[This week’s Communications column for the Vanuatu Independent.]

[Yes, it’s a re-hash of this rant. ed.]

As a computer geek, I’m supposed to be suffused with enthusiasm and excitement over the features of the latest software. By rights, I should be the one carrying the techno-tablets down from the mountain, telling you how the latest in frobnalising ephemetry is going to change everyone’s life. I’m the one supposed to show you where to sign up and what to do with it once you’ve got it.

I have a confession to make: I hate most software.

90% of software is crap. As author Theodore Sturgeon famously said, that’s because 90% of everything is crap.

I save a particular loathing for word processors. For any but the simplest tasks, their interfaces are utterly ridiculous. I haven’t liked a word processing interface since WordPerfect circa version 5, which ran on DOS (remember DOS?). If I had my own way, I’d still be using it.

Why, after over 20 years of development, do word processors suck so badly? Mostly, it’s because of the WYSIWYG approach. What You See Is What You Get, besides being one of the most ghastly marketing acronyms to see the light of day in the digital era, is ultimately a lie. It was a lie back in the early 1990s when it first hit the mainstream, and it remains a lie today. The fact of the matter is that trying to do structuring, page layout and content creation at the same time is a mug’s game. Even on a medium as well understood as paper, it’s just too hard to control all the variables with the tools available and still have a comprehensible interface.

Ask yourself, how many times have you had to reprint a report, just because you have to hit ENTER a few more times just to get a title or a table onto the right page? How many pointless hours have you spent keeping a manually generated table of contents up to date? Make a change on page 3, and now you have to re-calculate the page numbering on the entire thing.

The real sin that word processors are guilty of is not that they’re trying to do WYSIWYG – okay it is that they’re trying to do WYSIWYG, but they way they go about it makes it even worse. Rather than insisting that the user enter data, structure it and then lay it out, they cram everything into the same step, short-circuiting each of those tasks, and in some cases rendering them next to impossible to achieve.

The concept of a workflow seems to be lost on software designers.

Learning how to write, then structure, then format a document (or even just doing each through its own interface) is easier to accomplish than the all-in approach we use today. For whatever reason, though, we users are deemed too dumb to create a document without knowing what it’s going to look like right now, and for our sins, that’s what we’ve become. And so we are stuck with word processors that are terrible at structuring and page layout as well as being second-rate text authoring interfaces. They do nothing well, and many things poorly, in no small part because of the inherent complexity of trying to do three things at once.

Advanced users know this already: You can apply styles to a document which, in theory at least, allow you to avoid this kind of thing. But nobody takes the time to teach you this, because, honestly, they want to create the impression that all you have to do to create a professional document is to start writing.

Ask yourself: Which is more important – the words you write or the font they’re in?

It doesn’t help that every word processor’s technical implementation is poor. The Word 2003 document format is little better than a binary dump of memory at a particular moment in time. That means that even if you do apply styles and proper formatting, they’re still prone to getting corrupted, or interpreted differently from one version of the software to the next.

There’s no changing any of this, of course. The horse is miles away, and anyway the barn burned down in the previous millennium. We’ll be dealing with stupid formatting issues for years to come. We will continue to be unable to properly structure a document past about the 80th percentile, which is worse than not at all. We will continue to deal with visual formatting as our only means to infer context and structure, leaving us with very little capacity to do anything useful with the bloody things except to print them out and leave them on someone’s desk.

We could be doing so much more. Wouldn’t it be nice to be able to take a bunch of documents gathered during the course of a piece of work and simply dump them into a database? Wouldn’t it be nice if we could share the contents of a folder on the Internet in such a way that everything is automagically organised, searchable and yes, even moderately useful to others?

It’s possible, you know, even easy. Or it would be easy if we could just have software that didn’t make it easier to ignore useful formatting tools than to use them.

Doing proper formatting is not rocket science. It’s a simple matter of taking a little time to set a few properties before you start writing. Once you’ve got decent styles, you have them for life. From then on, if you want to use a heading, you don’t have to change the font size, then bold your text, then tab it in. One little click will do all that for you.

Much of the web works this way. And despite the fact that most early web authoring tools tried to turn themselves into word processors (somehow they imagined that their warts would be seen as beauty marks), the tide has since turned against them. These days, style and content are understood to be two different things.

If I had my way, we’d just stop using word processors at all. We’d start doing everything on the web and never print again. I’m more than half serious about this, actually. At least on the Web, the idea that content and presentation are separate things isn’t heresy. At least on the Web, we can archive, search, contextualise, comment, plan, structure and collaborate without having to wade through steaming piles of cruft all the time.

At least on the Web, we can choose which steaming piles we step into.

I wish I could start recommending that people stop using Word as an authoring medium. There are far better, simpler tools for every task, and the word processor has been appropriate for exactly none of them for too long now.

Sometimes you have to destroy the document in order to save it.