Back to: The revolution will not be hand-stitched | Forward to: A likely tale

CMAP: "Why do you use Microsoft Word?"

(Another in the irregular "Common Misconceptions about Publishing" series of essays ...)

You already know I hate Microsoft Word. But it takes actual exposure to generate such a volume of bile, and you probably won't be surprised to learn that there's a copy of Microsoft Office 2011 for Mac on this laptop. Even though I don't use it for writing books, or even business correspondence (I've got Scrivener for the former task and Pages or LibreOffice for the latter) I can't get away from it.


It's all because of the workflow of the trade fiction publishing business.

A major publishing imprint in the USA produces hundreds of books per year. (Tor, who I publish through, produce over 300 titles; they're part of a larger group, Macmillan, whose output probably numbers in the thousands.) Publishing is not a terribly profitable business, so the payroll isn't padded with lots of spare pairs of hands: they get by with as few people as they can, and outsource jobs as piece work where possible. From my (or your) point of view, I am a special sparkly pony and my books are the only ones that matter to me. But from the point of view of a publisher, I'm one of a faceless army of small, erratic suppliers who pump meat into one end of a sausage machine, while they turn the handle.

Here's what happens inside the sausage machine. Pay special attention to steps 6, 7, 10, and 12: these are the steps where my sausage-machine metaphor breaks down, because at these stages in the production cycle of a book, the publisher needs to send little hunks of sausage meat to other chefs for special treatment (those chefs being the copy editor and the author and the typesetter). Which implies document exchange has to be possible.

Now, in principle I could send the manuscript of my latest and greatest novel to my editor on parchment, hand-scribed with a quill pen. Or I could use cuneiform on clay tablets. Or LaTeX. As long as my editor can read it, what's the problem?

The problem is that the MS then has to be sent to the copy-editor, to be marked up with changes for consistency/bug squishing. It then has to be sent to the author, who will confirm or countermand the CE's changes. It then has to be sent to the typesetter, who prepares camera-ready copy (or, these days, a PDF file with imposition suitable for feeding to a modern PDF-to-press system). The PDF then has to be checked by the author and, ideally, a proofreader (most authors make terrible proofreaders). Finally, the additional bugs that have come to light need to be fixed, and a final production file generated.

You can do this by snail mail. But this adds about a month of slack time to the production process for a book, not to mention the (sadly increasing) risk of the postal service losing a slab of paper that embodies a couple of months of valuable work. (This has happened to me in the past. Luckily I had a document scanner and am paranoid about postal whoopsies. Other authors I know were less lucky.) It also vastly increases the cost of typesetting, because someone has to copy-type, scan, OCR and proof, or somehow convert a random alien file type into something that a desktop publishing/layout program like Adobe InDesign can import. Typesetters charge by the hour, and it is vastly preferable to supply them with electronic copy in a format which InDesign or Quark can slurp in, leaving the human operator to merely add style tags and check the text for issues like rivers, orphans, and widows.

Why Word?

The simple answer is: Word is ubiquitous. For decades, just about every PC sold shipped with Windows and a demo version of Microsoft Office. It's available on the Mac. Corporate IT departments (and the major publishers all have them, because the major publishers have all the infrastructure of any other billion dollar corporation) can support it. Workable clones such as LibreOffice (descended from StarOffice by way of OpenOffice: I've been using it in one form or another since 1996) exist for minority platforms. Word supports change tracking, notes, and basic collaboration features. Because the workflow of a book is linear, there's no need for simultaneous collaboration and complex revision control; just tracking who deleted or inserted a given phrase is sufficient.

So trade fiction publishing, as it has migrated to electronic workflow (using email instead of postal mail to shave about 10% off their production cycle) relies on Microsoft Word documents. Not because anyone loves it (except corporate IT), but because it's an omnipresent miasma, like air pollution and second-hand cigarette smoke.

Novelists are not, for the most part, IT people. Much less computer science graduates. I used to write novels in vim, using rcs for revision control and a bastardized version of Perl's POD document macros for formatting, along with a hand-rolled toolchain of makefiles and perl scrips to generate output in a variety of formats. And it worked okay, back when publishers did everything on paper. But as they moved to electronic workflow everything converged on Word .doc files; and there was no escape, because proofing is an important part of the book production cycle but proofing is not the same as writing.

Yes, the author can write their book in other programs. I use Scrivener. Other folks I know use XyWrite or Protext or GNU Emacs (there's no accounting for taste). But we need to deliver something our publishers can process—at a minimum, that means RTF. Then we need to check the returned copy-edited manuscript. Which these days comes through the email as a .docx file, so we need a .docx compatible proofing tool.

LibreOffice will cope with Word change tracking. So will Apple's Pages. Alas, Scrivener can't and won't (Scrivener is not a word processor, and emits a .doc file (or an epub or PDF) as the end product of its process: there's no easy way to go in the other direction). And our publishers expect to get a .doc or .docx file back, with change tracking switched on and our own changes stacked on top of the copy editors, for editorial review before they hit the "accept all changes" button and forward the final product to the typesetter.

I try to avoid Microsoft Word like the plague, but I have my limits. For me, the requirement that finally made me crack and throw money at the Beast of Redmond was the need (in mid-2012) to edit, revise, and indeed redraft the first six books of the Merchant Princes series into three omnibus volumes in twelve weeks. There was no slack time in the proposed production cycle for those books: they were to come out at one month intervals in the UK in the spring and early summer of 2013. And I'd agreed with my editor that we'd use the original as-published American ebooks as a baseline and apply change tracking, so that we need only focus editorial attention on the changes I was going to make.

I could have trusted LibreOffice to do the job right. But we're talking about a corpus of, in the final version, 620,000 words: about the length of "War and Peace", 1.7 times as long as "The Lord of the Rings". I made something like 12,500 changes in total. The probability of LibreOffice being up to the task was high, but the risk if a hitherto-unsuspected bug surfaced and rendered my change-tracked files indigestible by Microsoft Word was that I'd have to throw out three months' hard work and re-do from scratch. (Imagine Microsoft Word as Clint Eastwood, and me staring up the barrel of a .44 magnum. "Are you feeling lucky, punk?") Let's also add that I had other deadlines, and a book to finish the same year after I redrafted that trilogy six book series. Was retaining my ideological purity worth the risk of losing three months' work and screwing my next year?

I sometimes fantasize about Markdown taking off everywhere. It's not as if Markdown editors are thin on the ground; it's almost become a de-facto standard on the iPad due to iOS's lack of a rich text library prior to iOS 7. Markdown is expressive enough to write a novel (novels are structurally very simple). It's so lightweight that you can learn the basics in half an hour and print a crib sheet on the side of a coffee mug. There are powerful Markdown editing tools with syntax colourizing and folding and other features (Scrivener supports it; so do vim and emacs and BBEdit) and you can write it using any plain text editor. If we could just get everyone to use it, there are powerful proofing tools out there—it's a plain text based markup language, so the entire panoply of programmer's tools are available.

But it ain't gonna happen. Novelists are not only not IT people; they are on average quite old (it's rare to sell a first novel before you turn 30: most working novelists are middle-aged). They are emphatically not early adopters. And the corporate IT departments that support Microsoft applications across the organization will happily go on doing so, and go on ignoring newer and better alternatives that generate newer and more exciting management headaches, until change becomes not merely inevitable but overdue by about a decade and a half.



This is a lovely example of something which (a) would be much better changed but (b) has a co-ordination overhead to changing which is almost impossible to overcome. There are plenty of other examples, ranging from physical infrastructure (it would have been much better if the London underground had standardised on a bigger tunnel diameter) to constitutions (which are deliberately made hard to change, sometimes much too hard). I wrote about some of them earlier this year at (in part inspired, in splendid circularity, by an earlier post here).


Word does integrate with Microsoft's version of CVS (TFS) and does support simultaneous editing assuming you put the right back end in place.

There are also an insane number of Word add-ins that are used by people that do all sorts of weird and wonderful things. Word (and Office) is an ecosystem that plugs not only into the other Office products, but into other stuff like Exchange, SharePoint, Active Directory Rights Management Services.

Word does an insane number of things that most people aren't aware of because they only see the product through their own usage scenario. (you can identify these people because they also claim that Word is basically the same with an updated UI as it was back in Office 97). Replacing Word requires providing a better replacement ecosystem - not just a replacement word processor.


Was retaining my ideological purity worth the risk of losing three months' work and screwing my next year?

Did you try opening the output document in LibreOffice, make a few changes and see how if fared?

No, because you misunderstand the problem (or rather, got it back-to-front): if I'd done the edits in LibreOffice, I wouldn't know if the output document was compatible until someone else opened it in Word. By which time it'd be much too late.


There's an assumption here that Word is perfect and infallible compared to LibreOffice. Maybe publishing has enough data on this topic to know that it is. But I've had Word eat long and important documents before - I'd be nervous of losing 3 months of work even _without_ bringing in 3rd party stuff. Just the Mac/Windows Word transition is fraught enough.


I think the suggestion wasn't that you actually change your workflow for this publishing cycle (the reasoning sounds very convincing to me). Just that after you've done that using Word, tried to open it in LibreOffice, make one or two more changes, and then asked a friendly proofer if it still looked OK for them in Word.

Basically, 10 minutes to provide a data point for the future if LO is up to the task.

That said, I was in a similar situation. I wanted to write my master's in LaTeX; but since the UoL had begun catering to non-programmers in MSc degrees too, the template was in Word. And adherence to the provided style part of the grading. So I abandoned the plan of using LaTeX and used LibreOffice.

(It turns out, just as anecdotal evidence, that the things that sucked most and those that worked out really well were exactly the opposite of what I expected. Reference management? I expected LaTeX to excel, but Zotero rocked my world! Graphic integration? Notoriously suckish in LaTeX, but Word/LibreOffice took the cake. "Oh you updated the graphic input, let's reformat everything!")

My main gripe is that text navigation and manipulation in LO sucks. I need to research if there's a vim input mode plugin.


publicstrategist @ 1
Sometimes called: "QUERTY" reasons, I believe ... it's set in stone, everyone uses it, & it's too late to change.
Fabric-widths, fixed in the 19thC as standard Imperial measures are worldwide - giving very silly numbers in the equivalent metric/mks measures....


There's an assumption here that Word is perfect and infallible compared to LibreOffice.


There's an assumption here that Word is bug-compatible with Word. Not the same at all. (Did you read my previous essay on why I hate Word and want it to die?)


A lesson for all fans of the Hydrogen Economy


I'm convinced we won't get the hydrogen economy. What we'll get instead is the zero-fossil-carbon hydrocarbon economy. (If you've got hydrogen you can reform it into methane and longer-chain hydrocarbons via Fischer-Tropsch synthesis, and we've got trillions of dollars of installed hydrocarbon handling infrastructure. The atmospheric CO2 spike is caused by releasing locked-up fossil carbon into the air; a reformer-based cycle would take carbon out of the air before releasing it again, thereby not adding to the problem.)

But that's a digression ...


I used to think about writing a decent wordprocessor in Visual Basic for Applications (ie, as a giant word macro). That's probably the only way of replacing word in publishing - do it without them noticing. I suspect you couldn't override enough of word's UI, but it's a nice fantasy.


I hate Word. (I hate it slightly less as of last weekend, when I finally managed to wrap my brain around some aspects of the way it handles styles, but I still hate it.) It is, nevertheless, on my computer, and will be on successor machines, for exactly the reasons given in the OP. I expect that when I finally have to abandon my beloved SmartSuite I'll simply switch over to Scrivener, but I will still have Word installed for ease of change-tracking exchanges with my publisher.


The thing I suspect #3 and #5 didn't get is that this is a risk management problem, not a "does it work" problem.

I've been bitten by exactly this risk - open .docx, make a massive number of changes, save, send back to source, source can't read it. And I didn't keep any kind of editable version that wasn't now-broken .docx.

Changing to .doc or other formats wasn't possible, either.

(There's now one research group I want to work with precisely because they write their FP7 proposals in .tex using Github)


It looks to me as though the essential problem is that the other links in the chain have done it before, and are confident that the work can be handled by Word. And, if anything unexpected happens, corporate IT have a chance of coping. And if they can't cope, it's Microsoft who gets the bad name, not an author with strange tastes.


Markdown looks a bit like a wiki language. History seems to be moving against wiki markup. I notice that several wiki systems are moving to using a "user-friendly" editor for all user interactions and hiding (or removing) the wiki markup completely.

Also, a problem with wiki languages was that every wiki seemed to use a different one.

So I wouldn't be optimistic about the chance of Markdown taking off. Although it might be nice if it did.


From my point of view as a manager in a corporate (well, university) IT department, the key phrase in Charlie's post was "management headaches". My team is relatively small at 50 people and even so I have many, many more things to worry about than whether we use MS Word to write our documents. (Leaving aside the fact that everyone we deal with also uses it).

As Charlie says, this is a major part of why MS Word is so ubiquitous. It's an example of the network effect combined with the high cost that a change would incur.


The thing I suspect #3 and #5 didn't get is that this is a risk management problem, not a "does it work" problem.

I got it alright, it was not what the question was about ;-)

I quite simply got curious. After having created this massive document, how does LibreOffice handle the strain? And how well did Word actually fare? Might be useful in the glorious future where one could have an actual choice, e.g. if the pervasive dominance of Word in the publishing industry were to be reduced.


Re: #15

Markdown looks a bit like a wiki language. History seems to be moving against wiki markup.
Not quite. What Markdown resembles, completely intentionally, is the informal plain-text markup that people use in text emails and other places where there's no formatting controls (with the notable exception of links, which only resemble plain-text links with the <> form).

This is quite unlike the various wiki languages, which were invented de novo and disconnected from any natural markup patterns.

You don't really have to worry about Markdown not taking off, because it's already won in a significant number of places.


Not quite, Markdown resembles the informal plain-text markup that some people use in text emails. Personally, I never 'got' Markdown, because it uses different conventions to me. The fact that I'm not alone in using these conventions (e.g. *bold* /italics/ (the slashes push the text over) _underline_) means that a lot of people aren't ever going to want to use Markdown. The variety of other text only systems around (e.g. reStructuredText, and AsciiDoc) suggest that I'm not the only one to prefer something else.


Back to the main topic, I've multiple (well, three) questions:

Why not use a Wiki (e.g. MediaWiki or a similar program with a decent history and version tracking system)? The only requirment then would be a modern web browser (made within the last 10 years would surely satisfy), and an Internet connection (though I accept that this can't be taken for granted, planes and blackouts being just two potential reasons). Oh, and the publishers would need to setup a server somewhere, and so have to hire one or two extra people.

Or, why not use plain text and a proper version control system? Wait, never mind, "too complicated".

Finally, what's to stop you from accepting changes, and then the other end not seeing where you made changes? Are they actually going to compare documents, or just trust the track changes feature? (It's something that strikes me as very odd, the way that some people just automatically accept the track changes feature, and think that if it's not recorded, then the change was never made. I'd hate to think lawyers do that, but it wouldn't surprise me if some of them do.)


Completely Off Topic
Sad news
Doris Lessing just died
Another one gone.....


All that being said, then, what would you suggest to an amateur who isn't averse to the idea of professional publication but can't really justify the up-front cost of MS Office?


Finally, what's to stop you from accepting changes, and then the other end not seeing where you made changes?

1. The instructions from the managing editor say "don't do that, or else we'll make you eat your dinner again."

And in case that isn't sufficient ...

2. MS Word supports user IDs and roles, and permits the managing editor to lock changes applied by the CE so that the author can't hit "accept changes" without reaching for a really big pair of pliers.


what would you suggest to an amateur who isn't averse to the idea of professional publication but can't really justify the up-front cost of MS Office?

If you aren't averse to the idea of [eventual] professional publication then you're not an amateur, you're an aspiring professional.

I'd recommend writing in Scrivener. Cross-platform (Windows, Mac, Linux, iOS under development -- note the primary platform is OSX, Windows/Linux versions lag in terms of features), proprietary but cheap (US $45), uses RTF as an internal file format but can also work on MarkDown text files, and has everything you need to generate publication-quality epub files if you don't end up going all the way to a professional publisher.

If you do end up working with the big guys, then for proofing you can try and use LibreOffice -- it does change tracking fine -- and if you're worried it's not going to work in MS Word, Microsoft offer a 30 day trial download of Office for Mac or Windows. (NB: If you have a real publishing contract and they want Word documents, I assume they're paying you. In which case a paid-for copy of MS Office becomes a tax deductible business expense.)

The probability of LibreOffice being up to the task was high, but the risk if a hitherto-unsuspected bug surfaced and rendered my change-tracked files indigestible by Microsoft Word was that I'd have to throw out three months' hard work and re-do from scratch. (Imagine Microsoft Word as Clint Eastwood, and me staring up the barrel of a .44 magnum. "Are you feeling lucky, punk?")

Just to be pedantic, that risk is a hazard not a risk, the risk is 1 - probability that LibreOffice would be up to the task.

Risk is the chance it goes wrong. Hazard is how much it messes you up if it goes wrong. Dirty Harry - 1 in 6 chance there's a round left is the risk, hazard is your brains all over the place.

If you juggle, the risk of you dropping the ball doesn't (theoretically) change regardless of the ball if the weight stays the same. Fill the balls with acid and some sort of ground-hitting magic trigger so if you drop one it sprays acid all over you and the audience and the hazard goes way up because what can happen if it goes wrong is far nastier.


The gravitational pull of Word is, if anything, even more intense for those of us who write academic non-fiction. The sausage machine works (more or less) the same way, but with the added requirement of pre-publication review, and yet more markup, by referees and (potentially) a series editor from outside the press.

Academic publishing also, in my field at least, frequently takes the form of essay collections. Take on one of those and, as volume editor, you're going to be passing chapters back and forth with 12-18 contributors, and potentially 1 or 2 co-editors, before you ever compile the result into a manuscript you can feed into the publisher's sausage machine.

I'm no fan of Microsoft Word, but the (relative) ease of interchange its dominance imposes has saved wear and tear on my sanity.


It was a smart move. When I went back to school to get my Masters, most of our paper submissions required sending the professor a Word document. And like writers, most of them were middle aged or older, and not IT inclined.

I started out using Open Office, but when my professors opened the files on their end it had sometimes unpredictable formatting results. Especially since they were all using different versions of Word on their own machines. After getting marked down on a paper because my citations didn't show up correctly when they opened the document, I got a copy of Word and made sure to save it in the 2003 version of the .doc file so it would be compatible...

Since I graduate the college has since migrated to Google's suite of applications aimed at academia. Provided they actually get everyone to use them, that could solve a lot of problems.


Thanks. My favoured fiction-writing tool to date has been gedit; most of what I've written to date gets posted to various forums so it's easier to just work in plaintext with a little BBCode. I'll give Scrivener's free trial a shot though, even if they lose a few points for being the official sponsors of International Torture Yourself With Pointless Arbitrary Quotas and Deadlines Month.

And I don't know if I would call myself an aspiring professional, actually. I certainly don't mind the idea of being paid for it but if it turned into a job then I don't think it'd be fun anymore.


I'm sorry, but much as I enjoyed your rant about Word and Word-related practices, it's a phrase in the last paragraph that gave me a lovely warm glow: "most working novelists are middle-aged". Thanks for that!

I'm a happy Markdown user myself (there's a host of nice new apps that use a 'pseudo-preview' approach to Markdown, effectively allowing you to see the effect of your codes in semi-WYSIWYG mode; my favourite on iOS is currently 1Writer, but there are plenty of others).

Markdown doesn't take long to learn. What does take a long time to work out are the pros/cons of the many, many variants on Markdown that have gradually evolved (from MultiMarkdown onwards).

1Writer, for example, uses absolutely standard Markdown syntax - but then adds a couple of decidedly non-Markdown codes to show tasks with tickboxes/checkboxes. Which means, unfortunately, that Markdown can no longer be regarded as a standardised writing environment (except for the most basic forms of rich text).

However, I'm sure a Stross-backed variant would quickly gain credibility... ;)


Charlie is SO VERY RIGHT!

I mostly use Open Office, and convert its .odt files to Word 95 or PFD or whatever an editor demands.

Which does not always quite work.

Only my having programmed computers since 1966 (yes, 47 years ago) makes me HATE Microsoft so much, but have to use its de facto standards because, you know, I like to be PAID for my writing.


Having just finished a degree I never had any issue with using LibreOffice (on Ubuntu Linux, either 10.04 or 12.04) to write my essays etc. (and Zotero to do all my bibliography formatting) and sending MS Word files, or PDFs (as appropriate). In fact, I was complemented a number of times on my excellent referencing! I didn't say that it wasn't me, it was Zotero.

I even used LibreOffice to do most of my team work assignments, with no complaints.

Of course, my essays etc. weren't using any complex formatting.

PS For those not in the know. Zotero is an excellent reference and bibliography manager. It's also completely Free (libre & gratis) and is written for academic usage. It is cross platform, and can be used with MS Word and LibreOffice/, and other word processors (though not with 'cite as you write' I don't think). Having just finished a masters, and using this tool throughout, I can't speak more highly of it.


It seems to me that LibreOffice isn't much of a win from OGH's perspective anyway. It doesn't involve paying the Redmond tax and the document format is (AFAICT) stable, but it still suffers from a Word-like confusing interface and styles/formatting-codes fight that he complains about in his earlier essay.


the risk if a hitherto-unsuspected bug surfaced and rendered my change-tracked files indigestible by Microsoft Word was that I'd have to throw out three months' hard work and re-do from scratch.

It's precisely this sort of thing that makes me explicitly version-track my important .docx files. I don't even care that I can't diff them, just knowing I can inspect the history and access any (committed) past version with ease is a huge boon to my peace of mind.


I am an "aspiring professional" novelist (to use your phrase, Charlie) and a seasoned professional nonfiction editor and writer. In my latter capacity, I saw the horror of Word's incompatibility with other apps. A freelancer submitted a white paper he'd written in LibreOffice. He kept on insisting that the output was 100% Word compatible even as I muttered and swore and cleaned up the many formatting errors that crept in.

In my capacity as a professional nonfiction writer, I write mainly for the web and use TextWrangler to generate simple HTML. I use Word when necessary. I prefer not to use Word, but it's not a big deal when I do.

My big problem with Scrivener: The iOS version has taken forever to develop. I need to look into syncing with iPad. It seems simple but with one or two tricks.

Scrivener has one or two other little quirks that make me suspect it's not the perfect tool for me.

I keep getting tempted by Ulysses, which has a lot going for it.

Charlie, what writing apps do you use on the iPad? Do you work on your novels on the iPad?


I have disqualified myself from NaNoWRiMo in perpetuity, following an unfortunate fit of hypergraphia in June, during which I extruded 51,000 words of fiction in 167 hours, i.e. one hour less than a week. (Then finished the project and had to spend a week in bed.) I really, truly, no longer have anything to prove: I know I'm slower than Philip K. Dick at his fastest (he allegedly wrote "The Three Stigmata of Palmer Eldritch" in a 72 hour long amphetamine-and-LSD fuelled frenzy) but I reckon a first draft of a 109,000 word novel in 18 days is respectable. (If I could write at that pace all the time it'd be equivalent to emitting my entire writing career to this point in 11 months. And burning out keyboards on a 10 week cycle.)


iPad writing tools: I don't do much serious writing on huge iPad because scrivener for iPad isn't here yet. But I use Pages for casual correspondence and Editorial — a really nice markdown-aware editor, whose features I am still getting to grips with — for the rest. Also Notebooks (can synch with a folder hierarchy on Dropbox, i.e. a scrivener project formatted for markdown). And I'm always looking for new tools. Like, er, Baldur's Gate II Extended Edition, due out on iOS real soon now.


That's quite a bit more than respectable, I reckon!

Though of course it raises the question of whether just hammering it out as fast as humanly possible and worrying about quality afterwards gets the book into a publishable state faster and more easily than setting a more methodical pace and doing your level best to get it right the first time. If one's even capable of the former method, which as I rapidly approach thirty I'm forced to admit I'm not anymore.


109,000 words in 18 days is indeed very respectable! Which novel was that?

I need to reread Palmer Eldritch. Dick predicted Second Life, in particular the way so many used SL for Barbie McMansions they'd never have in RL.


Editorial is nice. I've played with it and written a couple of articles in it.

My favorite writing tool for iPad and iPhone is Drafts. It's great for quickly drafted paragraphs which can be easily pasted into or programmatically sent to other apps, with optional Markdown conversion. I'm writing this in Drafts on the iPad now.


Speaking as a huge fan of LibreOffice ... I fear you made the right decision in the economic crunch.

However, I hope you will report any and every tracking bug you do find, with the culprit document ;-) They do love to kill testable conversion bugs ...


By the way, a ten year-old example of what can happen when you pass PC Word to Mac Non-Word back to PC-Word.

This happened a couple of times to me in grad school, when I was emailing a text to a professor for edits. The email I got back tried to fill my entire hard drive. There was a bug in the system that caused the document to attempt unsuccessfully to download until it filled all free space in my computer, at which point my computer froze.

The cure was to to contact university IT and get them to delete the email, so that my computer would stop trying to download it, then to delete the file from my system (a little harder than you might think, when it's bigger than the rest of your system combined, and dear sweet Windows wants to keep it in the trash until you flush it).

That particular professor told me it was my fault, sent the file two more times, then tried it with several other files before I convinced him (through an indirect threat of violence) to either use Word or write his comments in pencil on a printout, rather than using whatever passed for Track Changes on his computer.

Fortunately, I haven't seen this problem in years, but I have seen more recent documents with multiple reviewers rendered unreadable due to someone being stupid about where they put their edits.

The bottom line is that there's a lot to be said for using one system for review and editing when multiple people are involved. There's also something to be said for the dope slap with wind-up for people who think otherwise, mess up days of work, and blame others for their own IT shortcomings.


Pros and cons:

Hammering it out fast means the prose quality is poor and it needs a redraft. It also leaves gaps that need filling in (more reason for a redraft).

On the other hand, there's less time for the author to forget important details -- if you take months to years over a novel you forget key insights or stuff that was obvious becomes obscured. So a fast novel is a coherent, well-focussed novel.

(I'm happy with my 18 day wonder, but it ain't going to be published in that form -- it's going to sit on the shelf for a year then be thoroughly, painstakingly, re-worked, over a period rather longer than 18 days, before it's ready to formally submit to an editor.)


Wow - hadn't seen Markdown before, but it reminds me of some of the stuff I used to do with fmt, only better. And the comments on the Markdown web site give a lot of credit to Aaron Swartz working on it back when he was a teenager, yet another place that kid was making the world better.

And people still use XyWrite? Wow, shades of the 1980s early electronic publishing business!


NaNoWRiMo should just give you a lifetime achievement award. But, yeah.


Here's hoping eventually the new breed of cloud-based, Markdown-friendly collaboration platforms -- like Draft or Editorially -- gain a little traction in the publishing world.

Markdown is about as easy as it gets, and while both systems are very easy to use, they offer real version control, collaborative editing and commenting, and export into whatever you want. Draft is a little like the writer's version of Github (but easier), and while both are pretty clearly oriented towards producing text for online use, a slightly customized version would seem to offer a lot of advantages for a publisher.

Writers could write in whatever the hell they wanted; once the text was in place in the system, everything would be versioned and recorded and there would be far fewer issues with conflicting edits/comments (a bigger problem in the copywriting world than yours).

I'm introducing a pair of clients to Editorially now, and happily writing my copy in a text editor (originally Emacs [proving taste is alive and well], though once I learned I could get Emacs keybindings in Sublime Text, I've been seduced).

I prefer ReStructuredText to Markdown, but MD is where the momentum lies; either is far, far better than the dreaded .docx.


If you are currently using OpenOffice, is there any advantage to moving to LibreOffice, either from a general productivity or Word-compatibility standpoint?


There's an assumption here that Word is bug-compatible with Word.

How much of an assumption is that? I know I've been bitten when formatting and similar got screwed over because someone had a different default paper size, or even a different print driver. Version to version compatibility seems better than it used to be 5-10 years ago, but I'd still be a bit cautious.

That said, I use a horrible hodgepodge of Libreoffice on Mac and Word on (virtualised) Windows to manage my documents. And get bitten by strangeness about the same on both systems.


Microsoft gave away its stuff to the libraries, financed the libraries here in the U.S. to get hooked up to the internet. Google came with with it, o yes it really did. And publishers, as you say -- and not only the trade publishers -- decided Word was teh Word.

Trade publishing lived by public library bottom line purchasing.*

Now the public libraries are going with the tech winds, so is Word, one would presume.

Love, C.

* Recall until just how recently books directed to what was called then the juvenile market was almost entire public libraries, for example. Of course, I am speaking only of what I know from personal experience in publishing, libraries and academia, in the U.S.


Just because I love me my markdown, I wanted to mention Draft -- -- which is trying to be a collaborative editing platform with integrated version control. And it uses Markdown. It's a little rough around the edges (stackedit,, is a lot slicker, but without the integrated version control or collaboration features), but it's an interesting idea.

And yeah, probably not going to overthrow Word anytime soon, but I am fortunate that in my career Word has never been either common or important.


Seems like Google Drive should be a great replacement for this system: it's accessible everywhere, it has version control and and you can decide who has access to which parts. It also means everything lives on the cloud and doesn't take up any storage on your system or require backing up. I guess you could download a version before submitting it and download a version after everyone else had had a go at it (copy editors, editors, etc)?


What I can't understand is why the different formats aren't compatible. Isn't this where computers came in? Why is it so hard to get a computer to
1) read a text in Program A
2) note all editing codes
3) replace them with the applicable codes for Program B
4) carry on?
Why is it not straightforward?


Chris Borthwick, see OGH's previous post about why he hates Word. Essentially to get Word-like formatting to work in anything else and vice versa, you need to recreate Word and all its foibles. Hence, it's not straightforward.


Joel Spolsky writes about how the .doc "format" came about here, and explains why it's buggy to translate, emulate or even render with other software:

The real takeaway is that word -> rtf is actually a feature complete file format, for which rendering, editors etc can be built.

"Everything that Word can do can be expressed in RTF, but it’s a text format, not binary, so you can change things in the RTF document and it’ll still work. You can create a nicely formatted document with placeholders in Word, save as RTF, and then using simple text substitution, replace the placeholders on the fly. Now you have an RTF document that every version of Word will open happily."

Managing to get non-tech staff to emit and absorb said format in Word has proven to be fraught, in practice.


Chris Borthwick:

You'll find a description of the Word '97-'07 file formats here....

...first note that there's 210 pages of it, and then consider that my employer is currently in the middle of a project (not directly related to file format conversion, I might be able to tell you what we're doing or why we're doing it but I'm subject to an NDA which would make a laundry agent gulp so I prefer not to take the chance...) which involves accepting files in that file format, rigorously checking for "syntax" correctness, and then storing it in a format suitable for further processing.

We've had a team which started at 5 people and currently standing at about a dozen members working on it for going on for 2 years now and we can currently process going on for 40% of files.

Non-trivial is a considerable understatement....


JayGee: the article referenced at 53 above gives methods for those kind of conversions, allowing saving into formats which should be portable into the future.


LibreOffice is under active development and has shed a lot of dead code and fixed a lot of bugs in the past six months. Just download it and compare startup times if you want an argument for why you should switch from OpenOffice ...!

Why is it not straightforward?

In order to do this, you need to know what the editing codes are for programs A and B.

Unfortunately, there's no reliable way to know what the editing codes for Microsoft Word mean. Traditionally Microsoft has resisted publishing them altogether; even when it does publish them, some of the editing codes are explained along the lines of "This ... typically results in small caps which are smaller than typical small caps at most font sizes", which is pretty useless — there's no way to translate that into the editing codes of any other program.

If the documentation for the editing codes for one of the programs is missing and/or vague, translating between them will never be reliable.

I wish I were joking about "typically ... smaller than typical small caps at most font sizes".


Exactly, it is trivial to realise that HTML bold is <b> and markdown bold is [b], but when Redmond refuses to tell you what the relevant code in Word is...


Thanks for the rationale, Charlie. I also hate Word, but have to use it. Like Jules Jones I prefer Lotus SmartSuite (which does everything MS Office does for less than a tenth of the price and with far less bloat), but I've recently adopted Scrivener for novel writing. It was a steep learning curve, but I think I'm there now and I love the binder column, the ease of repositioning scenes and the synopsis facility.

Like you, I have to squeeze out the final output of the novel as a doc file. I'm resisting buying an updated version of Word, however. Word97 still works for me and I don't see why I should upgrade. (Don't get me started on docx files or programmes which are deliberately incompatible with their own older versions.)


It goes beyond even the problem of Redmond not telling you, though. Word's file format has a dozen or more different ways of indicating bold, each of which performs differently under varying circumstances. If there was only one, it would be much easier to reverse-engineer, and achieving bug-compatibility would be less completely pot-luck too.

(On which note - my experience is that some versions of Word aren't even reliably bug-compatible with themselves; they can wind up producing different output from the same document because the user has changed a setting in an unrelated application....)


If you're on a mac, one of the hidden gems in OSX (versions since 10.4, anyway) is a terminal utility called textutil. Textutil is basically a command line (scriptable) file conversion tool for word processing documents. And the formats it supports are: txt, rtf, rtfd, html, doc, docx, odt, wordml, and webarchive.

RTFD is an Apple embrace-and-extend of Microsoft's RTF: it's RTF in a directory (folder) with added image files. HTML in this case means conformant HTML 5, and it does a vastly better job of generating HTML from Word .doc files than Word itself does. ODT is OpenDoc, i.e. the format used by LibreOffice/OpenOffice, doc and docx are the Microsoft formats, and webarchive is Apple's container for HTML plus image files in a folder in a zip archive (as used by Safari/WebKit).

(One warning, though: I can confirm that .doc to .odt conversion via textutil seems to lose change tracking information and comments. I just tested this. Whereas opening a .doc file with changes directly in LibreOffice retains the change tracking and comment information. So be careful.)


I realise that; my point was simplified for the non-programmer. Please do not risk starting me on style sheets vs mark-up, since that's at least the 1 hour rant...


Actually, markdown bold is **b** or __b__ not [b].

[b] is boldface in BBcode, which is a predecessor to Markdown and similar to it, designed for message boards.

I forget what the Markdown and BBcode are for nitpicking.


This is a nice overview of some of the headaches. Personally, I think we were better off with paper. I think the editing done was better, the copyediting done was far better, and the only real savings in time is in the typesetting phase. There is a giant difference in the work I do as an editor working on screen versus on paper. On paper, I tend to take longer and be more thoughtful. Also, when working on paper, I'm not constantly distracted by what else my computer is doing, e.g., dinging every time mail comes in, telling me I have to scan it, telling me there's an update to this software or that. And, of course, no Facebook to procrastinate with (if I leave my phone off).


I think you realise that the symbols used were illustrative rather than accurate.


I work on academic non-fiction as a free-lance copy-editor, and use Word for the reasons adduced above. Following on from Comment 64 above: while the move to on-screen editing has saved time in the type-setting phase, it has increased the time I spend on titles as I have to make more layout/formatting/styling decisions than I did when I worked on hard copy.


Thank you. Just tried LibreOffice and it does seem to be quite a bit faster.


In a past life I was IT manager for a small manufacturing company here in the US. Having spent years fighting off bugs, viruses and malicious macros from MS Office, I spent considerable effort convincing my organization to give something else a try. We selected IBM/Lotus's Lotus SmartSuite office suite and implemented it.

Two years later we unimplemented it and went back to MS Office. It wasn't because it didn't work -- it was fine. The presentation software was a bit weak but I viewed that as a feature: we had less opportunity to fall into the Powerpoint black hole. The finance folks actually preferred Lotus 1-2-3 to Excel and even reading MS Office files sent to us by vendors, customers and business partners presented no serious problems. So why did we change back?

I call it old fart syndrome. Most other businesses our close-to-retirement managing director encountered used MS Office and over time our being the odd sheep in the pen grew into a more serious issue for him. I believe that he was personally embarrassed that we were doing something different. The final straw was when his wife had difficulty opening pictures of a distant grandchild on his laptop. Note that this has nothing to do with what office suite software is on the machine, it's one of "those" issues you naturally have to deal with when supporting the use of computers by folks who didn't grow up with them and resent having to use them now. Within a month we were given our marching orders and three months after that there was just a legacy copy of Smartsuite sitting out there, "just in case."

Sometimes IT isn't the actual problem.


THANK YOU, Charlie. I'd never heard of markdown - I need to install the perl package on my system at home.

Over the last 12-14 years, I've tried all sorts of programs alleged to do web pages, from Quanta to *bleah* Dirt, er, Word, and all of them produce crap, and there's not a single one that once you try to view the page... left justifies *everything* in the source.


I still wish WordPerfect, which was (and probably still is) the best WP out there (starting with 5.x). If they'd had a clue, the "reveal codes" showed you what would have been *perfect* HTML... but they never made it save that way, when I was using it (last was 6.0.c, mid-nineties). Oh, and a marketing dept that could market their way out of a wet paper bag....

Again, thanks for letting me know about markdown.



@22: If neither Scrivner or Libreoffice are to your tastes, older versions of Word (eg, Word 2003) are usually available quite cheaply on Ebay and provide near-complete compatibility with the current version, especially for the rather simple kinds of tracked changes used in publishing fiction.


The bit where WP5.2 was actively unstable under Windoze(for workgroups)3.11 doesn't seem likely to have helped.


Have I mentioned that I hated [pre-WYSIWYG] versions of Word Perfect even more than I hate Microsoft Word?

Bloated crapware. And that was Word Perfect 4.1! Could only edit two files simultaneously, lost them in event of a power glitch, didn't retain state between sessions (i.e. open it up and you'd have to reload whatever you'd been working on and then go to wherever you'd been working when the machine hung), and forced you to memorize a bizarre farrago of control keys.

After some spinning I settled on Borland's Sprint, and wrote several novel-shaped things using it over about a 5 year period.) Blazingly fast, could open up to 20 files simultaneously, maintained state so that you fired it up and found yourself right where you'd been previously, auto-saved to a journal file if you stopped typing for 2 seconds, had a powerful editor macro language and a separate macro-driven formatter system, and came with a cascading menu system that was easy to learn and had plenty of keyboard shortcuts. (Oh, and if you were migrating from Word Perfect or Wordstar or MS Word it could remap its user interface to emulate their menu structures and shortcuts.)


Back in the late 1980s I sold a few technical articles to computer magazines. Most of them would take straight ASCII text, but a few would only take Word Perfect "document" files. I think the street price of Word Perfect was around $400 then, bearing in mind those were 1980s dollars, and that was about a month's rent and utilities for a nice house.

To run a current version of Word, I'd not only have to buy a copy of Word or Office, I'd probably have to buy a copy of Windows to run it on and set up a new VirtualBox session for it, unless it would run under WINE.

It's not so much that the industry has converged to a monoculture, as that monoculture is so clunky and ugly...


You could apply similar arguments to circa 1990 versions of WordStar and Word.

For me Word's greatest mistake is styles, style sheets and especially linked styles rather than mark-up codes.

Back in 1995 I was editting technical papers by multiple authors using WP. Once I'd managed the challenge of persuading them to put legal section numbers into their sections and not worry about how they were producing multiple sections (in separate files) all numbered "section 1" I was well coushty because when I copyed and pasted the ultimate section, say 5 with 1..4 in place, it re-numbered itself automatically. I've never found that trick easy with Word, even if no-one accidentally changes an existing style!


Yes, the author can write their book in other programs. I use Scrivener. Other folks I know use XyWrite or Protext or GNU Emacs (there's no accounting for taste).

Some of us, koff, still write in WordPerfect 5.1, then convert into .rtf/.doc for supplying to the publisher. This may seem like a cockeye3d way of doing things but, particularly for large and complex texts, the savings in time and effort during the actual writing outweigh many times over the palaver of conversion.

Although I'm accustomed to the publisher sending the copyedited text in Word, the one for my most recent magnum opus (about 10% longer than that omnibus of yours, Charlie!) instead used Adobe Reader, which I'd never before realized was suitable for this sort of thing. The process was more of a hassle than with Word, but it was easier to see my corrections and what was going on.


Charlie @ 72
Give the man a sausage voucher!
When I was donig my MSc dissertation (19943/4), we were SUPPOSED to use WordPervert .... euw.
I finally used LOTUS1-2-3 - much better.
In fact, it's a pity that model did not prevail over MS Word IMHO


"I used a spreadsheet in preference to a word processing package"!!?


John Grant reminds me -- what working writer is still using XyWrite? What is their name?

I will buy that person a gift.

I loved XyWrite. What a wonderful program.


Gee, an awful lot of the comments here make it obvious people don't use the post-2007 versions of Word. Yes, Word's interface is still distinctive, but then so is pretty much every alternative's. And yes, Word's management of styles and templates is also distinctive, with less excuse. And only a fool prefers the inbuilt referencing features to the Zotero addin.

But Word's bloat these days is not excessive compared with its functionality, while complaints about file formats and hidden editing codes really are a product of pre-2007 Word. The later XML-based file format is pretty sensible and completely open; it ought to play with others very easily and reliably. Those complaints about Word formats having undocumented features are obsolete.

None of which is praise for Redmond's products generally. It's just that I think a lot of the kvetching here is about problems that they have - belatedly - fixed.

The later XML-based file format is pretty sensible and completely open; it ought to play with others very easily and reliably. Those complaints about Word formats having undocumented features are obsolete.

For what it's worth, I was quoting from the documentation for the XML-based file format when I was talking about editing codes explained along the lines of "This ... typically results in small caps which are smaller than typical small caps at most font sizes".

I guess whether this is sensible or easy is a matter of opinion, but it's hard to describe it as completely open, documented and reliable when the official documentation is so unhelpful as to be partly useless.

Not an obsolete complaint at all.


Correction in-line, bold:-

"Gee, an awful lot of the comments here make it obvious people don't examine the detailed file formats of the post-2007 versions of Word."

Seriously, Word is not so uniquitous that I haven't needed to examine the format internals in some years.


erratum - " now so ubiquitous that..."


Word's bloat these days is not excessive compared with its functionality,

Oh, really?

I remember grinding out book-sized works on a fully functional word processor in 1988 that fitted on 4.5Mb of disk space (including all dictionaries, thesauri, and drivers) and ran in 384Kb of RAM on an 8086.

Okay, not WYSIWYG. You want WYSIWYG? Circa 1990, I was using Ventura Publisher Pro to build 500 page technical manuals with circa 150 vector diagrams, tables of contents, embedded objects, indices, and so on. Came on (gasp!) 14 floppies, took 8Mb of disk space, ran happily on a 12MHz 286 with 8Mb of extended memory. Not terribly fast, I know, but throwing another 100MHz at it would have worked wonders.

You want power? FrameMaker on a UNIX workstation circa 1994. Ran happily on a 32Mb machine with a 75MHz processor, was way better than MS Word at complex compound structured documents, did true WYSIWYG, professional publishing grade layout, text composition, change tracking, revision control, styles ...

Editing text is not a task that requires bags of memory. A novel runs to about 600-900Kb of UTF-8 text. Even if you add hierarchical structural data like XML tags, we can somehow manage to parse the resulting data structures in not too many megabytes of bloat.

This was in an era when GNU Emacs was the butt of a joke about it's bloatedness: that Emacs was an acronym for "eight megabytes and continually swapping". Emacs, remember, incorporates a web browser, email client, full-scale Lisp development environment, and enough bells and whistles that some hardcore lispers have actually turned GNU Emacs into a UNIX Window Manager (graphical desktop environment). It doesn't get much more bloated than taking over the freaking operating system's user interface layer, but nevertheless the whole thing fits into about 80Mb and doesn't consume that much RAM.

Today's Word is a 1.2Gb download and it won't run on a machine that can't spare it at least 512Mb of RAM. The case for the prosecution rests.


LibreOffice vs Apache OpenOffice:

I just spent the last year polishing up the article on Wikipedia. I have read all the coverage on OOo/AOO/LO I can find from the past thirteen years, in several languages. I have followed the AOO and LO dev lists for that time. I have compiled LO from source (AOO I failed to). I will go so far as to call myself an expert on the subject.

* LibreOffice has a very active development community, multiple corporate sponsors, and several contributing companies whose daily bread is converting businesses and government offices to LO. For a stultifyingly boring topic (office software), the dev list is really very interesting - how excellent programmers cope with a seriously legacy codebase.
* Apache OpenOffice is functionally a spoiler project, backed by IBM and no-one else. They are popular because they have the trademark. LO is better in every dimension. AOO doesn't write .docx files, either. The dev list is "hey, something might happen soon!" The blog is excuses for not releasing things, and flaming LO. Nobody knows what the fuck the Apache Software Foundation was thinking taking on this turd.

You want LibreOffice.


h4nd: RTF is not in fact entirely compatible between implementations. The way it works is:

* Microsoft defines a spec
* Other people follow this spec
* Microsoft doesn't follow this spec, adding lots of ad-hoc whatever
* Other people then have to chase their tails reverse-engineering what MS Word actually does
* Interoperability then breaks, particularly between any two things that aren't MS Word.


If it's impossible to change, how did publishing go from paper MS to Word? Is there a history of that transition that could help with a new transition to an HTML5 workflow?
(I suggets that as it is the truly universal document format, and while in some ways as complex as Word docs, exists in multiple interoperating versions and has an explicit agenda to be usable for many decades).


You've missed the Killer Feature problem, and that is that openness isn't a killer feature for publishing. A ubiquitous editor and a format with comments and basic revision control are the killer features.

The publishers have one tool to point towards, and because of how popular that tool is, they don't have to write any documentation or do any technical support. And they can assume that some huge super majority of all their authors, copy editors and other contractors already have a copy. If you cannot sort yourself out, it ain't their problem. No one likes having a technical support division. Not even the tech companies (see: outsourcing).

HTML5 doesn't have a ubiquitous tool to write it in. Seriously, what are you going to point people at?

Second, revision control. Publishers use this to keep track of who did what. For HTML5, there are any number of bolt-ons, but they're all developer tools. I love version control, but they are not simple tools (part of my job is doing git tech support for our engineering team; even highly technical folks often get turned about by version control).

(I will give you that HTML5 does have comments. But that's 1 out of 3, and it is still missing a standard UI for people who don't want to read and edit raw HTML...)

I suspect changing away from Word as the standard would require Microsoft to go into significant decline. This process may have started in the last decade or so, but assuming that it did, they've got a lot of assets and will be around for decades to come. MS is deeply entrenched. Good Enough is the enemy of both Perfect and Better.

For the record: I use vim+make+git+plain text (having used POD and LaTeX in the past, depending on exactly what I'm up to; I like markdown, but haven't been using it more than casually). I have roughly the same feelings as OGH with regard to Word.


Mr. Stross,

Re: Word bug compatibility (though I think you already know):

Scott Hanselman: I've got two Microsoft Word documents that I wrote in Word that I can no longer open in Word as Word says "Those aren't Word documents."

Version control. Please.


You couldn't make it up, could you?


I've managed to avoid word for most of my adult life, but now I find myself having to write a few assignments for a course on it.

Can anyone explain the rationale for the annoying habit it has of "helpfully" extending the selection highlight. Libreoffice (Which I use at home) doesn't do this, it just selects whatever I mouse over without trying to grab extra words or characters.

In other news, I started playing an augmented reality game the other day, It has me walking around zapping "portals" at random GPS marked real world locations. Very Halting State.


That extends throughout MS Office. It's M$ implimentation of a Unix GUI's "smart selection" where 1 click places the cursor, doubleclick selects a word, triple a line and quadruple an entire file.

It basically assumes that because you've dragged, you wanted to select the whole of the words you started and ended the drag on. You can turn it off, but the location of the "feature" varies with which version of Wurd you've had inflicted on you. File -> Options seems likely.


If it's impossible to change, how did publishing go from paper MS to Word? Is there a history of that transition that could help with a new transition to an HTML5 workflow?

This only happened in the past few years. Circa 2005 for the small publishers, circa 2010 for folks like Penguin and Macmillan.

To do a transition to HTML5 would require ubiquitous, good, WYSIWYG, HTML5 DTD-compliant editors -- which are virtually non-existent. I suspect it will happen eventually as the industry moves towards primarily delivering epub files rather than paper (as epub format is a containerized, DRM-enabled version of HTML5/CSS3), but it's still a long way off (years to decades).


Second, revision control. Publishers use this to keep track of who did what. For HTML5, there are any number of bolt-ons, but they're all developer tools. I love version control, but they are not simple tools (part of my job is doing git tech support for our engineering team; even highly technical folks often get turned about by version control).

I will note that revision control is a very minor consideration in trade fiction publishing.

95% of books have a single author. Of those books, probably 99% go through writing/editing/revision stages, but only one person works on the file at a time -- the workflow is strictly linear. There might be forks for different edits in UK/USA context, but that mostly only happens in children's/young adult lit these days -- adults in the UK are expected to be okay reading US grammar/spelling. It might in future be useful to support forks for translations, but in general translation rights are sold to local publishers and the translation process is effectively the creation of an entirely new book in a different language: there's no continuity. Finally, there might be fixes for typos/bugs in the text (previously, only when a book went from hardcover to mass market; these days it's both easier and harder to fix typos in an ebook). But overall, the process is usually entirely linear: rcs or sccs are massive overkill in terms of the functionality they provide for novel publishing.


A very useful term to describe Microsoft Word's smart-selection process is hlepful.

Someone (or some piece of software) is being hlepful when they over-anticipate a requirement you have and attempt to help you achieve it by offering counterproductive or just plain irritating advice or services.

Example: if you announce on Facebook that you've been diagnosed with some chronic illness, a whole bunch of your FB-friends will exhibit hlepfulness by pointing you at crank diets, herbal remedies (which may or may not work), or homeopathic solutions. Usually without actually understanding the nature of your condition.

Another example: you ask how to compose a style sheet in Nisus Writer, and a bunch of folks hlepfully explain how to do it in Microsoft Word (which you do not own).


(My filter bubble of) The Internet appears to settling on halping as the word of choice for that kind of situation. I'm regularly seeing it used as a closer to comments, to indicate the ludicrous misapprehension of the situation - and consequent completely useless (and possibly illegal) advice - preceding it was entirely deliberate.


I see a some talk here about people using Lotus SmartSuite.

Is there a version that will run on a modern PC? I ask because a friend created a lot of support materials for our shared RPG campaign, using it back when we both had XP boxes. It seems impossible to get it to work on any of the later windoze releases though.

Alternatively, does anyone know about some conversion tools that can port data from SmartSuite files to something a bit more generic?


The word I see used for "hlepful" is "helpy".

e.g. "That's not helpful, it's just helpy."


Geeks are very prone to helpiness, aka geeksplaining.

"I need to do A, B and C. I have D, E and F."
"Oh, you totally don't want to use F, you must be an idiot to use that. I use G, H and I. Much better."

G, H and I are incompatible with C, and don't work too well with A or B.


Lotus Smartsuite stalled over a decade ago, and the rights are held by IBM. Symphony lost the support for Word Pro file format, and was quietly supplanted by Open Office. Most of the rest of the Smartsuite stuff was withdrawn back in May 2013.

If you want to wander the less legal download sources, you might still be lucky, but can you trust them?

LibreOffice certainly was able to read .lwp files at version 4.0.1 It took me a while to find one of my old data files to check, but it did open.

Open Office does not seem to support the format.


I don't want to be merely contrarian, but I've mostly enjoyed using Word.

I felt closest to it with 5.1 on the Mac.

But I use the current version on Win7 very happily.

Intellectually I understand some of the criticisms here, but as a user (and ex-professional journalist), it does it for me.

Different strokes etc.


Is that yours? That's actually pretty good. 'Hlepful', that is.


I prefer 'hlepful' because it reminds me of 'heffalump'.


My Office 2010 (not just Word; I use Access and Excel too) experience was somewhat improved by the discovery that ctrl+key shortcuts still work, and by someone telling me about the customisable quick access toolbar (File -> Options -> Quick Access Toolbar).


I have a copy of Smartsuite somewhere, dating from the Windows 95 / NT 4 era that will, iirc, run on XP. Vista breaks a lot of software from that era, though, so it likely won't run on anything newer.

Your best bet is probably a temporary install on an XP box and convert the docs to another filetype. There are still plenty of XP machines in use, so you're bound to know somebody who has one.


Publishers hate Word just as much as everyone else. Really, trust me on this. But they love the power that change tracking gives them, and the idea of going back to proofreaders' markup simply isn't going to happen, as you well demonstrate.

But Word is ugly, unreliable, cumbersome, and prone to sulking. The bad news is that most other editors are nearly the same, especially those which try to mimic Word.

The good news is that HTML5 (and/or/EPUB3) is going to be the file format which will save humanity*, so expect to see a lot of editors start offering it, including (eventually) Word. This won't make the interface any better, or get rid of change tracking, but some added competition might make Microsoft look again at that thing called Usability.

* For some value of "humanity". And some value of "save".


I can now confirm that the current version of LibreOffice will read .lwp files produced by the Millenium Edition of Smartsuite.


About this Entry

This page contains a single entry by Charlie Stross published on November 17, 2013 11:01 AM.

The revolution will not be hand-stitched was the previous entry in this blog.

A likely tale is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.

Search this blog