[comp.text] WordMarc

kiely@lownlab.UUCP (James P. Kiely) (01/14/87)

I am planning to buy "WordMarc" (a.k.a. "Muse") as the
WYSIWYG word processor for a VAX 11/730.  Does anyone have
any experience with this package?  Is there anything inherently
good or bad about this particular word processor?  Would anyone
strongly suggest that I buy another package rather than "WordMarc"?  
Thanks.
-- 
 
NAME:     James P. Kiely                USPS:   Kiely Laboratories
USENET:   ...!harvard!lownlab!kiely             P.O. Box 624
PHONE:    +1 617 782 4136                       Allston, MA 02134-0624

roy@phri.UUCP (Roy Smith) (01/15/87)

In article <325@lownlab.UUCP> kiely@lownlab.UUCP (James P. Kiely) writes:
> I am planning to buy "WordMarc" (a.k.a. "Muse") as the
> WYSIWYG word processor for a VAX 11/730.

	I assume you are talking about the product marketed by MARC
Software.  We tried it out a couple of months ago and were terribly
disappointed.  Our requirement was to print technical stuff on an Apple
LaserWriter, which they claimed they supported.  At least at the time, that
was somewhere between a joke and an outright lie.

	Their idea of support for an Apple LaserWriter was to flip the
switch to Diablo 630 emulation mode.  When we heard this, we were horrified
and told them that what we meant by "supporting the LaserWriter" was
"producing PostScript output".  They told us that they were working on it
and would send us a preliminary version of the PostScript driver.

	Everything from the word "go" was a mess.  First, we couldn't read
the tape they sent us (on either of two tape drives).  It turned out to be
a combination of the block size on the tape not matching the documentation
(figured that one out with trial and error using "dd") and parts of the
tape being unformatted.  When I convinced them the tape must be bad, they
sent us another one, but we couldn't read that either.  It took about 3 or
4 tapes before we got one that we could even load.  They muttered something
about how they didn't have a tape drive on their Sun system and had to move
the files to a Celerity (which had a strange notion of block size) and there
was something wrong with the drive, etc.  I can understand a drive having
hardware problems, but to send 3 bad tapes in a row to a customer doesn't
exactly inspire confidence.

	I suppose there is nothing exactly wrong with it, but the tape
turned out to be in cpio format, with each of the files contained on the
tape being a tar image!  I guess that's legal, but why bother with that
brain damage?  There are lots of systems that come with cpio *or* tar, not
both, so this seems pretty stupid to me.  Well, anyway, we have both, so
that wasn't a real problem.

	Next problem.  The code only comes as binaries (OK, OK, I'm
spoiled; you can't really take points off for not distributing source on a
commercial product) so configuration is rather inflexible.  There are
several directories built in which didn't match our disk organization.  We
wanted to put the stuff in /usr/local/{bin,lib}, but the binaries expected
to find the stuff in /usr/wmc.  A few symlinks took care of that, so no big
deal.  Worse was this monster shell script they had for installing the
software with a cute, but totally unnecessary, set of menus that got in the
way more than they helped (Nested menus in sh?  Give me a break!)  I don't
remember the details, but we had to edit the script to get it to work on
our system -- there were some assumptions made that symlinks couldn't fix.

	The list of problems with installation goes on and on.  For
example, the distribution comes with a bunch of terminal drivers standard,
and there is a menu item for loading the optional drivers.  On the Sun
tape, the driver for the Sun is on the "optional" part -- this caused us
rather a bit of confusion.

	I will give them credit for trying, however.  It just happened that
about the time I was getting fed up (which was actually fairly early on,
but I kept at it to humor some people), there was a big Unix show in New
York.  They arranged to have the guy who was working on the PostScript
driver get assigned "show duty" so I had a chance to meet him in person.
Since their flight back wasn't until the evening of the day after the show
closed, he even spent the morning at our site, helping us get the software
up.  Several brownie points for sending a programmer to a customer site to
do the install.  Unfortunately, these are the only points MARC gets, and
they are not enough by far.  Would it have made a difference if the 'Niners
got a second field goal against the Giants?

	It turns out that they don't even have a PostScript printer in
house (or didn't at the time), and were working largely from documentation
and borrowing time on other customer's machines for testing!  Needless to
say, the driver didn't work very well and the guy spent most of the morning
patching it together.  Once it did get running, it was a disaster.  Seems
that WordMARC can't deal with proportionally spaced type; it's designed for
fixed-width fonts like on daisy wheel printers.  The PostScript driver was
originally set up to use Courier.  When I pointed out that we wanted to use
the other fonts, they made it use Times or Helvetica, but the software
still did the filling calculations as if the type was monospaced!  Needless
to say, the results were not satisfactory.

	One of our requirements is being able to set math and scientific
stuff.  They claim they can do this, but what they offer is pretty sad.
You designate a shift character (I think they use "[" by default).  Then,
to do alpha, you say [a[, or something like that.  That's not too bad until
you have to start remembering that a square root sign is [%[ and an
integral is [@[ (or whatever they are).  They are real proud of the fact
that they can do 15 levels of sub and super scripts (and this kind of
marketing bullshit even impressed some of the people here, much to my
grief) but the results are so horrible looking as to not even be readable,
let along aesthetically pleasing.  If you want a good laugh, ask to see some
samples of what they can do in math mode.

	OK, let's assume they have solved the minor problems of getting the
software to work, and making the installation reasonably painless.  You
still have a piece of software that I don't think is worth the effort.  For
some fairly large number of thousands of dollars, you get something which
can fill and adjust monospaced type and center lines.  It's got some sort
of on-line help, menu-guided front end, but I didn't think that worked very
well.  There are also some commands for searching and moving text, but
nothing out of the ordinary.  For my money, if that's all the functionality
you need you will be much better off with some flavor of Emacs in auto fill
mode.  I think even the commercial versions of Emacs are cheaper than
WordMARK, and you can always go for one of the PD versions, such as gnu.
If you need anything much fancier than that, WordMARK just won't do it for
you anyway, and you'll do better to go with something else.

	Frankly, other than InterLeaf (which is pretty wonderful but I
can't even begin to afford it) I don't know of any WYSIWYG systems
currently available that I would want to use.  Some of the Xerox systems
look pretty good from what little I've seen of them, but as far as I know,
Xerox machines don't talk to Unix machines, so we didn't look too closely.
Certainly, I don't know of any worthwhile WYSIWYG's that run under a
general-purpose timesharing OS on an ASCII terminal.  If you find one, let
me know -- we've spent a lot of time looking.

	What we're experimenting with now is TeX, previewed on Sun
workstations.  We're eagerly awaiting VorTeX, which as far as I understand
it, is a sort-of WYSIWYG version of TeX.  Maybe the answer is some flavor
of MacWrite/Draft/Paint/Word/Whatever on an Apple Macintosh (which we will
also be playing with shortly).  While I have fought fiercely to keep
dedicated word processors out of this place, I do recognize that in some
environments they are probably the right solution.  There are some people
around here who swear by Wang systems for scientific manuscripts, so I
suppose they might be worth looking into.

	Meanwhile, we sent back the WordMARC tapes and we're still
struggling along with troff and friends (eqn, tbl, bib, etc) and hoping
something better comes along that we can afford.
-- 
Roy Smith, {allegra,cmcl2,philabs}!phri!roy
System Administrator, Public Health Research Institute
455 First Avenue, New York, NY 10016

"you can't spell deoxyribonucleic without unix!"

usenet@soma.bcm.tmc.edu (USENET maintenance) (01/19/87)

We have been running WordMARC on our Masscomp system for over two years.
We find it useful and easy to use, but it does lack some features that
limits us. As roy@phri has already noted, WordMARC has no concept of
proportional printing. It also cannot handle a wide variety of font
sizes. It has no concept of a "page" and that limits its usefulness
with conventional laser printers. It can be an administrative nightmare. 
It has its own database of terminal and printer types and each available
terminal line must be defined for WordMARC to know what to do with it.

However, the menu system does make it easy to use. It does have the
capabilites of dealing with special characters. It does produce
a document file that can be shipped from UNIX to VMS to MS-DOS or
wherever a version of WordMARC is available and be used without alteration.
It does handle large document assembly in a reasonable fashion.

Also, I get the impression that the people at MARC are really trying to
meet the demands of their sophisticated customers. It just takes them time
to develop a plan of attack.

In short, if your user base is too busy to become fluent in page
description languages or troff commands, WordMARC is a good choice. But,
you do lose something.

Stan	     uucp:{shell,rice,cuae2}!soma!sob       Opinions expressed here
Olan         domain:sob@rice.edu or sob@soma.bcm.tmc.edu   are ONLY mine &
Barber       CIS:71565,623   BBS:(713)790-9004               noone else's.