Imagine a boot stamping on your ideas forever

Via Daring Fireball, author Charlie Stross writes about his hatred of Microsoft word.  I don’t know enough to evaluate the technical arguments about design and file formats, including the comments left by a commenter “globetrotter” who self-identifies as a MS program manager for early versions of Word. But the empirical experience of Word is sufficient to share Stross’ loathing for it, shared by the majority of his commenters.  I just about choked when I read globetrotter’s comment:

In Word you can open a 200,000 page document, and make 20 quick edits and save without it taking most of the afternoon.

Really? Now it may be true that Word can handle a 200K page plain text document better than competitors.  But if this 200K page document contains any embedded textboxes, figures, or tables that are the reason to use Word vs a plain text editor, it’s likely to crash horribly.  I’ve suspected for years that Word has a calendar function that crashes more frequently as grant deadlines approach.

Even when it doesn’t crash, I hate the way Word (and Office) change your text to “correct” it. Word and Powerpoint are responsible for a generation of biologists misusing the abbreviated forms of the binomial species names; I’ve lost count of the times I’ve seen E. coli rendered as E. Coli. And things like this comment are all too familiar:

 One of the things the Tech Ed supplied us was a monthly Q&A column. Going over his copy, I noticed that he’d eschewed the usual “Q” and “A” prefixes in favour of “Q” and “R”. “Question and Response, I guess… OK, bit quirky, bit individual, leave it in”. The second question and answer were also given as a pair of paragraphs beginning “Q” and “R”; the third question and answer, however, came as a pair of paragraphs beginning “S” and “T”.

At this point I saw what was going on, and changed the “R” prefix to an “A”.Word changed it back, right before my eyes. It happened several times before I got the changes to stick, even after I’d told it not to format the paragraphs as numbered lists.

Editing a multiple choice test in Word can be a real adventure.  Then there are things like mysteriously unselectable formatting elements, and unusably dense track changes markup.

So I hate Word too.  So why do we use it?  First some minor quibbles:

Was Word first?

Stross writes:

Steve Jobs approached Bill Gates to write applications for the new Macintosh system in 1984, and Bill agreed. One of his first jobs was to organize the first true WYSIWYG word processor for a personal computer — Microsoft Word for Macintosh.

Xerox Bravo was first, but the Alto never made it to the market. Similarly, the WYSIWYG in the Xerox Star and Apple Lisa could arguably be discounted. But if we use the original Mac as the personal computer where WYSIWYG Word first appeared, both MacWrite and WriteNow (originally for the NeXT) were available before the first version of Word for the Mac. I wrote my PhD dissertation in WriteNow before Word was available. Perhaps Stross is saying Microsoft promised the first WYSIWYG for a personal computer, but was late to deliver it.

TeX and scientific writing


Not all commenters agree, but you see this kind of thing in discussions of Word suckitude

I know that for scientific publications, latex is the rule (a friend of mine does a lot of mathematic translations).

LaTex is the rule in some fields, but not others. NSF’s survey of doctoral scientists from 2008 (pdf) shows that 164,000/651,200 employed PhD scientists and engineers were in the life sciences.  Biologists use Word. The NIH grant templates are in word. Our journals accept Word files. Our journals even link to downloadable Word files as supplemental data (ugh).  But if my Facebook friends around grant time are representative, there are plenty of other biologists who hate Word.

If we hate it, why do we use Word?

To some extent, it’s for the same reason everyone else does: because everyone else does. The ability to edit Word documents is assumed when you are collaborating with others.  Collaboration with people who didn’t want to use Google Docs is why I did my most recent upgrade to Word; there were docx files that my older version could no longer convert. If you’re not collaborating with others who demand the ability to work in Word, the journals and granting agencies take pdfs these days, so you can switch to anything that can generate pdfs. Not that long ago, journals and granting agencies expected hardcopy, so you could do your word processing in anything that could print.  Several colleagues and I clung to WriteNow  for a long time.

For biologists like me, I think grants drove people to use Word more than publications.  After all, for publications, we still submit things in dull double-spaced text without inline figures and tables. The publisher has the fancy stuff to do the layout. But for the grants we used to print and now submit as pdfs, there is no other publisher/design house to handle the layout. And as much as we as scientists may try to value substance over style, all other things being equal, a grant that is easier on the eyes gets the edge.  Inline figures and tables for a grant are even more important now that page limits have been reduced at NIH. And even if a campus had their own service to do layout on our grants, we probably wouldn’t use it because it wouldn’t let us use every last available procrastinated minute before the application had to be sent.

But we can make inline figures and tables with a variety of software packages we hate much less than Word.  As a Mac user, why don’t I use Pages, for example? Some of it is the collaboration issue described above. But the other is summarized in one word:


Scholarly writing means citing sources and at least in my part of the life sciences, EndNote is the de facto standard, and uses academic discounts similar to those Microsoft uses to keep people from switching.  There are alternatives to EndNote, but as far as I can tell, plugins to support EndNote-style Cite while you write functionality are relatively new. The expectation that collaborators will be able to share EndNote libraries reinforces the  disincentives to change.

Could Word lose its dominance in science writing?

Of course. I’m old enough to remember paying typists to do manuscripts and grants. That’s not coming back, but it means I also lived through the days when people advocated training all students to use WordStar, and I remember the competitive market for plastic sheets that sat over your keyboard with the Wordstar key bindings.  Perhaps WordStar’s dominance then was not as great as Word’s is now, but there is space for competitors, based on limitations of Word. I already use a combination of Google Docs and Word, where Word is used to tweak the final version, put in references, and so on.  Reference managers that work with Google Docs are starting to show up.  As I write this, Apple is preparing an announcement about iWork on iCloud.