[TUHS] Re: ed: multiple addresses (with semicolons)

[TUHS] Re: ed: multiple addresses (with semicolons)

[Douglas McIlroy]

> More was I curious about the documentation of address chains in books.

It was even discussed in Lomutu and Lomuto, "A Unix Primer", a pleasant
book whose level is accurately described in the title.

Doug

[TUHS] Re: ed: multiple addresses (with semicolons)

[markus schnalke]

Hoi.

[2022-07-11 19:30] Douglas McIlroy <douglas.mcilroy@dartmouth.edu>
>
> > More was I curious about the documentation of address chains in books.
> 
> It was even discussed in Lomutu and Lomuto, "A Unix Primer", a pleasant
> book whose level is accurately described in the title.

Thanks for the recommendation of this book. I now have a copy and
am looking forward to reading it. It must have been a great book
for users at the time (mainly covers editing and formatting) and
now it looks to be a great book for someone interested in
understanding how the experience of using Unix must have been to
normal users back then.

Although the semicolon address separator is explained in section
10.2.1 (page 143), I couldn't find address chains or supplying
more addresses than necessary to commands mentioned in the book.
Of course, as I haven't fully read it yet, I could have missed it,
although I checked the relevant pages.

Nonetheless, I'm happy with this new (old) book, which I luckily
received in nearly mint condition. :-)


meillo

[TUHS] Re: ed: f command (facts vs. file)

[markus schnalke]

Hoi.

[2022-07-18 12:52] markus schnalke <meillo@marmaro.de>
> [2022-07-11 19:30] Douglas McIlroy <douglas.mcilroy@dartmouth.edu>
> > 
> > Lomutu and Lomuto, "A Unix Primer"
> 
> Thanks for the recommendation of this book. I now have a copy and
> am looking forward to reading it.

Now I enjoy reading through the book. Thereby I came across this
piece of information (section 7.3.3, p. 87):

	(Historically, f stands for ``facts'', but most people
	prefer to think of it as an abbreviation for ``file''.)

Why ``facts''?


The manpage in 1st Edition does not list an f command at all.

The manpage in 3rd Edition explains f as ``the filename command''.

Assember and C sources of various editions all talk about filename
and none about facts.


Neither does ``facts'' seem to come from qed.

I couldn't find an f command there at all.
https://www.multicians.org/mspm/bx-9-06.681115.qed-editor.pdf
http://www.bitsavers.org/pdf/honeywell/multics/swenson/6906.multics-condensed-guide.pdf


What's the background for the comment about ``facts''?


meillo

[TUHS] Re: ed: f command (facts vs. file)

[Phil Budne]

> What's the background for the comment about ``facts''?

https://www.bell-labs.com/usr/dmr/www/qedman.pdf
(for DMR's QED for GE-TSS on the GE 635):

    _History of QED_
    The original QED was implemented at the University of
    California, Berkeley [4]. Substantially redesigned versions
    were written by the second author (KLT) for the CTSS system
    at MIT [1] and in BCPL for MULTICS. The latter version has
    also been available under GE-TSS using I/O routines supplied
    by A. W. Winikoff.

    _This version of QED_
    The present incarnation of QED was implemented in GMAP by
    the first author (DMR). It offers noticeable improvements
    in speed, program size, and text packing density over the
    BCPL version, of which it is a direct descendant. New
    facilities include a redesigned Global command and a
    numerical capability

The F(acts) command appears on page 9 (pdf page 10)

    The F command causes QED to type out
    1) the number of words on QED's "free list",
    2) the highest memory location used for text storage, and
    3) the current core allocation.

    When the third number gets near 32K, take care not to exceed the core
    maximum.

[TUHS] Re: ed: f command (facts vs. file)

[markus schnalke]

Thanks, that explains it.


meillo


[2022-07-22 13:14] Phil Budne <phil@ultimate.com>
>
> > What's the background for the comment about ``facts''?
> 
> https://www.bell-labs.com/usr/dmr/www/qedman.pdf
> (for DMR's QED for GE-TSS on the GE 635):
> 
>     _History of QED_
>     The original QED was implemented at the University of
>     California, Berkeley [4]. Substantially redesigned versions
>     were written by the second author (KLT) for the CTSS system
>     at MIT [1] and in BCPL for MULTICS. The latter version has
>     also been available under GE-TSS using I/O routines supplied
>     by A. W. Winikoff.
> 
>     _This version of QED_
>     The present incarnation of QED was implemented in GMAP by
>     the first author (DMR). It offers noticeable improvements
>     in speed, program size, and text packing density over the
>     BCPL version, of which it is a direct descendant. New
>     facilities include a redesigned Global command and a
>     numerical capability
> 
> The F(acts) command appears on page 9 (pdf page 10)
> 
>     The F command causes QED to type out
>     1) the number of words on QED's "free list",
>     2) the highest memory location used for text storage, and
>     3) the current core allocation.
> 
>     When the third number gets near 32K, take care not to exceed the core
>     maximum.
>