tsmith@solbourne.com (Terry Smith) (05/11/91)
I asked for comments from Xroff users. Here are the responses.
Thanks to everyone.
tsmith@solbourne.com
-------------Cut Here--------------
From dbs@kodak.pa.dec.com Fri May 3 12:20:15 1991
Received: from Solbourne.COM (stan) by harmony.Solbourne.COM (4.1/SMI-4.0)
Date: Fri, 3 May 91 11:17:00 -0700
From: dbs@Decwrl.dec.com
Message-Id: <9105031817.AA05279@dbsmax.pa.dec.com>
To: tsmith@Solbourne.COM
Subject: Re: xroff
Status: R
Terry,
Stick with eroff. The x in xroff stands for Xerox; it's a *roff that
was developed to drive a Xerox 2700, an early 1980's laser printer.
I'm sure that Image Network has a PostScript driver by now, but you're
probably better off with the Elan product.
--Dan
-------------------------------------------------------------------------
Date: Fri, 3 May 91 14:34:34 -0500
From: Brain in Neutral <bin@primate.wisc.edu>
Message-Id: <9105031934.AA06347@uakari.primate.wisc.edu>
To: tsmith@Solbourne.COM
Subject: Re: Xroff -- Request for Comments/Gripes
Status: R
The x is for xerox, as in xerox 2700 or 4045 printers. Nothing to
do with X windows. Xroff now has a PostScript post-processor now
as well as xerox.
We have it, use it a lot, it works pretty well.
Yours,
Paul DuBois
dubois@primate.wisc.edu
-----------------------------------------------------------------
From: steve@devnull.mpd.tandem.com (Steve Williams)
Message-Id: <9105040021.AA00332@wave9>
Subject: Re: xroff
To: tsmith@Solbourne.COM (Terry Smith)
Date: Fri, 3 May 91 19:21:31 CDT
X-Mailer: ELM [version 2.2 PL16]
Status: R
Strange you should ask. Just a few weeks ago I posted a query to
comp.text and a local newsgroup seeking opinions about eroff. Since
then we've gotten a fully-functional version of xroff, tried it and
returned it, and ordered and rec'd eroff.
Xroff has one large problem, which is that it uses its own fonts, not
taking advantage of the built-in LaserWriter fonts (or any builtins).
Every file -- EVERY file -- that you send to the printer has complete
font definitions within it. That means added file size and transmission
time, but worse, the fonts don't match the standard PostScript fonts.
We didn't pursue it much beyond that, so other aspects of their product
may be wonderful or horrible. Their previewer looked nice, but it, too,
depended on their proprietary fonts.
We went ahead and ordered a few thousand bucks worth of Elan's eroff and
X11 "environment" (basically their previewer), with a 30-day moneyback
etc. I just rec'd it yesterday and we haven't got it fully installed,
so I can't comment much. What sort of comments would your group make?
Why are you considering a switch?
-----------------------------------------------------------------------------
Steve Williams steve@mpd.tandem.com
or: cs.utexas.edu!halley!steve
-----------------------------------------------------------------------------
From: wpr@devnull.mpd.tandem.com (Bill Robinson)
Subject: Re: xroff
To: steve%wave9.mpd.tandem.com@devnull.mpd.tandem.com (Steve Williams)
Date: Mon, 6 May 91 8:10:57 CDT
Cc: tsmith@Solbourne.COM
In-Reply-To: <9105040021.AA00332@wave9>; from "Steve Williams" at May 3, 91 7:21 pm
Steve's message says it all. Xroff itself seemed like a decent product.
Printing efficiency aside, we didn't particularly like their fonts,
and to match the appearance of documents produced with other software,
we have to stick to the PostScript fonts. So the previewer was essentially
useless to us, since the line breaks differed from our printed versions
with PostScript fonts.
I posted a notice to the Documentation Special Interest Group of UNIX
International asking for people to share their experiences with screen
previewers. The e-mail generated by that and some postings on comp.text
(not a large amount, to be sure) suggested that eroff is generally preferred
to xroff, although I can certainly imagine that for some needs xroff might
be preferred. If you have any particular information about eroff's
strengths or quirks that would help us, we'd love to hear from you.
----------------------------------------------------------------
From: chip@chinacat.Unicom.COM (Chip Rosenthal)
Newsgroups: comp.text
Subject: Re: Xroff -- Request for Comments/Gripes
Date: 4 May 91 00:37:15 GMT
Organization: Unicom Systems Development, Austin, TX
In article <1991May3.163405.23681@solbourne.com>
tsmith@solbourne.com (Terry Smith) writes:
>Any Xroff users out there?
The Image Network folks were kind enough to let me eval a copy of
xroff when I was DWB shopping. I ended up selecting eroff, but I've
got nice things to say about the Image Network people and the xroff
product.
>We're currently using eroff, but want to evaluate Xroff. It looks like
>any other roff to me, but with a few X-named scripts doing some of the work.
>I don't see any real connection to X Windows other than it has a previewer
>that runs in an xterm.
First, you are right, the Image Network xroff program has no relation
to the X Window Pig(tm). xroff is their port of DWB. I tested it
out on a plain vanilla XENIX/386 box.
I like xroff. It and Elan seemed to be the two most robust DWB's I
looked at. Xroff also included a nice set of fonts - from DigiFont
I believe. I'd kill to be able to have a copy of the `Palatino' font
Image Network ships. The HP Palacio is a poor substitute. (I'll
leave my story of how the DigiFont sales prevention people kept me
from doing business with them for another time.)
The thing which was a total knock-out punch for xroff is the fact that
the backend is not ditroff compatible - nor was it documented.
Therefore, if you've got any tools for printer handling, fonts, etc.
they are useless. I needed this capability, and thus xroff was rendered
unusable for my needs.
xroff is similar in capability (and unfortunately price) to Elan eroff.
All the preprocessors (pic, tbl, etc.) are there. If not for the
proprietary backend, the choice between eroff and xroff would probably
have been a coin flip. I think I would have ended up picking xroff
because of the fonts.
BTW...why are you looking to drop eroff?
--
Chip Rosenthal 512-482-8260 |
Unicom Systems Development | I saw Elvis in my wtmp file.
<chip@chinacat.Unicom.COM> |
-----------------------------------------------------------------
From npn@sirius.att.com Mon May 6 10:05:33 1991
Date: Mon, 6 May 91 10:58 EDT
Original-From: sirius!npn (Nils-Peter Nelson +1 908 582 6078)
To: tsmith@Solbourne.COM
Subject: troff under X
I've been fooling with xditivew, the X11R4 troff previewer.
Substantial sections had to be re-written to get it
to work acceptably. This has been done ( including adding
all the pic code). We are going to include a version with
an OPEN LOOK interface as source code with Picasso 2.0.
If you're looking for something cheap to use internally,
I suppose I can't beat free beer. If you're looking for a
product to sell to your customers, both Picasso 2.0 and the
troff previewer (xtv) load without change on Sun4, 386,
MIPS, Silicon Graphics. Both are excellent companions to
the basic DWB package.
In its current state, xtv handles equations, tables, pic,
multiple font troff, and generally behaves very satisfactorily.
Personally, I love it.
---------------------------------------------------------------
Newsgroups: comp.text
Subject: Re: Xroff -- Request for Comments/Gripes
Date: 7 May 91 17:06:31 GMT
Organization: Elegant Communications Inc.
In article <1983@chinacat.Unicom.COM> chip@chinacat.Unicom.COM (Chip Rosenthal) writes:
> In article <1991May3.163405.23681@solbourne.com>
> tsmith@solbourne.com (Terry Smith) writes:
> >Any Xroff users out there?
I'm not an Xroff user, so you might want to take my comments in
stride. I do however have extensive (user) experience with V7-troff,
DWB-1.0, DWB-2.0, and SQtroff.
> I like xroff. It and Elan seemed to be the two most robust DWB's I
> looked at.
Just what do you mean by robust? I've only once encountered a
non-robust DWB, and that was the initial SysV/386 DWB-2.0. The first
version of tbl from that release usually dumped core. Some vendors
fixed it before the released it, others didn't. Other than that, I've
rarely seen anything in any DWB behave rudely or dump core, thus I'd
call them all robust.
> The thing which was a total knock-out punch for xroff is the fact that
> the backend is not ditroff compatible - nor was it documented.
> Therefore, if you've got any tools for printer handling, fonts, etc.
> they are useless. I needed this capability, and thus xroff was rendered
> unusable for my needs.
This does sound like a *VERY* important consideration. There are lots
of tools for dealing with di-troff output.... Even though SoftQuad
enhanced their SQtroff output, they did fully document it (not that
it's hard to understand in the first place!), and they also provided a
filter to convert it back into di-troff(5).
--
Greg A. Woods
woods@{eci386,gate,robohack,ontmoh,tmsoft}.UUCP ECI and UniForum Canada
+1-416-443-1734 [h] +1-416-595-5425 [w] VE3TCP Toronto, Ontario CANADA
Political speech and writing are largely the defense of the indefensible-ORWELL
------------------------------------------------------------------
Newsgroups: comp.text
Subject: Re: Xroff -- Request for Comments/Gripes
Date: 10 May 91 03:35:51 GMT
Organization: Unicom Systems Development, Austin, TX
[ Maybe I can flip out trn by following up two articles at once :-) ]
In article <1991May7.170631.14030@eci386.uucp>
woods@eci386.UUCP (Greg A. Woods) writes:
>Just what do you mean by robust? I've only once encountered a
>non-robust DWB, and that was the initial SysV/386 DWB-2.0.
I meant I threw a bunch of documents at it and they worked. If I
remember correctly, the only DWB which seemed to have any problems
was the ISC troff. It could have very well been pilot error. I did
take pains to document my results to (what was then) ISC/Hollis since
they were kind enough to let me eval a copy, and nobody ever called
back telling me I screwed up.
In article <748@tivoli.UUCP>
lark@greylock.tivoli.com (Lar Kaufman) writes:
>In article <1983@chinacat.Unicom.COM> chip@chinacat.Unicom.COM
>(Chip Rosenthal) writes:
> > The thing which was a total knock-out punch for xroff is the fact that
> > the backend is not ditroff compatible - nor was it documented.
>Actually, this is probably less of a consideration than you may think.
>Xroff provides an adequate selection of good fonts, as well as drivers for
>almost any laser printer you could wish
1) I rolled my own `S' font by picking and choosing glyphs I wanted. With,
xroff, I could not have done that.
2) My backend lets me place any arbitrary bitmap image into a document,
providing I'm willing to throw a bit of code into understanding this
new format. With xroff, I could not have done that.
3) The xroff I evaluated at the time didn't even have support for
included bitmap images. I had a postprocessor which would. With
xroff, I could not use it.
4) I've got the source to my postprocessor - I can make it do whatever I
want. With xroff, I cannot use it.
For some folks, locking up the backend is a knockout blow. Obviously
that isn't an issue for you. You basically confirmed what I said in
my original message. If you don't need access to the ditroff
postprocessor then xroff seems to be a very nice package.
--
Chip Rosenthal 512-482-8260 |
Unicom Systems Development | I saw Elvis in my wtmp file.
<chip@chinacat.Unicom.COM> |
--------------------------------------------------------------------
Newsgroups: comp.text
Subject: Re: Xroff -- Request for Comments/Gripes
Keywords: troff ditroff xroff sqtroff
Date: 9 May 91 01:59:27 GMT
Reply-To: lark@tivoli.com (Lar Kaufman)
Organization: TIVOLI Systems, Inc.
In article <1983@chinacat.Unicom.COM> chip@chinacat.Unicom.COM
(Chip Rosenthal) writes:
> In article <1991May3.163405.23681@solbourne.com>
> tsmith@solbourne.com (Terry Smith) writes:
> The thing which was a total knock-out punch for xroff is the fact that
> the backend is not ditroff compatible - nor was it documented.
> Therefore, if you've got any tools for printer handling, fonts, etc.
> they are useless. I needed this capability, and thus xroff was rendered
> unusable for my needs.
This does sound like a *VERY* important consideration. There are lots
of tools for dealing with di-troff output.... Even though SoftQuad
enhanced their SQtroff output, they did fully document it (not that
it's hard to understand in the first place!), and they also provided a
filter to convert it back into di-troff(5).
Actually, this is probably less of a consideration than you may think.
Xroff provides an adequate selection of good fonts, as well as drivers for
almost any laser printer you could wish, and other printer drivers as
well - even popular PC printers (there is an Xroff for MS-DOS).
I've published manuals using Xroff and found it quite satisfactory -
though I like the later SQtroff stuff, too.
-lar
Lar Kaufman I would feel more optimistic about a bright future
(voice) 512-794-9070 for man if he spent less time proving that he can
(fax) 512-794-0623 outwit Nature and more time tasting her sweetness
lark@tivoli.com and respecting her seniority. - E.B. White
-----------------------------------------------------------------------
Newsgroups: comp.text
Subject: Re: Xroff -- Request for Comments/Gripes
Date: 6 May 91 21:01:03 GMT
Reply-To: jenkins@jpl-devvax.JPL.NASA.GOV (Steve Jenkins)
Organization: Jet Propulsion Laboratory, Pasadena, CA
In article <1991May3.163405.23681@solbourne.com> tsmith@solbourne.com (Terry Smith) writes:
>Any Xroff users out there?
When I was working on my dissertation in 1987, I had a hell of a time
with xroff on eqn expressions like 'roman Q dot'. I finally figured
out that Image Network hadn't implemented the ct numeric register that
tells about character height and depth. After I had spent some time
debugging and constructing a workaround, I called Image Network and
heard something to the effect of "oh, so *that's* why that doesn't
work". They didn't think ct was used anyplace, so they didn't put the
information into the font files.
It may be a fine product now, so draw your own conclusions. I use LaTeX :-).
--
Steve Jenkins N6UNI jenkins@jpl-devvax.jpl.nasa.gov
Caltech/Jet Propulsion Laboratory (818) 354-0162
------------------------------ END ---------------------------------
--
Terry Smith Solbourne Computer, Inc
tsmith@solbourne.COM Longmont, Colorado 80501
*****----------*****
"When it comes to stewed prunes, are three enough or four too