[9fans] off topic: troff book

73 views
Skip to first unread message

James A. Robinson

unread,
Dec 27, 2000, 2:42:24 PM12/27/00
to
Has anyone here read Unix Document Processing and Typesetting by
B. Srinivasan? If so, would you recommend it? The only other book I've
seen which talks about troff is The UNIX Programming Environment, and I'm
interested in seeing what others have to say about the troff/tbl/eqn/pic
typesetting environment.


Jim

John E. Gwyn

unread,
Jan 2, 2001, 12:44:35 PM1/2/01
to
"James A. Robinson" wrote:
> ... The only other book I've seen which talks about troff is The

> UNIX Programming Environment, and I'm interested in seeing what
> others have to say about the troff/tbl/eqn/pic typesetting
> environment.

There have been several books on the Documenter's WorkBench (DWB),
which is the name AT&T used for the separately licensed package of
troff and associated tools. You can license the latest version
(3.1 I think) although it's a bit pricey for personal use. There
were papers in the Bell Labs CSTR series describing most of these
tools, and the papers are available on-line. The current Research
version of most of those tools is bundled into Plan 9 Release 3,
also available under a free license on-line, and the Plan 9
documentation includes user guides for some of them (and UNIX-style
manual pages for all). The GNU project has some independently
developed troff-workalike tools ("groff" etc.) which are freely
available.

http://www.unipress.com/toolkit/dwb.html Lucent/Unipress DWB
http://plan9.bell-labs.com/plan9dist/ Bell Labs Plan 9
http://www.gnu.org/software/groff/groff.html GNU troff (groff)

I use DWB (2.0 plus local improvements) for my own technical
documentation. You need to be aware that it is meant to be used
primarily as a set of specialized programming languages, not via a
graphical user interface (although interactive drawing tools do
exist). In many ways it is similar to Knuth's TeX. Its main
advantage over nearly all GUI-based formatters (Word, etc.) is its
relative lack of presumptions about the way you want your document
formatted. For example, it does not insist on modeling all text
as characters packed into lines; you can place any character
anywhere on the page, including overstriking other characters.
The programmability is largely tamed via macro packages (much
like with TeX), most often the "MS" or "MM" packages, provided
with DWB. One of the things I like most about DWB is the ability
to easily create specialized preprocessors to build graphs, etc.
which can be done within Makefiles, for example.

Allan J. Heim

unread,
Jan 5, 2001, 4:48:45 AM1/5/01
to
> Has anyone here read Unix Document Processing and Typesetting by
> B. Srinivasan? If so, would you recommend it?

I'm sure it's fine. "Word Processing on the Unix System" (I don't have
the author's name handy) was also really good, but it's years out of
print. A look at Amazon reveals "Document Formatting and Typesetting on
the Unix System" by Narain Gehani, and I seem to recall that it was
useful.

Having been a troff typesetter in my youth, I'd ask you to consider
TeX/LaTeX as an alternative. It seems more complex to set up, but it's
very sophisticated typesetting software, and the community
(http://www.tug.org/) is very active. There's even a graphical front-end
for LaTex, called "LyX".

--
Allan J. Heim mailto:a...@netscape.com

Boyd Roberts

unread,
Jan 5, 2001, 10:12:18 AM1/5/01
to
From: Allan J. Heim <a...@netscape.com>

>
> Having been a troff typesetter in my youth, I'd ask you to consider
> TeX/LaTeX as an alternative. It seems more complex to set up, but it's
> very sophisticated typesetting software, and the community
> (http://www.tug.org/) is very active. There's even a graphical front-end
> for LaTex, called "LyX".

knuth had lost his mind [tex]. lamport was a screwhead [latex].

gimme raw troff anyday...


Steve Kilbane

unread,
Jan 6, 2001, 12:42:20 PM1/6/01
to
Boyd boyded thus:

> knuth had lost his mind [tex]. lamport was a screwhead [latex].
>
> gimme raw troff anyday...

Hold on. TeX and LaTeX have their good and bad points, but exactly
how did they screw up in ways that troff didn't? Off the top of my
head, the only thing that comes to mind is that troff would read
from stdin, while tex would insist on a named file, getting in the
way of pipelining. But that's more a style of use, rather than of
the systems themselves.

steve


Boyd Roberts

unread,
Jan 6, 2001, 1:15:18 PM1/6/01
to
From: Steve Kilbane <st...@whitecrow.demon.co.uk>

>
> Hold on. TeX and LaTeX have their good and bad points, but exactly
> how did they screw up in ways that troff didn't? Off the top of my
> head, the only thing that comes to mind is that troff would read
> from stdin, while tex would insist on a named file, getting in the
> way of pipelining. But that's more a style of use, rather than of
> the systems themselves.

they could read /dev/drum for all i care.

they are _ghastly_.

my 'lamport was a screwhead' comes from my two years at PRL.

lame defense, but i'm entitled to my opinon.

if tex was so great, where was the need for latex?


Jim Choate

unread,
Jan 6, 2001, 1:37:24 PM1/6/01
to

http://www.latex-project.org/intro.html

The relation is 'language definition' and 'compiler implimentation'.

I use Scientific Notebook which combines LaTeX and Maple V. Very cool for
a technical doodle pad. Strongly suggested for all students. less than
$100. I don't work for the company or receive any sort of payment, just my
opinion.

____________________________________________________________________

Before a larger group can see the virtue of an idea, a
smaller group must first understand it.

"Stranger Suns"
George Zebrowski

The Armadillo Group ,::////;::-. James Choate
Austin, Tx /:'///// ``::>/|/ rav...@ssz.com
www.ssz.com .', |||| `/( e\ 512-451-7087
-====~~mm-'`-```-mm --'-
--------------------------------------------------------------------


Steve Kilbane

unread,
Jan 7, 2001, 5:24:23 AM1/7/01
to
Boyd's case for [La]TeX being worse than troff:
> they are _ghastly_.

..which is somewhat less reasoned than I'd hoped for. Got anything
other than personal dislike?

> my 'lamport was a screwhead' comes from my two years at PRL.

It also comes across as an attack on Lamport, rather than on
LaTeX, so it still needs backing up.

> lame defense, but i'm entitled to my opinon.

Absolutely. You're welcome to prefer troff to *tex, but I'd
be really interested if you could justify that preference based
on the packages' corresponding merits. C'mon, you can do it. :-)

> if tex was so great, where was the need for latex?

Same reason you had troff -man. It's a macro language;
you provide macro packages to make life easier.

steve


rob pike

unread,
Jan 7, 2001, 10:34:23 AM1/7/01
to
I didn't switch to tex when it came out because it just seemed like
the same thing packaged differently. It didn't seem like an
improvement. As time has passed, I have come to understand that it
does equations better, which many people care about, and is free,
which many people care about. Those are its advantages. It also has
many disadvantages, including a screwy chatty user interface and a
serious misjudgement about how fonts should work, which saddles it
with very few fonts - until recently, only one - which is, in turn,
certainly the ugliest book font that ever got widespread use. Latex
extended this mistake into the same story about layout, so that today
90% of the technical books and papers in the world look exactly the
same: grey, unbalanced, and drab, like an oil painting done by a rank
amateur.

Because this at one level so unimportant, but at another is the entire
point of the exercise - why create a layout language that produces
only one layout; why create a font language if only one person
in the world can use it to make a font - it rankles.

-rob

Boyd Roberts

unread,
Jan 7, 2001, 6:52:22 PM1/7/01
to
>I didn't switch to tex when it came out because it just seemed like
>the same thing packaged differently. It didn't seem like an
>improvement. As time has passed, I have come to understand that it
>does equations better, which many people care about, and is free,
>which many people care about. Those are its advantages.

'bout sums it up from what i realised when i had to write a paper
using latex. what does it give me except another tedious interface
with a lot of formating noise? what value does it add?

after reading lamport's book i decided he didn't understand
the problem. hence, my earlier comment, fueled a bit by
the member or research staff / member of technical staff
woes (i was an MTS).


Luis Fernandes

unread,
Jan 8, 2001, 4:54:27 AM1/8/01
to
>>>>> "rob" == rob pike <r...@plan9.bell-labs.com> writes:

rob> [...lamentations of TeX...]

rob> Latex extended this mistake into the same story about
rob> layout, so that today 90% of the technical books and papers
rob> in the world look exactly the same: grey, unbalanced, and
rob> drab, like an oil painting done by a rank amateur.

Kinda like the look-and-feel of 9wm...

:) <-- please take special note of the flame-retardant smiley

John A. Murdie

unread,
Jan 8, 2001, 5:45:21 AM1/8/01
to

Rob Pike writes:
>I didn't switch to tex when it came out because it just seemed like
>the same thing packaged differently. ... It also has

>many disadvantages, including a screwy chatty user interface

I find this the most immediate unpleasantness of TeX.

>As time has passed, I have come to understand that it
>does equations better,

The LaTeX format (layout) has (macro) packages for a wider range of
equations, but I still find troff's eqn much easier to read and write.
In fact, I think the best part of the troff approach is the use of little
languages by preprocessor, and the worst thing about troff is its use of
interrupt-like traps. If you've written a troff macro package, I think
you'll know that more than expected care is necessary to get trap use right.

Some kind of (procedural markup?) programming language is necessary in
a typesetting system. I often wonder whether something like pic with
better text primitives, and galleying (pagination) functionality, would be
a better approach. Yes, I know, I should make time to experiment.

>Because this at one level so unimportant, but at another is the entire
>point of the exercise - why create a layout language that produces
>only one layout; why create a font language if only one person
>in the world can use it to make a font - it rankles.

Yes, though there are other TeX formats (layouts), such as CONTeXT,
LaTeX is almost universally used. Most effort seems to be put into
writing extra packages for use with the LaTeX format. There is some
use of MetaFont for specification of special symbols, but the TeX
world seems to be moving away from the use of bitmap fonts generated
from MetaFonts and towards the use of commercially-provided PostScript
or TrueType typefaces.

John A. Murdie
Department of Computer Science
University of York
England

Mark C. Otto

unread,
Jan 8, 2001, 10:22:25 AM1/8/01
to
I grew up with TeX but would agree with Rob on its pluses and minuses over
troff. A package that seems to build on both troff and TeX is lout,
http://snark.ptc.spbu.ru/~uwe/lout/lout.html. The language looks a lot like
troff, lout produces postscript output, and is written in ANSI C. I know it
would have to have significant advantages over troff, but it might fit well with
plan9.

Mark

John A. Murdie

unread,
Jan 8, 2001, 10:47:19 AM1/8/01
to

Mark C. Otto writes:
>A package that seems to build on both troff and TeX is lout,

I've followed lout almost since day one. It's very nicely written,
and I've never had any problems installing it on various Unix
derivatives. Try notating a diagram in its pic-equivalent language,
however, and I think you'll find that it's far less elegant. This
is a result of it using a general applicative language style instead
of a little language.

Douglas A. Gwyn

unread,
Jan 8, 2001, 11:18:56 AM1/8/01
to
rob pike wrote:
> - why create a layout language that produces
> only one layout; why create a font language if only one person
> in the world can use it to make a font - it rankles.

As I recall, Knuth started development of TeX and Metafont
because he was dissatisfied with the appearance of one edition
of TAOCP. I have never seen any evidence that he is familiar
with troff or even with Unix.

I will agree that TeX is capable of producing nicer
mathematical copy than eqn|troff with a reasonable amount of
effort. I personally don't think that repays the loss of
Knuth's time that could have been spent working on additional
volumes of TAOCP.

James A. Robinson

unread,
Jan 8, 2001, 12:35:20 PM1/8/01
to
> As I recall, Knuth started development of TeX and Metafont
> because he was dissatisfied with the appearance of one edition
> of TAOCP. I have never seen any evidence that he is familiar
> with troff or even with Unix.

He started working on what he calls a 'proto TeX' in 1977, and I believe
the PDP-11 showed up at Stanford in 1976. He states in the introduction
of TeX: The Program that he had many comments from Howard Trickey, who
later ported TeX to Unix at Stanford. I expect he could answer whether
or not Troff was known to folks at Stanford. I think this time period
was before the rewrite of troff to be device independent. That may have
something to do with Knuth's development of his own system.


Jim

Howard Trickey

unread,
Jan 8, 2001, 2:35:43 PM1/8/01
to
From: "James A. Robinson" <jim.ro...@stanford.edu>

> He started working on what he calls a 'proto TeX' in 1977, and I believe
> the PDP-11 showed up at Stanford in 1976. He states in the introduction
> of TeX: The Program that he had many comments from Howard Trickey, who
> later ported TeX to Unix at Stanford. I expect he could answer whether
> or not Troff was known to folks at Stanford

I didn't get to Stanford until 1980, well after the start of Tex. Stanford
had a bunch of Unix machines by then, and I'm sure, even at the time when
Knuth started Tex. So, though I don't know for sure, I find it hard to
believe that Knuth wasn't told about troff as soon as he made it known
within the department that he was going to work on typesetting.

Knuth didn't work on Unix, however. He worked exclusively on the 'SAIL'
system (running on a DEC-20), a rather strange system that had, among other
things, a special terminal with a TV display, a keyboard with a lot of
special characters (e.g., an 'alpha' key), and its own visual editor. There
was no C compiler for SAIL.

All I know about the start of Tex was that, as someone else noted, he was
shown a galley for a reprinting of one of his Art of Computer Programming
books, done with some 'new' computer typesetting system, and it offended his
sensibilites. He really wanted his books to look like 'old-time' math
textbooks, with their Modern-style fonts, and he was told by the publisher
that he couldn't get it. I think he was also offended by the bad spacing in
math formulas in the galley he was shown. So his motivation in doing Tex
was to get (a) something that ran on SAIL; (b) something that used fonts
like the books of his youth; (c) something that did a good job of spacing in
formulas. Troff at the time would have needed a lot of modification to meet
all of these requirements, and he vastly underestimated the amount of time
that the Tex/Metafont project would take him.

Knuth is a very parsimonious person, and I think many of the not-so-great
choices in Tex were due to that parsimony. For instance, he hated wasting
time, so he had Tex run with a chatty interface that let him edit the input
stream on the fly as errors were encountered, so that he wouldn't have to
start a run over. In implementation, he was forever squeezing things in
memory (admittedly, the computers in 1977 didn't have a lot of memory to
work with), so Tex77 only allowed single letter names for fonts. Etc. Knuth
is also a perfectionist: thus, he would prefer to have Tex fail utterly with
something so ugly that you have to fix it rather than allow it to output a
paragraph with too much interword spacing.

- Howard Trickey


Dan Cross

unread,
Jan 8, 2001, 3:23:33 PM1/8/01
to
Rob Pike wrote:
>> - why create a layout language that produces
>> only one layout;

By this, surely you mean LaTeX? Why create -ms, which
basically produces one layout?

>> why create a font language if only one person
>> in the world can use it to make a font - it rankles.

Hmmm..... I'm not sure that's fair. I suspect that more
people know how to use Metafont than perhaps know raw troff.
I could be wrong, perhaps.

btw- I kind of like the CM fonts; I think they're rather
nice looking. I like troff's output also, though.

Doug Gwyn writes:
>As I recall, Knuth started development of TeX and Metafont
>because he was dissatisfied with the appearance of one edition
>of TAOCP.

I believe that's the story in, ``Digital Typography.''

>I have never seen any evidence that he is familiar
>with troff or even with Unix.

...except for the fact that he sits in front of a Sun
workstation? :-)

Come, surely Knuth knows about troff. I don't know that
he was familiar with it at the time he embarked on the
TeX project, though.

>I will agree that TeX is capable of producing nicer
>mathematical copy than eqn|troff with a reasonable amount of
>effort. I personally don't think that repays the loss of
>Knuth's time that could have been spent working on additional
>volumes of TAOCP.

I don't know, he considered it important enough to spend
10 years on it. Remember: There Is No Royal Road to Programs,
and he seems to have done a pretty good job on ellipses
using that diamond shaped pen, eh? :-)

- Dan C.

Andy Newman

unread,
Jan 9, 2001, 4:38:51 AM1/9/01
to

(To those of us somewhat interesting in typesetting systems this has
been an interesting thread - thanks). I have a reprint of an article
written for the AMS in 1978 entitled "Mathematical Typography" (Bulletin
(New Series) of the AMS, vol 1, no. 2, pp. 337-372), it says "This article
is the Josiah Willam Gibbs Lecture given by Mr.Knuth in February, 1978."
The first sentence is,

"Mathematics books and journals do not look as beautiful
as they used to."

The lecture discusses some of the problems he and the AMS faced in
publishing, shows some different layout styles and launches into
an overview of the math problems in layout systems, line filling,
font and curve representations and anti-aliasing in raster systems.

At one point he states "...I know that TeX provides only small
refinements over what is available in other systems. Yet several
dozen small refinements add up to something that is important to me..."

The article has an extensive bibliography that includes references
to eqn and papers on "....some experiments in typesetting physics
journals with the Bell Labs system...". I.e, he knew.

Boyd Roberts

unread,
Jan 9, 2001, 5:18:31 AM1/9/01
to
i think doug is right. knuth wanted tex and metafont
'cos he didn't like the typography of TAOCP.


Steve Kilbane

unread,
Jan 10, 2001, 3:15:32 AM1/10/01
to
Rob:

> I didn't switch to tex when it came out because it just seemed like
> the same thing packaged differently. It didn't seem like an
> improvement.

I starting using troff and latex for exactly the same reason: that's
what was used to typeset documents in the place I was at, at the time.
First used troff, moved to a texhouse. Later moved back to a troff
place. In both cases, I rewrote local house styles in what I'd just
been using, and what I was most familiar with.

It's been a long time, so I don't remember any details, but I vaguely
recall that my primary experience was this: each made some things
much easier than the other, and other things much harder.

> It also has
> many disadvantages, including a screwy chatty user interface

I loathed the interface, too. I think troff's pipeline approach
is a much cleaner, clearer interface than tex's abort-and-ask
style. Similarly, preloaded packages seem bizarre. However, the
actual languages didn't seem to have much between them.

> and a
> serious misjudgement about how fonts should work, which saddles it
> with very few fonts - until recently, only one - which is, in turn,
> certainly the ugliest book font that ever got widespread use.

This surprises me, I have to admit. I've never sat down and tried
to add new fonts to either system. My experiences with troff were
that there were hardly any fonts; tex seemed to come with loads,
but I couldn't figure out how to load them. These might be an
artifact of the local installations, more than anything else.


> Because this at one level so unimportant, but at another is the entire
> point of the exercise - why create a layout language that produces
> only one layout;

I'll go along with this, for latex at least. I didn't enjoy creating
page styles from the ground up in troff, and reading latex styles is
a nightmare, so by and large, I've used the default ones. However, I've
had reason to do a fair amount of latex in standard fiction submission
format: fixed width font, double-spaced, no right justification, no
hyphenation, and emphasis via underlining. These are all exactly the
opposite of what tex considers to be "right", and it fights every step
of the way.

Bottom line, though: I've never yet found a typesetting system that
allows me to do everything I want, without blood and tears. It just
seems to be a matter of picking which things annoy you less.

steve


Douglas A. Gwyn

unread,
Jan 11, 2001, 4:50:36 AM1/11/01
to
Steve Kilbane wrote:

> ... My experiences with troff were
> that there were hardly any fonts; ...

troff does not supply fonts, merely makes available
whatever fonts are supported by the specified target
device. Note that there is no presumption that these
are raster fonts; they might be molded type metal.

sa...@bear.com

unread,
Jan 12, 2001, 4:32:00 AM1/12/01
to
In article <2001010820...@augusta.math.psu.edu>,
9f...@cse.psu.edu wrote:
> Rob Pike wrote:
[snipped]

>
> btw- I kind of like the CM fonts; I think they're rather
> nice looking. I like troff's output also, though.
>

I believe Knuth used some best printing-industry people,
and finest font designers around at that time.

> Doug Gwyn writes:
>
> >I have never seen any evidence that he is familiar
> >with troff or even with Unix.
>
> ...except for the fact that he sits in front of a Sun
> workstation? :-)
>
> Come, surely Knuth knows about troff. I don't know that
> he was familiar with it at the time he embarked on the
> TeX project, though.
>

No, that is not true. Knuth felt that troff was burdened with
couple of generations' legacy at Bell Labs, and he wanted to
start with a clean slate without fear of hurting any Bell Labs
colleague's feeling (about troff).

> >I will agree that TeX is capable of producing nicer
> >mathematical copy than eqn|troff with a reasonable amount of
> >effort. I personally don't think that repays the loss of
> >Knuth's time that could have been spent working on additional
> >volumes of TAOCP.
>
> I don't know, he considered it important enough to spend
> 10 years on it. Remember: There Is No Royal Road to Programs,
> and he seems to have done a pretty good job on ellipses
> using that diamond shaped pen, eh? :-)
>

He was very dissatisfied with the quality of scientific journals --
bad spacing, paragraph layout etc.
Adobe todays uses TeX paragraph layout algorithm.
And finally, he demonstrated that even large programs can be
explained in full detail (unlike most programs written by early
Bell Labs guys), though Pascal is not a great language on which
WEB (the implementation language of TeX) is based.


- Saroj Mahapatra


Sent via Deja.com
http://www.deja.com/

Douglas A. Gwyn

unread,
Jan 12, 2001, 12:17:57 PM1/12/01
to
sa...@bear.com wrote:
> No, that is not true. Knuth felt that troff was burdened with
> couple of generations' legacy at Bell Labs, and he wanted to
> start with a clean slate without fear of hurting any Bell Labs
> colleague's feeling (about troff).

? Where did you hear that? It doesn't even make sense.

> And finally, he demonstrated that even large programs can be
> explained in full detail (unlike most programs written by early
> Bell Labs guys), though Pascal is not a great language on which
> WEB (the implementation language of TeX) is based.

There is of course CWEB.

The necessity for WEB was largely due to Pascal's requirement
to define things out of their natural order. It is much less
of a problem for C.

Whether programs are well documented or not really has nothing
to do with typesetting languages.

Boyd Roberts

unread,
Jan 12, 2001, 1:08:38 PM1/12/01
to
> The necessity for WEB ...

my god, i'd forgotten about _that_ horrible thing.


Message has been deleted

Boyd Roberts

unread,
Jan 12, 2001, 2:09:38 PM1/12/01
to
From: <sa...@bear.com>

>
> No, that is not true. Knuth felt that troff was burdened with
> couple of generations' legacy at Bell Labs, and he wanted to
> start with a clean slate without fear of hurting any Bell Labs
> colleague's feeling (about troff).

nonsense, research does not work like that. if you're afraid
of 'hurting' people's feelings you'll get no research done
at all. do you understand the scientific method?

do you think einstein would have held back not to 'hurt'
newton's feelings? are you mad? admittedly, newton
was dead, but it _could_ hurt his standing, retrospectively.

> And finally, he demonstrated that even large programs can be
> explained in full detail (unlike most programs written by early
> Bell Labs guys), though Pascal is not a great language on which
> WEB (the implementation language of TeX) is based.

LP is nonsense. you write the code. the code is the 'truth'.
you add comments to explain the global view and stuff that's
not 'obvious'. not 'obvious' is a judgement call. you get
'judgement' after you've spent a few years in the caper.

unfortunately, that last sentence is not always true.

then again, no documentation is better than bad/incorrect
documentation.

i fought tooth and nail with my old man about what was
the point of 'pure' research. i thought it was pointless.
finally, i realised he was _right_.

he did surface physics. kinda useful now, given we're at
the limit of how many transistors you can fit on a given
chunk of silicon.

do you think he thought about 'hurting' the feelings of the
guys who built the electron gun (at york) that 'no-one could
fix' when the problem landed on his desk and he fixed it?

'doing a boyd' maybe, but somebody's got to do it.


sa...@bear.com

unread,
Jan 16, 2001, 11:25:47 AM1/16/01
to
In article <3A5F3607...@arl.army.mil>,

"Douglas A. Gwyn" <gw...@arl.army.mil> wrote:
> sa...@bear.com wrote:
> > No, that is not true. Knuth felt that troff was burdened with
> > couple of generations' legacy at Bell Labs, and he wanted to
> > start with a clean slate without fear of hurting any Bell Labs
> > colleague's feeling (about troff).
>
> ? Where did you hear that? It doesn't even make sense.
>

It is somewhere in his "Digital Typography" book.

> > And finally, he demonstrated that even large programs can be
> > explained in full detail (unlike most programs written by early
> > Bell Labs guys), though Pascal is not a great language on which
> > WEB (the implementation language of TeX) is based.
>

[snipped]

> Whether programs are well documented or not really has nothing
> to do with typesetting languages.
>

I was referring to TeX and MetaFont implementation.

sa...@bear.com

unread,
Jan 16, 2001, 11:31:00 AM1/16/01
to
In article <01b901c07cc8$bc634660$0ab9...@cybercable.fr>,

bo...@planete.net (Boyd Roberts) wrote:
> From: <sa...@bear.com>
> >
> > No, that is not true. Knuth felt that troff was burdened with
> > couple of generations' legacy at Bell Labs, and he wanted to
> > start with a clean slate without fear of hurting any Bell Labs
> > colleague's feeling (about troff).
>
> nonsense, research does not work like that. if you're afraid
> of 'hurting' people's feelings you'll get no research done
> at all. do you understand the scientific method?
>

That was precisely the point! TeX was done from the ground up.

> do you think einstein would have held back not to 'hurt'
> newton's feelings? are you mad? admittedly, newton
> was dead, but it _could_ hurt his standing, retrospectively.
>
> > And finally, he demonstrated that even large programs can be
> > explained in full detail (unlike most programs written by early
> > Bell Labs guys), though Pascal is not a great language on which
> > WEB (the implementation language of TeX) is based.
>
> LP is nonsense. you write the code. the code is the 'truth'.
> you add comments to explain the global view and stuff that's
> not 'obvious'. not 'obvious' is a judgement call. you get
> 'judgement' after you've spent a few years in the caper.
>
> unfortunately, that last sentence is not always true.
>
> then again, no documentation is better than bad/incorrect
> documentation.
>
> i fought tooth and nail with my old man about what was
> the point of 'pure' research. i thought it was pointless.
> finally, i realised he was _right_.
>

Sorry, how many programs written by you can be read from cover to
cover and fully understood?


[snipped]


Thank you,

Reply all
Reply to author
Forward
0 new messages