unshifted character, not the vertical bar
by the shear force of economics of the PC.
ASCII keyboards had \ unshifted long before the PC. The ASR-33 didn't
have it (it didn't have pipe at all - backslash was on shift-L), but
every DEC keyboard I can find did, as did the ADM-3a, and incidentally
the Symbolics Lisp Machine's keyboard.
I had suspected the reason that [\] ended up as the unshifted characters
is because {|} were not available on uppercase-only keyboards, but I
can't find any uppercase terminals that had separate keys for them (the
ASR-33 had them on shift- KLM). I expect that's also why ^, on shift-N,
was used for pipes rather than the vertical bar in the earliest versions
of Unix that had them.
What terminals did you use, back in those days?
(Incidentally, I can find *literally no* pictures of a Teletype 37,
and no sufficiently high-quality closeups of a 38 to determine its
keyboard layout)
8:49 p.m.
New subject: Slashes (was: MS-DOS)
On Fri, Jul 8, 2016 at 2:23 PM, Random832 <random832(a)fastmail.com> wrote:
What terminals did you use, back in those days?
Hard copy (that I remember, but we have proven how poor memory can be):
- ASR/KSR 33s originally, we had one 37 but it was not on a system I
used
- IBM console's printers Model numbers I've forgotten (but with an APL
ball)
- Later Xerox DaisyWheel printers Model numbers I've forgotten
- Later DEC Decwritter I and II
Glass TTY (not complete but for a quick memory dump)
- Lear Siegler ADM3 (lots of them - many people made them as kits) and
why we have hjkl as the movement keys in vi - the arrows were embossed on
those key tops
- PE "Fox" (CMU has a large lot deal and these became the standard there
in the late 70s)
- Triple Drip "Graphic Wonders" (man I miss these with a dedicated PDP11
and an amazing keyboard --- best game platform I ever knew)
- Tektronix 401x series (just about all models of them) but 4014 was
used the most
- DEC VT52
- Eventually, VT-100, PT-100 and a number of other VT-100 knock offs
- Eventually Tek 4025s (until the RT, one of the best keyboards - used
in Magnolia BTW)
- Heathkit H19 (still have mine that I built)
- Eventually Wyse 100's, 99GTs and Wyse 60s -- later being the best Wyse
(I still have one)
- Ann Arbor Ambassador (my all time favorite - wish I still had one)
- Too many different graphics terminals to remember
- Numerous other "dumb terminals" who's brand names I have long
forgotten.
- although for some reason I remember the Kimtron KT-7 being a popular
one - memory is they were dirt cheap at the time
- Eventually later models of DEC terminals, but the keyboard always
sucked and had those strange DEC private connectors on them, so I tried to
avoid them.
Clem
9:09 p.m.
New subject: Slashes (was: MS-DOS)
I started on ADM-1 (upper case only but they did have cursor addressing) and ASR-33
teletypes. I remember using all the backslash escapes to write my first C program while
Mike Muuss looked on in the UGL at Hopkins. Later we got HP terminals with upper and
lower case.
Hopkins had a KSR37 in the EE department with a greek type box on it even. It was stored
in a closet dubbed (obviously) “The KSR Room.” The pennywhistle modem I had lived
there. We used to place collect calls to the Pentagon TIP (we’d tell the operator we
were calling a computer and if it beeped it accepted the charges). Later they upgraded
the printing with a Diablo-ish (daisy wheel) printer. We had such printers over in
various other labs I had access to (Psych department, etc.).
Hopkins actually had one of the “braindamaged Hazeltines” (leave poor tilde alone) and a
few ADM-3’s and for some idiotic reason the department bought a couple of SWTPC
implementations of the TVTypewriterII which were just awful. Tektronix donated a bunch of
stuff to us so we ended up with both 4014-ish things and some real raster Tek graphics
terminals.
When I went over to BRL we primarily dealt with some VT52 clones which were preferred
because they put the control key next to the A which we liked. Eventually, we got the
Teletype 5620 the commercialization of the Blit/jerq DMD terminals. By the time I left
most of us had either Suns or SGIs on our desk though.
At home I had an ADM-3 followed by one of the VT-52 clones and also a ASR 37 that I got
surplus (It had a Rocky Flats property tag on it). When I moved to NJ I ditched them
all and just used at terminal emulator on my DOS PC-AT for the longest time. I actually
had a 9600 SLIP line and a router to the Ethernet in my house (I used one of the RU subnet
numbers).
From: TUHS [mailto:tuhs-bounces@minnie.tuhs.org] On Behalf Of Clem Cole
Sent: Friday, July 8, 2016 4:49 PM
To: Random832
Cc: TUHS main list
Subject: Re: [TUHS] Slashes (was: MS-DOS)
On Fri, Jul 8, 2016 at 2:23 PM, Random832 <random832(a)fastmail.com> wrote:
What terminals did you use, back in those days?
Hard copy (that I remember, but we have proven how poor memory can be):
* ASR/KSR 33s originally, we had one 37 but it was not on a system I used
* IBM console's printers Model numbers I've forgotten (but with an APL ball)
* Later Xerox DaisyWheel printers Model numbers I've forgotten
* Later DEC Decwritter I and II
Glass TTY (not complete but for a quick memory dump)
* Lear Siegler ADM3 (lots of them - many people made them as kits) and why we have hjkl as
the movement keys in vi - the arrows were embossed on those key tops
* PE "Fox" (CMU has a large lot deal and these became the standard there in the
late 70s)
* Triple Drip "Graphic Wonders" (man I miss these with a dedicated PDP11 and an
amazing keyboard --- best game platform I ever knew)
* Tektronix 401x series (just about all models of them) but 4014 was used the most
* DEC VT52
* Eventually, VT-100, PT-100 and a number of other VT-100 knock offs
* Eventually Tek 4025s (until the RT, one of the best keyboards - used in Magnolia BTW)
* Heathkit H19 (still have mine that I built)
* Eventually Wyse 100's, 99GTs and Wyse 60s -- later being the best Wyse (I still
have one)
* Ann Arbor Ambassador (my all time favorite - wish I still had one)
* Too many different graphics terminals to remember
* Numerous other "dumb terminals" who's brand names I have long forgotten.
* although for some reason I remember the Kimtron KT-7 being a popular one - memory is
they were dirt cheap at the time
* Eventually later models of DEC terminals, but the keyboard always sucked and had those
strange DEC private connectors on them, so I tried to avoid them.
Clem
9:16 p.m.
New subject: Slashes (was: MS-DOS)
Ron Natalie scripsit:
Hopkins actually had one of the “braindamaged
Hazeltines” (leave
poor tilde alone)
Not the terminal's fault. It was just implementing ASCII-63, in which
ESC was up by DEL, in the current position of ~. Other terminals did
ASCII-67, the current standard.
--
John Cowan http://www.ccil.org/~cowan cowan(a)ccil.org
A poetical purist named Cowan [that's me]
Once put the rest of us dowan. [on xml-dev]
"Your verse would be sweeter / If it only had metre
And rhymes that didn't force me to frowan." [overpacked line!] --Michael
Kay
9:45 p.m.
New subject: Slashes (was: MS-DOS)
That may have been the origin, but Hazeltine persisted even in later versions of the
terminal which implemented later ASCII and had a printing ~ at 126. That's why the
comment and associated code is in the terminal driver. In order to print a ~ you had to
send it twice.
-----Original Message-----
From: John Cowan [mailto:cowan@ccil.org] On Behalf Of John Cowan
Sent: Friday, July 8, 2016 5:17 PM
To: Ron Natalie
Cc: 'Clem Cole'; 'Random832'; 'TUHS main list'
Subject: Re: [TUHS] Slashes (was: MS-DOS)
Ron Natalie scripsit:
Hopkins actually had one of the “braindamaged
Hazeltines” (leave poor
tilde alone)
Not the terminal's fault. It was just implementing ASCII-63, in which ESC was up by
DEL, in the current position of ~. Other terminals did ASCII-67, the current standard.
--
John Cowan http://www.ccil.org/~cowan cowan(a)ccil.org
A poetical purist named Cowan [that's me]
Once put the rest of us dowan. [on xml-dev]
"Your verse would be sweeter / If it only had metre
And rhymes that didn't force me to frowan." [overpacked line!] --Michael
Kay
9 Jul
9 Jul
4:47 p.m.
New subject: Slashes (was: MS-DOS)
On Fri, 8 Jul 2016, Clem Cole wrote:
In those days many of us, including me, did (and still
do) refer to the
asterisk as "splat" and the exclamation point as "bang" from the
sound
made by them when they printed yellow oiled paper @ 10 cps from the
console TTY.
I still remember when the pipe command was "^" (pointy hat).
--
Dave Horsfall DTM (VK2KFU) "Those who don't understand security will
suffer."
5:03 p.m.
New subject: Slashes (was: MS-DOS)
Dave Horsfall scripsit:
I still remember when the pipe command was
"^" (pointy hat).
Thus forcing the rest of us to quote grep patterns like '^foo$' forever.
--
John Cowan http://www.ccil.org/~cowan cowan(a)ccil.org
An observable characteristic is not necessarily a functional requirement.
--John Hudson
5:21 p.m.
New subject: Slashes (was: MS-DOS)
On Jul 9, 2016, at 12:03 PM, John Cowan
<cowan(a)mercury.ccil.org> wrote:
Dave Horsfall scripsit:
It’s the $ that forces the quoting.
I still remember when the pipe command was
"^" (pointy hat).
Thus forcing the rest of us to quote grep patterns like '^foo$' forever.
10 Jul
10 Jul
2:38 p.m.
New subject: Slashes
Milo Velimirovic <milov(a)cs.uwlax.edu> writes:
Not in POSIX-compatible sh (works in tcsh even):
$ echo see$
see$
(Neither does ^ need quoting there.)
--
Christian Neukirchen <chneukirchen(a)gmail.com> http://chneukirchen.org
On Jul 9,
2016, at 12:03 PM, John Cowan <cowan(a)mercury.ccil.org> wrote:
Dave Horsfall scripsit:
It’s the $ that forces the quoting. I still remember when the pipe command was
"^" (pointy hat).
Thus forcing the rest of us to quote grep patterns like '^foo$' forever.
2856
days inactive
2858
days old
15 comments
10 participants
participants (10)
-
Christian Neukirchen -
Clem Cole -
Dave Horsfall -
John Cowan -
Milo Velimirovic -
Nemo -
Random832 -
Ron Natalie -
Steffen Nurpmeso -
Steve Nickolas