Re: Typography

Tim Streater <timstreater@xxxxxxxxxxxx> wrote:

real-address-in-sig@xxxxxxxxxxxxxxx (Rowland McDonnell) wrote:

Tim Streater <timstreater@xxxxxxxxxxxx> wrote:

real-address-in-sig@xxxxxxxxxxxxxxx (Rowland McDonnell) wrote:

Tim Streater <tim.streater@xxxxxxxxxxxx> wrote:

I think there was one kicking around but it wasn't helpful at all.

Hmm. Sounds more to me as if there wasn't a straightforward LaTeX
manual, 'cos if Lamport's book had been there (the one I'm thinking of),
you would have found it useful - especially if you were resorting to
using a homebrew macro set with its own \chapter command.

The book I'm remembering (and this was a long time ago) was a sort of
loose-leaf spiral-bound job with a blue-ish cover.

In which case, that was probably either Lamport's book and it would have
served you very well if you'd read it; or it was Knuth's book and you
have to be a studious hacker to get anything much out of it.

I rather think it was Knuth's book. What you say chimes with my
recollection of the content.

Righto - Knuth would not have been a lot of use if you were wanting to
learn to use LaTeX. Knuth's The TeXbook is the book you want to read if
you want to create LaTeX, not use it.

Can you recall if the cartoon lion was in modern or classical clothing?

Don't even recall the lion, sorry.

No worries - just an identifying feature you might have recalled, is
all. In Knuth's book, the cartoon lion is in classical clothing; in
modern clothing in Lamport's book. He's on the front cover, y'see.

[snip]

Depends how you define useful. I have simple tastes.

So do I, I assure you. That's why I find very stripped-down text
editors really annoying.

And so they are. I only used notepad because I had to.

I've used Notepad and concluded that almost anything else would be
preferable.

When we started
using unix at SLAC in 1990 or so,

*Started*? Good grief. That late?

At physics labs they had two main computing requirements:

1) Off-line analysis - giant programs written in FORTRAN. For that they
needed well-known, well understood compilers, and CPU cycles. In those
days, that meant CDC 6600, 7600, followed by IBM 370/168, 3090, and CRAY.

Uhuh.

2) On-line control systems. At the time I left SLAC, that pretty much
meant the VAX, which had a well-understood, well-documented OS, and most
particularly an OS with a very good real-time response. The physicists
wanted to know what was going to be the latency between an interrupt
from their equipment, and the on-line control machine starting to
process it. At the time (don't know the situation now), that meant VMS
and not unix.

Righto. Well, well, well.

there was this great debate about
which editor to use. In those days I expected to have to *learn* an
editor, so I didn't want to learn one that was not going to be the local
standard.

Strange. There never would have been one standard - obviously. The
sensible thing is to learn to use one editor and stick with it.

My experience up to then was that one had to spend oodles of time
learning the editor. And that very often you had to consult the experts.
I didn't want to learn one only to have another adopted as the standard.

But why not just stick with whatever you'd learnt, rather than use
whatever had been decreed as `standard'? I mean, that's basically what
you did, isn't it?

Up til then I hadn't really used the Mac enough to appreciate that the
old command line stuff had had its day.

It hasn't, not completely.

And here I am nearly 20 years
later still arguing with guys about wsywig.

Yeah, but hang on a bit - after nearly 20 years of pure wysiwyg, Macs
finally got a command line interface with OS X. And to my mind, it's a
step forward, not backwards. There are some things which you need a
command line for.

Reminds me of the early 70s
at CERN when we had an on-line editing system using *gasp* video
terminals. Then it was punched cards that fogeys were reluctant to stop
using. I remember one guy with his large tea-lady trolley with some 15
or so card trays on it (2000 cards/tray). Even with a CDC card-reader at
1200 cards/min, it still used to take 25 minutes just to read his job in.

I knew a fruitcake in the 1980s who kept all his software backed up on
paper tape, stored in a filing cabinet. He got sniggered at behind his
back. And then one day the lab suffered a nasty flood. Everybody who
didn't have an offsite backup (i.e., everyone) was in trouble - all the
magnetic media was history. D. didn't care - he just dried out his
paper tapes carefully, ironed them flat, and restored all his software
over the course of a couple of weeks. Chunka-chunka-chunka-chunka goes
the tape reader...

SO as I had work to do I spent 5 minutes learning NotePad and
got on with that.

I've worked with similar text editors, and I find I can get more work
done more quickly and more accurately with text editors that have more -
well, I hate to say it, but `features'.

Of course. I expect features too. It's just that I expect the simple
stuff to be easy.

<shrug> It is with Emacs. With a tiny amount of practice, certain
finger motions end up drilled in to you. Not that I've used it for a
very long time, mind.

[snip]

I may look up more complex features but that's all. My expectation
is that, if I don't use it for 6 months, I may well forget where
everything is and have to find it again - but it'll be on a
sodding menu somewhere, won't it, eh?

Well, no, and anyway, what does the command do? Without a manual,
how can one possibly tell?

I remember what the command does.

How can you find out in the first place?

If I see as it might be "Search and Replace..." on a menu that might
give me a clue.

Well, yeah, that kind of thing, okay.

[snip]

I've used emacs twice (by accident, second time), and in each case
it took 20 mins for me to figure out how to quit it (the man pages
were as usual hopeless).

If you had taken the trouble to work through the Emacs tutorial, you
would have had a different experience entirely, I assure you.

Why should I want to use an editor whose simplest command, quiting, has
some dopey key-sequence rather than something standard?

Well, you might as well ask why you'd want to use a computer whose
simplest commands use some dopey key sequences like squiggle-Q or
require you to move your hands from the keyboard and grab a mouse (how
slow!) for the simplest commands, rather than something well-established
and standardised across all platforms like Emacs.

Because, funnily enough, squiggle-Q is what every Mac program uses. The
cross-platform issue is not one that interests me.

I put the point the way I did to make the point that there are other
ways of looking at it. Driving your text editor from the keyboard makes
some things work very very quickly indeed. Okay, okay doing a text
selection operation is something I do prefer to use a mouse for.

[snip]

Speed has never been a problem with PM, at least for what I have been
using it for.

You've never set it up to run as a server process doing thousands of
transactions an hour.

Correct. But then I've never needed to.

Thing is, you'd not notice speed as an issue unless you did do something
like that, surely?

Rowland.

--
Remove the animal for email address: rowland.mcdonnell@xxxxxxxxxxxxxxx
Sorry - the spam got to me
http://www.mag-uk.org http://www.bmf.co.uk
UK biker? Join MAG and the BMF and stop the Eurocrats banning biking
.