schultz@mmm.UUCP (John C Schultz) (09/28/88)
I am sure this question has been asked before, but the question has
just arisen for me again so here goes.
What do people feel are the advantages of WYSIWYG phototypsetting
(e.g. MacWrite, FRAMEMAKE) vs what I will call "programmed"
phototypesetting (e.g TeX, LaTeX)?
I will summarize if enough responses, either to these general groups
or e-mail to myself.
To start things off, here are my opinions, hopefully cleansed of
religious opinions.
pro WYSIWYG
- easy to learn
- hardcopy is identical to display
- immediate "results"
- easy creation/addition of pictures, images, etc.
con WYSIWYG
- difficult to implement style and formatting changes
- the writer becomes concerned more with the format than content
- manual labeling of page numbers, references, figures, sections, etc.
- speed - printing speed is generally limited by printer
- hardcopy is sometimes NOT identical to the display (tabs for instance)
- good ones can be very expensive (e.g. Interleaf)
pro PROGRAMMED
- worry about content - then formatting
- computer re-numbering of pages
- computer cross-referencing and numbering figures, tables, eqs, etc.
- softcopy output IS possible (eg. texx)
- computer generated index, list of tables, figures, etc.
- speed - compilation makes printing fast (duplicates are easy)
- custom forms for common document "feel" - e.g. memo.sty in LaTeX for memos
- user does not have to align table or columns of numbers (\tabular in LaTeX)
- TeX and derivatives are public domain (or close to it)
con PROGRAMMED
- long "compile time"
- uncertain page breaks
- difficult to debug
- many hardcopies are needed to correctly format complex documents
- TeX and derivatives are public domain (or close to it)
--
john c. schultz schultz@mmm.UUCP (612) 733-4047
3M Center, Bldg 518-1-1, St. Paul, MN 55144-1000
The opinions expressed herein are, as always, my own and not 3M's.hugo@chocorua.dartmouth.edu (Peter Su) (09/28/88)
In article <1141@mmm.UUCP> schultz@mmm.UUCP (John C Schultz) writes: > >pro WYSIWYG > >- hardcopy is identical to display >- immediate "results" Niether of these are always true. With the Mac at least, the screen cannot come close to displaying what will go one when you print your document to a laser printer, so WSYWIG is a misnomer. Also, most Mac processors are so painfully slow at doing anything substantial that it isn't worth bothering. Mac word processors tend to over emphasize the importance of bells and whistles and underemphasize the importance of real *text* processing. Of course, this true for most IBM programs as well. (Sorry, pet peave...) >pro PROGRAMMED > >- worry about content - then formatting Hardly ever the case. For example, the formatting of equations is very closely tied to how they are formatted. Both TeX and troff make you worry a lot about formatting type commands as you input your text. LaTeX and Scribe are better, but still have the same problem. Interestingly, all of your cons are true. This indicates to me that text processors are no where near as good as anybody would like to think. I think that systems that combine the two approaches, and allow the user to specify documents in a more modular way (like outline processor type systems) will do better. Pete
cosell@bbn.com (Bernie Cosell) (09/30/88)
In article <1141@mmm.UUCP> schultz@mmm.UUCP (John C Schultz) writes: }I am sure this question has been asked before, but the question has }just arisen for me again so here goes. } }What do people feel are the advantages of WYSIWYG phototypsetting }(e.g. MacWrite, FRAMEMAKE) vs what I will call "programmed" }phototypesetting (e.g TeX, LaTeX)? A common prelude to a good flame-war. Here goes... }To start things off, here are my opinions, hopefully cleansed of }religious opinions. Well, my religious opinion is that the common WYSIWYG systems are mostly adequate for small scale, not too formal documents, but for anything large-scale or constrained to be _really_ well done, I think that you have to go with one of the top-notch "programmed" ones. Note I think that we're just turning a critical corner in both arenas: programmed typesetters with built-in (and very nice) real-time previewers (and so you can see real-quick WYG), and WYSIWYG systems that are _really_ just glorified highly-specialized front ends to an underlying programmed formatter (and so they can, at least, cope with some of the more complicated stuff, like footnote and tables of contents and such) We seem to be converging on a pretty good middle from both ends. }pro WYSIWYG } }- easy to learn "easy to learn" is a bit of TANSTAAFL -- the easy ones are unbelievably cretinous (but surely easy), the _good_ WYSIWYG systems have manuals as big as SCRIBE's. It may be easier to sit down and start typing a paragraph, but the WYSIWYG equivalent of \section{} or \enumerate{} can be a bit complicated. ....other good points.... } }con WYSIWYG } }- difficult to implement style and formatting changes }- the writer becomes concerned more with the format than content }- manual labeling of page numbers, references, figures, sections, etc. }- speed - printing speed is generally limited by printer }- hardcopy is sometimes NOT identical to the display (tabs for instance) }- good ones can be very expensive (e.g. Interleaf) I find these CONs to be a pretty good summary of what I've seen, and I find them (for my purposes) to be _fatal_, for the most part. In particular, most of the technical people around here are (at best) mediocre writers; they have proven (time and time again) that they are ZEROs at document design. When it comes to picking fonts and point sizes and numbering conventions and indentations and margins and .. and... folks mostly have AWFUL taste (but plenty of opinions! :-). Almost all of the WYSIWYG stuff we have comes out QUITE badly done -- Ok if you're just whipping off an informal document, just stuff that just doesn't stand up even to casual scrutiny, much less serious critiquing. Also, other CONs are that _most_ WYSIWYG systems are pretty anemic formatters -- generally they provide little (if any) suport for footnotes, reference management, tables of contents, indexes, etc. Again, not real problems in smallish, not-too-formal documents, but murder for the "real thing". }pro PROGRAMMED } }- worry about content - then formatting More to the point, with good ones you *never* worry about formatting. They are predictable and competent. For almost all authors, the world would be better served if they NEVER worried about formatting and just let the computer handle it. ... more good points... } }con PROGRAMMED } }- long "compile time" Long is in the eye of the beholder. Two-page documents are vanishingly fast on even the piggy programmed systems. 500 page documents are a bear, but then try doing a 500 page document on your favorite WYSIWYG system. I've found that the delay is basically commensurate with the time it takes to print the thing. That is, I don't find the delays in running the "compilation" as being particularly signficant compared with the time to have the laser printer dump the stuff out and for me to walk to the printer room and pick it up, etc. }- difficult to debug Amen!!! Even a relatively user-friendly system (like SCRIBE) is a pretty subtle monster. Fighting with troff and TeX (my two most frequent opponents) is virutally a Black Art. Makes debugging a busted timesharing system using only the testword switches (this on a PDP-1, years ago) simple by comparision. UGH!!! }- uncertain page breaks }- many hardcopies are needed to correctly format complex documents These two relate to something that is a PET peeve of mine. Mostly, the authors -- the folks pouring out the raw text -- are lucky if they can manage sentences with an identifiable verb, and often having paragraphs that make sense from one to the next is too much to ask for. That such folk then presume to debate whether a figure would be better floated here or there, or if a footnote should have been divided in a different place, or if... or if... Yes, there are _some_ people who are appropriately endlessly frustrated by the vagaries and systematic mistakes that the formatters make, but there CERTAINLY aren't enough of _them_ around to justify WYSIWYG systems. And I also admit that there are _some_ typographic chores (complicated equations or tables, for example) that *are* hard to get right. They're not all that common (unless you're writing a textbook on tensor calculus or something like that), and even then a good previewer will home you in pretty quickly. (although around here, generally when that happens the job is best left to a "guru" for the appropriate system -- they can get REAL close the first time, and know the magical incantations it takes to tune things. I've never tried to do real hairy tables or formulae or the like in a WYSIWYG system, so I'm not sure quite how it goes there) }- TeX and derivatives are public domain (or close to it) This kind of thing turns out to matter quite a lot -- ever see how much a SCRIBE site license costs? I shudder to think what we could have done with the money that has been spent buying zillions of copies of various WYSIWYG systems for the MACs that dot the landscape around here. Phew! __ / ) Bernie Cosell /--< _ __ __ o _ BBN Sys & Tech, Cambridge, MA 02238 /___/_(<_/ (_/) )_(_(<_ cosell@bbn.com
chris@mimsy.UUCP (Chris Torek) (09/30/88)
In article <47700025@uxe.cso.uiuc.edu> mcdonald@uxe.cso.uiuc.edu writes: >On the IBM-PC and clones (as well as top end workstations) is IS possible >to get essentially true WYSIWYG. There are display available at >1200x1600 (portrait format) (relatively common) up to 3000x4000 >(at 300 dot per inch and not common at all). Monitors capable of even the former resolution typically cost more than the PC. (The Philips monitor used on Suns can scan about 1k by 1k and costs about $1k :-) .) The prices *are* improving, but 3k by 4k is indeed rare (and expensive), and that, to me, is a bare minimum for printer resolution. If you intend to publish, 300 dpi is not fine enough: the degradation introduced in the process of printing demands about an order of magnitude more resolution. 30k by 40k monitors, anyone? :-) -- In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7163) Domain: chris@mimsy.umd.edu Path: uunet!mimsy!chris
tbray@watsol.waterloo.edu (Tim Bray) (09/30/88)
>What do people feel are the advantages of WYSIWYG phototypsetting >(e.g. MacWrite, FRAMEMAKE) vs what I will call "programmed" >phototypesetting (e.g TeX, LaTeX)? Clearly for serious typesetting the programmed (more properly procedural) typesetting systems are more appropriate. But neither is the right answer. The right answer is to design the document using *descriptive* markup, then when you're finished, use a program to translate it into troff, TeX, or whatever works best where you are. An example of a standard for descriptive markup is SGML. The reasons for this are too long to go into here. Check out the excellent article in the Nov. '87 CACM by Coombs et al. Tim Bray, New Oxford English Dictionary Project, U. of Waterloo
mrd@sun.soe.clarkson.edu (Mike DeCorte) (10/02/88)
One thing that I find very interesting and unbelievebly frustrating is that people are very intimidated by TeX. There is a prof here at Clarkson how publishes magazine and uses one of the PC desk top publishing systems, I think Versatype (I don't know anything about how to use it and the name could even by wrong). When I asked him if he had thought about using TeX to do his journal, his response was that he looked at TeX briefly and didn't like the idea of all the control stuck in his text. People feel that the WYSIWYG's are easier to use even if they arn't. -- Michael DeCorte // (315)268-2292 // P.O. Box 652, Potsdam, NY 13676 Internet mrd@sun.soe.clarkson.edu // Bitnet mrd@clutx.bitnet ------------------------------------------------------------ Clarkson Archiver Server archiver-server@sun.soe.clarkson.edu archive-server%sun.soe.clarkson.edu@omnigate.bitnet dumb1!dumb2!dumb3!smart!sun.soe.clarkson.edu!archive-server ------------------------------------------------------------