On 21-Apr-24 11:42, Peter Moylan wrote:
> On 21/04/24 19:19, Phil wrote:
>> On 21/04/2024 06:53, Peter Moylan wrote:
>>> On 20/04/24 15:46, Ruud Harmsen wrote:
>>>> Fri, 19 Apr 2024 23:12:44 +0200: Bertel Lund Hansen
>>>> <gadekryds@lundhansen.dk> scribeva:
>>>>
>>>>> Ruud Harmsen wrote:
>>>>>
>>>>>>> It started when the company AOL handed out huge numbers of floppy
>>>>>>> disks
>>>>>>> which enabled beginners to sign up for a Usenet account. (And
>>>>>>> possibly
>>>>>>> other services; I don't know that part of the story.) All of a
>>>>>>> sudden
>>>>>>> lots of clueless newbies turned up in newsgroups, replying "Me
>>>>>>> too" to a
>>>>>>> lot of postings. As a result many of us started using "AOL" as an
>>>>>>> abbreviation for "Me too".
>>>>>>
>>>>>> Interesting. But strange. Never heard of this before,
>>>>>
>>>>> Do you know "ACK" with the same meaning, but another history?
>>>>
>>>> The ASCII sign "acknowledged"? Roger.
>>>
>>> I'm reasonably certain that ACK and NAK were used in comunications well
>>> before ASCII was invented, but now I can't find proof.
>>>
>>> Giving "ACK history" to Google results in pointers to the African Church
>>> of Kenya and to Assumption College Kilmore.
>>
>> ACK, but not NAK or ENQ, features in a long list going back to 1939 here:
>>
>> <http://www.railwaycodes.org.uk/features/telegraph.shtm>
>>
>> Some curiously specific codes in there:
>>   "BILTONG -- Arrange special collection today on account of Woodley and
>> Co.  of Cardiff and despatch by first train in refrigerator cars
>> consigned to Woodley and Co.  at the places named, the following
>> consignment of meat."
>
> I imagine that Woodley and Co were such frequent receivers of biltong
> that it made sense to have a special code for them.
>
Would biltong have needed refrigeration?
I'm not sure the weather in Cardiff gets hot enough to warrant the extra
costs (except on rare occasions).

--
Sam Plusnet

On 2024-04-21, Madhu wrote:

> * Adam Funk <7pfbfkxnem.ln2@news.ducksburg.com> :
> Wrote on Sat, 20 Apr 2024 01:44:55 +0100:
>
>>
>> I'd use "backtick" if it's paired around something, e.g.,
>>
>> FILES=`ls some/path`
>
> pro: type it unshifted, con: non-nestable

That's probably the reason (for the IDE recommendation). I'd forgotten
about nesting because I never nest command outputs (I spread them over
more variable assignments to make the scripts easier to read).

>> although my IDEs tell me to use
>>
>> FILES=$(ls some/path)
>>
>> instead (I can't remember why it's better); and "grave accent" or
>> "accent grave" (depending on whether I'm speaking/writing in English
>> or French) if it's going over a vowel.

--
Dear Ann [Landers]: if there's an enormous rash of necrophilia that
happens in the next year because of this song, please let me know.
99.9% of the rest of us know it's a funny song! ---Alice Cooper

On 2024-04-20, Mark Brader wrote:

> Garrett Wollman:
>>>>> [Some people] may not know that backtick is inexplicably
>>>>> called "GRAVE ACCENT" in the Unicode table.
>
> Peter Moylan:
>>> Backtick is a new one to me. Is that used anywhere?
>
> Adam Funk:
>> I'd use "backtick" if it's paired around something, e.g.,
>>
>> FILES=`ls some/path`
>
> As far as I'm considered that's a "grave" (rhymes with "Slav"),
> but the other name I've heard some people use is "backquote",
> not "backtick".

That too!

--
It's like a pair of eyes. You're looking at the umlaut, and it's
looking at you. ---David St. Hubbins

On 2024-04-21, Sam Plusnet wrote:

> On 21-Apr-24 11:42, Peter Moylan wrote:
>> On 21/04/24 19:19, Phil wrote:

>>> ACK, but not NAK or ENQ, features in a long list going back to 1939 here:
>>>
>>> <http://www.railwaycodes.org.uk/features/telegraph.shtm>
>>>
>>> Some curiously specific codes in there:
>>>   "BILTONG -- Arrange special collection today on account of Woodley and
>>> Co.  of Cardiff and despatch by first train in refrigerator cars
>>> consigned to Woodley and Co.  at the places named, the following
>>> consignment of meat."
>>
>> I imagine that Woodley and Co were such frequent receivers of biltong
>> that it made sense to have a special code for them.
>>
> Would biltong have needed refrigeration?
> I'm not sure the weather in Cardiff gets hot enough to warrant the extra
> costs (except on rare occasions).

I thought the historical point of such dried meat products was to make
them non-perishable. But in this case, maybe that's the codebook
author's little joke.

--
Now everybody's got advice they just keep on giving
Doesn't mean too much to me
Lots of people have to make believe they're living
Can't decide who they should be ---Boston

On 2024-04-21, Athel Cornish-Bowden wrote:

> On 2024-04-21 12:13:22 +0000, Sn!pe said:
>
>> Peter Moylan <peter@pmoylan.org.invalid> wrote:
>>
>>> On 20/04/24 15:06, lar3ryca wrote:
>>>> On 2024-04-19 20:04, Sn!pe wrote:
>>>
>>>>> Blame the BBC for the idiotic 'forward slash'
>>>>> (to distinguish it from 'backslash')
>>>
>>> Plus broadcasters in several other countries.
>>>
>>>> IBM [1]
>>>>
>>>>>> . period, full stop, dot
>>>>>>
>>>>>> ! exclamation mark, bang
>>>>>
>>>>> Also 'pling'.
>>>>
>>>> [1] acronym: (I Blame Microsoft)
>>>
>>> I was about to say the same. It was Microsoft who trained
>>> Windows users to think that the name of '\' is "slash".
>>
>> I wasn't aware of that; it's yet another crime to lay at their door.
>
> Even worse than inventing \ in the first place.

It's useful as an escape character.

--
SHALL WE PLAY A GAME?

On 21/04/2024 21:24, Adam Funk wrote:
> On 2024-04-21, Sam Plusnet wrote:
>
>> On 21-Apr-24 11:42, Peter Moylan wrote:
>>> On 21/04/24 19:19, Phil wrote:
>
>>>> ACK, but not NAK or ENQ, features in a long list going back to 1939 here:
>>>>
>>>> <http://www.railwaycodes.org.uk/features/telegraph.shtm>
>>>>
>>>> Some curiously specific codes in there:
>>>>   "BILTONG -- Arrange special collection today on account of Woodley and
>>>> Co.  of Cardiff and despatch by first train in refrigerator cars
>>>> consigned to Woodley and Co.  at the places named, the following
>>>> consignment of meat."
>>>
>>> I imagine that Woodley and Co were such frequent receivers of biltong
>>> that it made sense to have a special code for them.
>>>
>> Would biltong have needed refrigeration?
>> I'm not sure the weather in Cardiff gets hot enough to warrant the extra
>> costs (except on rare occasions).
>
> I thought the historical point of such dried meat products was to make
> them non-perishable. But in this case, maybe that's the codebook
> author's little joke.
>
>

Yes, that last, I think.
"H Woodley & Co, established in 1884, developed into the largest meat
business in Wales and continued as a dynamic chain of butcher's shops
until 1990." (Wales Online).

That code list was the first hit from a Google search on "telegraphy
ack", on the reasoning that some sort of ACK/NAK protocol might have
been used on teleprinters. It seems not, though.

One more from the railway code list. Right at the bottom, in a
supplementary list of Unofficial Codes:

GYFO -- Strongly urge all speed ('Get your finger out')

--
Phil B

On 2024-04-21 20:26:11 +0000, Adam Funk said:

> On 2024-04-21, Athel Cornish-Bowden wrote:
>
>> On 2024-04-21 12:13:22 +0000, Sn!pe said:
>>
>>> Peter Moylan <peter@pmoylan.org.invalid> wrote:
>>>
>>>> On 20/04/24 15:06, lar3ryca wrote:
>>>>> On 2024-04-19 20:04, Sn!pe wrote:
>>>>
>>>>>> Blame the BBC for the idiotic 'forward slash'
>>>>>> (to distinguish it from 'backslash')
>>>>
>>>> Plus broadcasters in several other countries.
>>>>
>>>>> IBM [1]
>>>>>
>>>>>>> . period, full stop, dot
>>>>>>>
>>>>>>> ! exclamation mark, bang
>>>>>>
>>>>>> Also 'pling'.
>>>>>
>>>>> [1] acronym: (I Blame Microsoft)
>>>>
>>>> I was about to say the same. It was Microsoft who trained
>>>> Windows users to think that the name of '\' is "slash".
>>>
>>> I wasn't aware of that; it's yet another crime to lay at their door.
>>
>> Even worse than inventing \ in the first place.
>
> It's useful as an escape character.

Actually, to be fair to Micro$oft -- though why would one want to be
fair to Micro$oft? -- LaTeX makes abundant use of \, and I think it has
since the beginning, i.e. since 1984. Maybe it was already in plain TeX.

--
Athel -- French and British, living in Marseilles for 37 years; mainly
in England until 1987.

Athel Cornish-Bowden <me@yahoo.com> wrote:

> On 2024-04-21 20:26:11 +0000, Adam Funk said:
>
> > On 2024-04-21, Athel Cornish-Bowden wrote:
> >
> >> On 2024-04-21 12:13:22 +0000, Sn!pe said:
> >>
> >>> Peter Moylan <peter@pmoylan.org.invalid> wrote:
> >>>
> >>>> On 20/04/24 15:06, lar3ryca wrote:
> >>>>> On 2024-04-19 20:04, Sn!pe wrote:
> >>>>
> >>>>>> Blame the BBC for the idiotic 'forward slash'
> >>>>>> (to distinguish it from 'backslash')
> >>>>
> >>>> Plus broadcasters in several other countries.
> >>>>
> >>>>> IBM [1]
> >>>>>
> >>>>>>> . period, full stop, dot
> >>>>>>>
> >>>>>>> ! exclamation mark, bang
> >>>>>>
> >>>>>> Also 'pling'.
> >>>>>
> >>>>> [1] acronym: (I Blame Microsoft)
> >>>>
> >>>> I was about to say the same. It was Microsoft who trained
> >>>> Windows users to think that the name of '\' is "slash".
> >>>
> >>> I wasn't aware of that; it's yet another crime to lay at their door.
> >>
> >> Even worse than inventing \ in the first place.
> >
> > It's useful as an escape character.
>
> Actually, to be fair to Micro$oft -- though why would one want to be
> fair to Micro$oft? -- LaTeX makes abundant use of \, and I think it has
> since the beginning, i.e. since 1984. Maybe it was already in plain TeX.

I got my first IBM PC XT in 1984 and MS DOS certainly used backslash
as an integral part at that time; I suspect that DOS predates LaTeX.

--
^Ï^. Sn!pe, PA, FIBS - Professional Crastinator

My pet rock Gordon just is.

Sn!pe wrote:

> Athel Cornish-Bowden <me@yahoo.com> wrote:

>> On 2024-04-21 20:26:11 +0000, Adam Funk said:
>>
>> > On 2024-04-21, Athel Cornish-Bowden wrote:
>> >
>> >> On 2024-04-21 12:13:22 +0000, Sn!pe said:
>> >>
>> >>> Peter Moylan <peter@pmoylan.org.invalid> wrote:
>> >>>
>> >>>> On 20/04/24 15:06, lar3ryca wrote:
>> >>>>> On 2024-04-19 20:04, Sn!pe wrote:
>> >>>>
>> >>>>>> Blame the BBC for the idiotic 'forward slash'
>> >>>>>> (to distinguish it from 'backslash')
>> >>>>
>> >>>> Plus broadcasters in several other countries.
>> >>>>
>> >>>>> IBM [1]
>> >>>>>
>> >>>>>>> . period, full stop, dot
>> >>>>>>>
>> >>>>>>> ! exclamation mark, bang
>> >>>>>>
>> >>>>>> Also 'pling'.
>> >>>>>
>> >>>>> [1] acronym: (I Blame Microsoft)
>> >>>>
>> >>>> I was about to say the same. It was Microsoft who trained
>> >>>> Windows users to think that the name of '\' is "slash".
>> >>>
>> >>> I wasn't aware of that; it's yet another crime to lay at their door.
>> >>
>> >> Even worse than inventing \ in the first place.
>> >
>> > It's useful as an escape character.
>>
>> Actually, to be fair to Micro$oft -- though why would one want to be
>> fair to Micro$oft? -- LaTeX makes abundant use of \, and I think it has
>> since the beginning, i.e. since 1984. Maybe it was already in plain TeX.

> I got my first IBM PC XT in 1984 and MS DOS certainly used backslash
> as an integral part at that time; I suspect that DOS predates LaTeX.

Plain TeX certainly uses the \ as an escape character, like LaTeX. The
first running version was in 1978.

https://www.ams.org/publications/dtp-tex-part-1.pdf

DOS was first released by Seattle Computing in 1980

https://en.wikipedia.org/wiki/Timeline_of_DOS_operating_systems

It had a lot of resemblances to CP/M, I'm told. I actually had a little
contact with CP/M c. 1982, but I don't remember what non-alphanumeric
characters it used or anything else about it.

--
Jerry Friedman

In article <acfbe47bb5d1f10bc5abb4ff0d0d2567@www.novabbs.com>,
jerryfriedman <jerry.friedman99@gmail.com> wrote:
>Garrett Wollman wrote:
>> The subject we are apparently discussing here is what dialectologists
>> (at least those who were presenting at the American Dialect Society's
>> annual meeting in January) refer to as "foreign /a/", and it is well
>> documented that foreign /a/ is realized as [A] (FATHER) in most
>> American dialects and [&] (TRAP) in RP and RP-influenced dialects;
>
>What interested me was that "locale" went the opposite way.

Did it, or is it just a BATH word? I have no idea, but borrowings
from French are often treated differently from other varieties of
"foreign", especially if they're older. OED dates the noun sense to
1731, which I believe predates the TRAP-BATH split.

-GAWollman

--
Garrett A. Wollman | "Act to avoid constraining the future; if you can,
wollman@bimajority.org| act to remove constraint from the future. This is
Opinions not shared by| a thing you can do, are able to do, to do together."
my employers. | - Graydon Saunders, _A Succession of Bad Days_ (2015)

In article <fe3cc0b6c6a5efcae02a6b034fc68e65@www.novabbs.com>,
jerryfriedman <jerry.friedman99@gmail.com> wrote:
>Sn!pe wrote:
>> I got my first IBM PC XT in 1984 and MS DOS certainly used backslash
>> as an integral part at that time; I suspect that DOS predates LaTeX.

PC-DOS did not have hierarchical directories until PC-DOS 2.0 was
released in 1983 to coincide with the introduction of the PC XT,
because a single directory was clearly going to be problem for
organizing a 10 MB hard drive. Hard drives on CP/M systems of the
time were partitioned into separate virtual disks, so CP/M never got
hierarchical file systems. (Likewise the DEC operating systems of the
seventies that inspired much of the user interface of both CP/M and
PC-DOS; if they had any directory support, it was one directory per
user, as also with IBM mainframe operating systems. Multics and Unix
were outliers in this respect.)

There was some dispute within Microsoft about whether DOS 2.0 should
follow Unix or DEC conventions, to the point that there was a
CONFIG.SYS setting SWITCHAR which would determine whether standard DOS
command-line utilities would use a '/' or a '-' to introduce options.
DOS itself supported either '/' or '\' as a directory separator but
the utilities would always output '\'. (Note that you needed the
switch character not to be '/' if you wanted to specify pathnames that
way on the command line. Most people did not bother to change it and
most third-party software didn't support it.)

>Plain TeX certainly uses the \ as an escape character, like LaTeX. The
>first running version was in 1978.

And C used the backslash as an escape character before that. I might
go so far as to say that meaning was well-established. (Many other
languages of the 1970s used different escape characters, or entirely
lacked one, so you can't say it was universal. Someone has presumably
done the research from the meetings of CBEMA committees to determine
when it was introduced, since it was already in ASCII-63 and they left
out plenty of other characters that would have been more useful at the
time, but if so I haven't read it.)

-GAWollman
(a PC-DOS 1.1 and 2.0 user)
--
Garrett A. Wollman | "Act to avoid constraining the future; if you can,
wollman@bimajority.org| act to remove constraint from the future. This is
Opinions not shared by| a thing you can do, are able to do, to do together."
my employers. | - Graydon Saunders, _A Succession of Bad Days_ (2015)

Garrett Wollman <wollman@hergotha.csail.mit.edu> wrote:

> In article <fe3cc0b6c6a5efcae02a6b034fc68e65@www.novabbs.com>,
> jerryfriedman <jerry.friedman99@gmail.com> wrote:
> >Sn!pe wrote:
> >> I got my first IBM PC XT in 1984 and MS DOS certainly used backslash
> >> as an integral part at that time; I suspect that DOS predates LaTeX.
>
> PC-DOS did not have hierarchical directories until PC-DOS 2.0 was
> released in 1983 to coincide with the introduction of the PC XT,
> because a single directory was clearly going to be problem for
> organizing a 10 MB hard drive. Hard drives on CP/M systems of the
> time were partitioned into separate virtual disks, so CP/M never got
> hierarchical file systems. (Likewise the DEC operating systems of the
> seventies that inspired much of the user interface of both CP/M and
> PC-DOS; if they had any directory support, it was one directory per
> user, as also with IBM mainframe operating systems. Multics and Unix
> were outliers in this respect.)
>
> There was some dispute within Microsoft about whether DOS 2.0 should
> follow Unix or DEC conventions, to the point that there was a
> CONFIG.SYS setting SWITCHAR which would determine whether standard DOS
> command-line utilities would use a '/' or a '-' to introduce options.
> DOS itself supported either '/' or '\' as a directory separator but
> the utilities would always output '\'. (Note that you needed the
> switch character not to be '/' if you wanted to specify pathnames that
> way on the command line. Most people did not bother to change it and
> most third-party software didn't support it.)
>
> >Plain TeX certainly uses the \ as an escape character, like LaTeX. The
> >first running version was in 1978.
>
> And C used the backslash as an escape character before that. I might
> go so far as to say that meaning was well-established. (Many other
> languages of the 1970s used different escape characters, or entirely
> lacked one, so you can't say it was universal. Someone has presumably
> done the research from the meetings of CBEMA committees to determine
> when it was introduced, since it was already in ASCII-63 and they left
> out plenty of other characters that would have been more useful at the
> time, but if so I haven't read it.)
>
> -GAWollman
> (a PC-DOS 1.1 and 2.0 user)
>

Very interesting, thank you. :)

--
^Ï^. Sn!pe, PA, FIBS - Professional Crastinator

My pet rock Gordon just is.

Garrett Wollman wrote:

> In article <acfbe47bb5d1f10bc5abb4ff0d0d2567@www.novabbs.com>,
> jerryfriedman <jerry.friedman99@gmail.com> wrote:
>>Garrett Wollman wrote:
>>> The subject we are apparently discussing here is what dialectologists
>>> (at least those who were presenting at the American Dialect Society's
>>> annual meeting in January) refer to as "foreign /a/", and it is well
>>> documented that foreign /a/ is realized as [A] (FATHER) in most
>>> American dialects and [&] (TRAP) in RP and RP-influenced dialects;
>>
>>What interested me was that "locale" went the opposite way.

> Did it, or is it just a BATH word? I have no idea, but borrowings
> from French are often treated differently from other varieties of
> "foreign", especially if they're older. OED dates the noun sense to
> 1731, which I believe predates the TRAP-BATH split.

I didn't think of that. Wikipedia says you're right.

https://en.wikipedia.org/wiki/Trap%E2%80%93bath_split

(However it says the lengthening of some BATH vowels started in the
17th century.)

--
Jerry Friedman

Sn!pe <snipeco.2@gmail.com> wrote:

> jerryfriedman <jerry.friedman99@gmail.com> wrote:
>
> > >> / slash, forward slash [which I find annoying]
> >
> > > Also 'solidus', 'stroke', 'oblique'.
> >
> > > Blame the BBC for the idiotic 'forward slash'
> > > (to distinguish it from 'backslash')
> > ..
> >
> > I don't know the history, but I doubt the BBC is to blame for the
> > use of "forward slash" in the U.S.
> >
> > I know two intelligent people, one in a fairly senior position
> > with my employer, who say they can't remember which slash is
> > forward. I'm inclined (sorry) to agree that if the two marks
> > were called "slash" and "backslash", fewer people would get
> > confused, but I don't know of any studies.
> >
>
> A simple aide-memoire for such people: slash is the one
> that is much easier for a right-handed person to write by hand.
> Backslash is just plain boodly awkward (thank you, Micro$oft).

But you you can't escape them, if you are a regular,

Jan

On 23/04/24 01:46, Garrett Wollman wrote:

> There was some dispute within Microsoft about whether DOS 2.0 should
> follow Unix or DEC conventions, to the point that there was a
> CONFIG.SYS setting SWITCHAR which would determine whether standard
> DOS command-line utilities would use a '/' or a '-' to introduce
> options. DOS itself supported either '/' or '\' as a directory
> separator but the utilities would always output '\'. (Note that you
> needed the switch character not to be '/' if you wanted to specify
> pathnames that way on the command line. Most people did not bother
> to change it and most third-party software didn't support it.)

Precisely. CP/M was influenced by DEC conventions (it even had a PIP
command to do file operations), and the DOS for the PC was just a
development from CP/M. Using '/' to specify command-line options was a
well-established convention by then. (Unix didn't do it that way, but
back then Unix wasn't well-known, and anyway it wasn't available for the
PC.)

There was no conflict with filename notation, because the file system
was a flat system with no subdirectories. (The disk was a small floppy
disk, remember, so you didn't have a lot of files anyway.) By the time
subdirectories were added to DOS, the '/' was already in use for other
things, so a new separator character had to be found. The backslash was
an obvious choice.

--
Peter Moylan http://www.pmoylan.org
Newcastle, NSW

On 2024-04-21, Phil wrote:

> On 21/04/2024 21:24, Adam Funk wrote:
>> On 2024-04-21, Sam Plusnet wrote:
>>
>>> On 21-Apr-24 11:42, Peter Moylan wrote:
>>>> On 21/04/24 19:19, Phil wrote:
>>
>>>>> ACK, but not NAK or ENQ, features in a long list going back to 1939 here:
>>>>>
>>>>> <http://www.railwaycodes.org.uk/features/telegraph.shtm>
>>>>>
>>>>> Some curiously specific codes in there:
>>>>>   "BILTONG -- Arrange special collection today on account of Woodley and
>>>>> Co.  of Cardiff and despatch by first train in refrigerator cars
>>>>> consigned to Woodley and Co.  at the places named, the following
>>>>> consignment of meat."
>>>>
>>>> I imagine that Woodley and Co were such frequent receivers of biltong
>>>> that it made sense to have a special code for them.
>>>>
>>> Would biltong have needed refrigeration?
>>> I'm not sure the weather in Cardiff gets hot enough to warrant the extra
>>> costs (except on rare occasions).
>>
>> I thought the historical point of such dried meat products was to make
>> them non-perishable. But in this case, maybe that's the codebook
>> author's little joke.
>>
>>
>
> Yes, that last, I think.
> "H Woodley & Co, established in 1884, developed into the largest meat
> business in Wales and continued as a dynamic chain of butcher's shops
> until 1990." (Wales Online).
>
> That code list was the first hit from a Google search on "telegraphy
> ack", on the reasoning that some sort of ACK/NAK protocol might have
> been used on teleprinters. It seems not, though.
>
> One more from the railway code list. Right at the bottom, in a
> supplementary list of Unofficial Codes:
>
> GYFO -- Strongly urge all speed ('Get your finger out')

Heh. I wonder if there are more inside jokes in the last that only
railway people get.

--
Men, there is no sacrifice greater than someone else's.
---Skipper

Bertel Lund Hansen suggested that ...
> Hibou wrote:
>
>> Le 19/04/2024 à 13:11, James Harris a écrit :
>>>
>>> AISI as humans we ought to  to be able to look at a particular
>>> punctuation character and 'know' whether it comes before or after
>>> another regardless of the context. Then an index would be usable whether
>>> the book happened to be about computing, classification theory, or
>>> whatever.
>>
>> I think that's a specialist requirement. Numbers have magnitude, so
>> order themselves naturally; we use the alphabet for sorting, too - but
>> there's no need for order if letters are used only for writing words.
>> Why should inverted commas, ticks, 'therefore' signs, pound signs,
>> hashes, and daggers come before or after each other? A big problem in
>> imposing an order on them is that there are an enormous number of them,
>> no-one uses more than a small subset, and so people are never going to
>> know where to put most of them.
>>
>> <https://en.wikibooks.org/wiki/Unicode/List_of_useful_symbols>
>
> I think that James Harris is mainly concerned about ASCII characters.
> I'd suggest just using the order of their code calue. Programmers at
> least will understand immediately.

A certain M Brader has shown these comments as significant:

"Things are getting too standard around here.
Time to innovate!"
-- Ian Darwin and David Keldsen

"If the standard says that [things] depend on the
phase of the moon, the programmer should be prepared
to look out the window as necessary."
-- Chris Torek

"The nice thing about standards is
that you have so many to choose from..."
-- Andrew Tanenbaum

"Oh what a tangled web we weave,
a literate geekiness to achieve."
-- Steve Summit

/dps "Have we answered James' RFP yet?"

--
"That's a good sort of hectic, innit?"

" Very much so, and I'd recommend the haggis wontons."
-njm

On Tue, 23 Apr 2024 09:44:21 +1000, Peter Moylan
<peter@pmoylan.org.invalid> wrote:

>On 23/04/24 01:46, Garrett Wollman wrote:
>
>> There was some dispute within Microsoft about whether DOS 2.0 should
>> follow Unix or DEC conventions, to the point that there was a
>> CONFIG.SYS setting SWITCHAR which would determine whether standard
>> DOS command-line utilities would use a '/' or a '-' to introduce
>> options. DOS itself supported either '/' or '\' as a directory
>> separator but the utilities would always output '\'. (Note that you
>> needed the switch character not to be '/' if you wanted to specify
>> pathnames that way on the command line. Most people did not bother
>> to change it and most third-party software didn't support it.)
>
>Precisely. CP/M was influenced by DEC conventions (it even had a PIP
>command to do file operations), and the DOS for the PC was just a
>development from CP/M. Using '/' to specify command-line options was a
>well-established convention by then. (Unix didn't do it that way, but
>back then Unix wasn't well-known, and anyway it wasn't available for the
>PC.)

I don't remember hearing of a DEC connection with CP/M.

I do remember something like this, from Google -
Windows is actually, in a complicated historical way, a descendent
of VMS in roughly the same way Linux is descended from UNIX: a
reimplementation of the core concepts.

Elsewhere, there was mention in particular of VAX =>Windows NT,
and NT was what I vaguely remembered hearing.

I computed on DEC-10s for 8-10 years, then on VAXs for 20+.
Logically, it seems like I must have used both systems for a
couple of overlapping years, but I can't remember that at all.

I also have not yet recalled using subdirectories on either,
though I think that the VAX must have had them.

I know that I found it convenient to have multiple accounts to
keep things separate. I also used a home-made STOrage program
that concatenated text files, with extra comments -- those
helped me keep a clean directory. Also, I could send a .STO
file to the printer and each separate original text would start
a page with the file name, original date, and the comment. (If I
didn't adapt that program to the VAX, maybe it was because subfiles
were available.)

For quite a while, I knew how to reSTOre those files to disk
with their original dates, but then the DEC operating system
made that command privileged, or hid it, or some such, and I
lost the ability.

For a while, some version of Windows had the ability to
put comments that would show up in directories alongside the
file names. I remember finding it a bit awkward, and not too
useful; and then the option disappeared (I think).

>
>There was no conflict with filename notation, because the file system
>was a flat system with no subdirectories. (The disk was a small floppy
>disk, remember, so you didn't have a lot of files anyway.) By the time
>subdirectories were added to DOS, the '/' was already in use for other
>things, so a new separator character had to be found. The backslash was
>an obvious choice.

--
Rich Ulrich

On 23/04/24 13:04, Rich Ulrich wrote:
> On Tue, 23 Apr 2024 09:44:21 +1000, Peter Moylan
> <peter@pmoylan.org.invalid> wrote:

>> Precisely. CP/M was influenced by DEC conventions (it even had a
>> PIP command to do file operations), and the DOS for the PC was
>> just a development from CP/M. Using '/' to specify command-line
>> options was a well-established convention by then. (Unix didn't do
>> it that way, but back then Unix wasn't well-known, and anyway it
>> wasn't available for the PC.)
>
> I don't remember hearing of a DEC connection with CP/M.
>
> I do remember something like this, from Google - Windows is
> actually, in a complicated historical way, a descendent of VMS in
> roughly the same way Linux is descended from UNIX: a reimplementation
> of the core concepts.

Yes, that's all that I meant. The people who developed this stuff were
influenced by their education, including their self-education. That
latter point is important, because strongly motivated people in the
computer industry usually made sure that they had read up on a variety
of computer architectures and operating systems and languages, to see
what ideas were around.

Gary Kildall, who wrote CP/M, graduated in 1972, not long after the
PDP-11 was released. The PDP-11 had a massive influence on the ideas
about computer design, because its instruction set was noticeably
different -- and, especially, more regular -- from what you'd find in
other minicomputers or in mainframes. The software that came with the
PDP-11, especially the multitasking RSX-11D, was also innovative. That
had a lot of influence.

From the 1970s onward the education market was, I believe, strongly
dominated by IBM and by DEC, and that showed up as a clear difference in
designs by what one might call "IBM-influenced people" and
"DEC-influenced people".

This shows up especially strongly in microprocessor development. By
looking at the instruction set and processor architecture, you can
immediately tell which designers had an IBM-influenced background, and
which others had a DEC-influenced background.

> Elsewhere, there was mention in particular of VAX =>Windows NT, and
> NT was what I vaguely remembered hearing.

There were rumours at the time that Microsoft hired someone away from
DEC to head its Windows NT work, but I haven't been able to confirm this
with web searches.

> For a while, some version of Windows had the ability to put comments
> that would show up in directories alongside the file names. I
> remember finding it a bit awkward, and not too useful; and then the
> option disappeared (I think).

Files in OS/2 have something called "extended attributes". Initially one
of the biggest motivations for this was to create a difference between
"file name" and "title", in the days when we had file systems that only
allowed 8.3 names but GUI users wanted longer names. Those days are long
gone, but the extended attributes continue to be useful. For example,
files have a "file type" that is not the same as what comes after the
last dot in the name, and this supports an object-oriented desktop. Rexx
scripts have an extended attribute that is the (semi-)compiled version
of the script. A file's icon is also an extended attribute.

Some of this would have gone into NT initially, because NT and OS/2 were
the same operating system until the big fight between Microsoft and IBM,
but it seems that Microsoft killed off the object-orientation fairly
quickly, preferring to go for a GUI that looked familiar to users of
more primitive versions of Windows.

--
Peter Moylan http://www.pmoylan.org
Newcastle, NSW

On Tue, 23 Apr 2024 16:13:32 +1000
Peter Moylan <peter@pmoylan.org.invalid> wrote:

> On 23/04/24 13:04, Rich Ulrich wrote:
> > On Tue, 23 Apr 2024 09:44:21 +1000, Peter Moylan
> > <peter@pmoylan.org.invalid> wrote:
>
> >> Precisely. CP/M was influenced by DEC conventions (it even had a
[]
>
> > Elsewhere, there was mention in particular of VAX =>Windows NT, and
> > NT was what I vaguely remembered hearing.
>
> There were rumours at the time that Microsoft hired someone away from
> DEC to head its Windows NT work, but I haven't been able to confirm this
> with web searches.

Use Wikipedia!
https://en.wikipedia.org/wiki/Windows_nt#Naming

Dave Cutler

VMS --> WNT (shift letters by 1)

--
Bah, and indeed Humbug.

On 23/04/24 17:38, Kerr-Mudd, John wrote:
> On Tue, 23 Apr 2024 16:13:32 +1000 Peter Moylan
> <peter@pmoylan.org.invalid> wrote:
>
>> On 23/04/24 13:04, Rich Ulrich wrote:
>>> On Tue, 23 Apr 2024 09:44:21 +1000, Peter Moylan
>>> <peter@pmoylan.org.invalid> wrote:
>>
>>>> Precisely. CP/M was influenced by DEC conventions (it even had
>>>> a
> []
>>
>>> Elsewhere, there was mention in particular of VAX =>Windows NT,
>>> and NT was what I vaguely remembered hearing.
>>
>> There were rumours at the time that Microsoft hired someone away
>> from DEC to head its Windows NT work, but I haven't been able to
>> confirm this with web searches.
>
> Use Wikipedia! https://en.wikipedia.org/wiki/Windows_nt#Naming
>
> Dave Cutler

Thanks. I missed seeing that detail.

> VMS --> WNT (shift letters by 1)

That reminds me of the HAL (hardware abstraction layer) in 2001. (But
Arthur C Clarke has denied that that name was derived from IBM.)

--
Peter Moylan http://www.pmoylan.org
Newcastle, NSW

Rich Ulrich submitted this idea :
> On Tue, 23 Apr 2024 09:44:21 +1000, Peter Moylan
> <peter@pmoylan.org.invalid> wrote:
>
>> On 23/04/24 01:46, Garrett Wollman wrote:
>>
>>> There was some dispute within Microsoft about whether DOS 2.0 should
>>> follow Unix or DEC conventions, to the point that there was a
>>> CONFIG.SYS setting SWITCHAR which would determine whether standard
>>> DOS command-line utilities would use a '/' or a '-' to introduce
>>> options. DOS itself supported either '/' or '\' as a directory
>>> separator but the utilities would always output '\'. (Note that you
>>> needed the switch character not to be '/' if you wanted to specify
>>> pathnames that way on the command line. Most people did not bother
>>> to change it and most third-party software didn't support it.)
>>
>> Precisely. CP/M was influenced by DEC conventions (it even had a PIP
>> command to do file operations), and the DOS for the PC was just a
>> development from CP/M. Using '/' to specify command-line options was a
>> well-established convention by then. (Unix didn't do it that way, but
>> back then Unix wasn't well-known, and anyway it wasn't available for the
>> PC.)
>
> I don't remember hearing of a DEC connection with CP/M.

I worked with PS/8 in about 1970; I don't remember which of the big
machines inspired PS/8, but it was a huge upgrade from the PDP-8
operating system. I think PS/8 and it's inspiration may have been the
model for CP/M, which I had a brief exposure to some where around 1980.

> I do remember something like this, from Google -
> Windows is actually, in a complicated historical way, a descendent
> of VMS in roughly the same way Linux is descended from UNIX: a
> reimplementation of the core concepts.

Especially the file system, which was quite different from the FAT
designs used in DOS and continued forward on diskettes and later
thumbdrives.

> Elsewhere, there was mention in particular of VAX =>Windows NT,
> and NT was what I vaguely remembered hearing.

See previous comment.

> I computed on DEC-10s for 8-10 years, then on VAXs for 20+.
> Logically, it seems like I must have used both systems for a
> couple of overlapping years, but I can't remember that at all.
>
> I also have not yet recalled using subdirectories on either,
> though I think that the VAX must have had them.

Yes, at least in the 1990s on VMS, although the VAXen I used in the
'80s were running BSD 4.1 or 4.2. Both OSes were multiuser. Some
DEC10s were multiuser running timesharing OSes. In the early '70s,
Tymshare was doing that in Portland.

> I know that I found it convenient to have multiple accounts to
> keep things separate. I also used a home-made STOrage program
> that concatenated text files, with extra comments -- those
> helped me keep a clean directory. Also, I could send a .STO
> file to the printer and each separate original text would start
> a page with the file name, original date, and the comment. (If I
> didn't adapt that program to the VAX, maybe it was because subfiles
> were available.)
>
> For quite a while, I knew how to reSTOre those files to disk
> with their original dates, but then the DEC operating system
> made that command privileged, or hid it, or some such, and I
> lost the ability.
>
> For a while, some version of Windows had the ability to
> put comments that would show up in directories alongside the
> file names. I remember finding it a bit awkward, and not too
> useful; and then the option disappeared (I think).

I think the low-level support remains, and sometimes get used for other
purposes; I'm not digging out my NTFS reference, but there was a field
called something like "extra" (well, something like). I don't remember
such a field in FAT.

>> There was no conflict with filename notation, because the file system
>> was a flat system with no subdirectories. (The disk was a small floppy
>> disk, remember, so you didn't have a lot of files anyway.) By the time
>> subdirectories were added to DOS, the '/' was already in use for other
>> things, so a new separator character had to be found. The backslash was
>> an obvious choice.

/dps

--
But happiness cannot be pursued; it must ensue. One must have a reason
to 'be happy.'"
Viktor Frankl

Watch this space, where Peter Moylan advised that...
> On 23/04/24 13:04, Rich Ulrich wrote:
>> On Tue, 23 Apr 2024 09:44:21 +1000, Peter Moylan
>> <peter@pmoylan.org.invalid> wrote:
>
>>> Precisely. CP/M was influenced by DEC conventions (it even had a
>>> PIP command to do file operations), and the DOS for the PC was
>>> just a development from CP/M. Using '/' to specify command-line
>>> options was a well-established convention by then. (Unix didn't do
>>> it that way, but back then Unix wasn't well-known, and anyway it
>>> wasn't available for the PC.)
>>
>> I don't remember hearing of a DEC connection with CP/M.
>>
>> I do remember something like this, from Google - Windows is
>> actually, in a complicated historical way, a descendent of VMS in
>> roughly the same way Linux is descended from UNIX: a reimplementation
>> of the core concepts.
>
> Yes, that's all that I meant. The people who developed this stuff were
> influenced by their education, including their self-education. That
> latter point is important, because strongly motivated people in the
> computer industry usually made sure that they had read up on a variety
> of computer architectures and operating systems and languages, to see
> what ideas were around.
>
> Gary Kildall, who wrote CP/M, graduated in 1972, not long after the
> PDP-11 was released. The PDP-11 had a massive influence on the ideas
> about computer design, because its instruction set was noticeably
> different -- and, especially, more regular -- from what you'd find in
> other minicomputers or in mainframes. The software that came with the
> PDP-11, especially the multitasking RSX-11D, was also innovative. That
> had a lot of influence.
>
> From the 1970s onward the education market was, I believe, strongly
> dominated by IBM and by DEC, and that showed up as a clear difference in
> designs by what one might call "IBM-influenced people" and
> "DEC-influenced people".
>
> This shows up especially strongly in microprocessor development. By
> looking at the instruction set and processor architecture, you can
> immediately tell which designers had an IBM-influenced background, and
> which others had a DEC-influenced background.
>
>> Elsewhere, there was mention in particular of VAX =>Windows NT, and
>> NT was what I vaguely remembered hearing.

> There were rumours at the time that Microsoft hired someone away from
> DEC to head its Windows NT work, but I haven't been able to confirm this
> with web searches.

The name is out there. I think it is on some of the documents. IIRC,
the first name was Dave, but it will take a few more ticks for me to
come up with the last name. He was in charge of the filesystem.

>> For a while, some version of Windows had the ability to put comments
>> that would show up in directories alongside the file names. I
>> remember finding it a bit awkward, and not too useful; and then the
>> option disappeared (I think).
>
> Files in OS/2 have something called "extended attributes". Initially one
> of the biggest motivations for this was to create a difference between
> "file name" and "title", in the days when we had file systems that only
> allowed 8.3 names but GUI users wanted longer names. Those days are long
> gone, but the extended attributes continue to be useful. For example,
> files have a "file type" that is not the same as what comes after the
> last dot in the name, and this supports an object-oriented desktop. Rexx
> scripts have an extended attribute that is the (semi-)compiled version
> of the script. A file's icon is also an extended attribute.

ISHRA

>
> Some of this would have gone into NT initially, because NT and OS/2 were
> the same operating system until the big fight between Microsoft and IBM,
> but it seems that Microsoft killed off the object-orientation fairly
> quickly, preferring to go for a GUI that looked familiar to users of
> more primitive versions of Windows.

I miss the Lisa's tear-off-a-sheet metaphor for starting a new
document.

/dps

--
Killing a mouse was hardly a Nobel Prize-worthy exercise, and Lawrence
went apopleptic when he learned a lousy rodent had peed away all his
precious heavy water.
_The Disappearing Spoon_, Sam Kean

On Tuesday, Kerr-Mudd, John queried:
> On Tue, 23 Apr 2024 16:13:32 +1000
> Peter Moylan <peter@pmoylan.org.invalid> wrote:
>
>> On 23/04/24 13:04, Rich Ulrich wrote:
>>> On Tue, 23 Apr 2024 09:44:21 +1000, Peter Moylan
>>> <peter@pmoylan.org.invalid> wrote:
>>
>>>> Precisely. CP/M was influenced by DEC conventions (it even had a []
>>> Elsewhere, there was mention in particular of VAX =>Windows NT, and
>>> NT was what I vaguely remembered hearing.
>>
>> There were rumours at the time that Microsoft hired someone away from
>> DEC to head its Windows NT work, but I haven't been able to confirm this
>> with web searches.
>
> Use Wikipedia!
> https://en.wikipedia.org/wiki/Windows_nt#Naming
>
> Dave Cutler
>
> VMS --> WNT (shift letters by 1)

ISHRA A!
(again, that is)

/dps

--
Ieri, oggi, domani

On 24/04/24 11:10, Snidely wrote:
> Watch this space, where Peter Moylan advised that...
>> On 23/04/24 13:04, Rich Ulrich wrote:

>>> For a while, some version of Windows had the ability to put
>>> comments that would show up in directories alongside the file
>>> names. I remember finding it a bit awkward, and not too useful;
>>> and then the option disappeared (I think).
>>
>> Files in OS/2 have something called "extended attributes".
>> Initially one of the biggest motivations for this was to create a
>> difference between "file name" and "title", in the days when we had
>> file systems that only allowed 8.3 names but GUI users wanted
>> longer names. Those days are long gone, but the extended attributes
>> continue to be useful. For example, files have a "file type" that
>> is not the same as what comes after the last dot in the name, and
>> this supports an object-oriented desktop. Rexx scripts have an
>> extended attribute that is the (semi-)compiled version of the
>> script. A file's icon is also an extended attribute.
>
> ISHRA

Sorry, that one whooshed me.

>> Some of this would have gone into NT initially, because NT and OS/2
>> were the same operating system until the big fight between
>> Microsoft and IBM, but it seems that Microsoft killed off the
>> object-orientation fairly quickly, preferring to go for a GUI that
>> looked familiar to users of more primitive versions of Windows.
>
> I miss the Lisa's tear-off-a-sheet metaphor for starting a new
> document.

The current version of OS/2, ArcaOS 5.1.0, still comes with a
"Templates" folder on the desktop. The templates are somewhat similar to
that tear-off-a-sheet metaphor. For example, to install a new printer
driver you drag a copy of the "Printer" template, and to create a new
folder you can drag from the "Folder" template. That latter case is
especially useful if you have created a new folder class that inherits
from the "Folder" object but adds some special features.

Those templates have been present in OS/2 for a long time, probably
since Warp 3 (which was released at about the same time as Windows NT),
but the impression I have is that this feature is not much used. Perhaps
this is because, with the shrinking market for OS/2, the users that are
left are the sort of people who work a lot with command-line shells
rather than GUI shells.

The main exceptions are the Color Palettes and the Font Palettes. For
example, most of the software that I write comes with a configuration
notebook. If people don't like my choices of font in that notebook, they
can drag a new font from a font palette and drop it onto a notebook
page. That is one example where GUI working is more convenient than
working in a command-line shell.

--
Peter Moylan http://www.pmoylan.org
Newcastle, NSW