Thoughtful, detailed coverage of the Mac, iPhone, and iPad, plus the TidBITS Content Network for Apple consultants.

The User Over Your Shoulder - Malign Neglect

More than ten weeks after U.S. customers began receiving the Word 6.0.1 update, it became available in New Zealand. Anyone calling Microsoft from New Zealand was told they had to deal with Microsoft NZ, and then - airplanes be damned - it was popped on the slow boat to save a buck. Amusingly, Microsoft NZ then overnight-expressed us our copy, as if this could somehow make up for the irrationally long wait.

After the usual harrowing installation procedure, I bravely but tremblingly turned Super Boomerang back on, and started Word; so far, no crash, so perhaps the thing I was most unhappy with is fixed. Meanwhile, as I'm trying to type while keeping my fingers crossed, I glance over some of what Microsoft has to say about the other improvements in this update:

  • "Word Count was significantly slower in 6.0 than 5.1. Performance is now par with 5.1. Fixed by changing how we check for an escape out of the action."
  • "The View menu took longer to drop than other menus. The extra time was used drawing the bullet symbol. We now preload the bullet."
  • "The MS LineDraw font was corrupt. It has been replaced."
  • "Other applications could not open Word 6.0 files saved as Word 5.1 files. Saving a file as Word 5.1 left the file with a W6BN File Type. The file now gets a WDBN File Type so other applications can recognize the file type."

And so forth. Microsoft is implying I should be grateful for these fixes. But why? After all, what's being fixed in each case seems to have been a pretty silly error in the first place. Microsoft isn't rescuing me from anything except itself! Perhaps Microsoft thinks of this update as valiant customer support, but to me it suggests that Microsoft did sloppy work and left its customers to act as beta-testers.

Don't get me wrong; I'm not a Microsoft-basher. I actually tried to write an article for TidBITS taking the decidedly minority tack of praising Word 6.0, as to its overall design and functionality at least. No praise, though, for this attitude that releasing an update is as good as getting it right the first time, which seems to be part of a general sloppiness among many commercial software developers. In the rush to publish, the user is lost sight of, while also being taken advantage of: by releasing buggy software so as to beat the competition to market, the developers raise capital to fix it at their leisure. And they have the leisure, because once you've bought bad software you're hooked, waiting for the fix.

Granted, no software is bug-free (as every programmer knows), and software held back too long from release is vaporware. But many of the mistakes which the 6.0.1 update fixes were downright shoddy, and this is why I attribute them more to an attitude gone awry than the normal vagaries of the development cycle. And don't forget, users who upgraded were paying big bucks to make this shoddiness part of their lives.

The makers of my other love/hate word processor, Nisus Writer, are not so different. Subscribers to the Nisus mailing list, probably among the program's most thoughtful and intensive users, actually got organized enough to come up with survey of improvements they'd like to see. However, they were told quite explicitly by a representative of the company that their views had been largely cast aside at a meeting as being too marginal (revealingly, by the marketers and engineers, not by the techies).

And another representative of the same company recently wrote me personally and asked me to stop contributing to that mailing list, because my criticisms of the company were annoying the other readers. Funny how no readers had written directly to me about it; and never mind that the review of Nisus that I wrote with Adam in TidBITS-116, TidBITS-117, and TidBITS-118 - which the company happily distributed with its demos (without telling me) - won and still wins the program many converts. It seems that when my admittedly enthusiastic tone of writing extends to certain home truths, I'm anathema.

And what are these home truths? Mostly that the new version, Nisus Writer 4.0, for which users shelled out what I think is an outrageous sum, is demonstrably much slower and considerably buggier than the version it replaced (Nisus 3.47). True, it's been getting better after several maintenance upgrades, but that's largely thanks to the vociferous complaints of paying customers, like me, who found the bugs and drawbacks in 4.0 that the company missed (or deliberately set aside).

Well, I'm sorry. I think putting customers in the position of paying big bucks to act as unwilling beta testers for Nisus - or for Microsoft, or for anybody - is outrageous. And I think we've paid for the right to scoff. Of course one should resist the ever-present tendency to flame incoherently. But if the relationship between developers and customers has gone wacko, only the clamor of the customers can do something about it.

There was a day, not long ago, when the fact that your computer did anything at all seemed a miracle. Your jaw dropped in admiration, and you felt love and warmth for the dedicated artists who turned a dead box of chips and wires into ingenious magic. I'm not saying that that day is entirely past, but I am saying that the gee-whiz factor can now be tempered with a considerable dose of practical reality. The simple fact is that computer programs are not magic but artifacts, mere human creations with a straightforward functional purpose. If you've paid for them, they can (and should) be viewed and criticised like any other commercial artifact like a house, a car, a shirt, a cigarette lighter. If it's shoddy, if it doesn't do what you need it to do, by jingo it's your money - you shouldn't have to stand for it.

Yet we do stand for it. I constantly get email from folks who have noticed a bug or a shortcoming in a program, and I say: great, and have you written the developers about this? Too often the answer is "No," or "Gee, I didn't think of that." I myself have more than once shelled out a couple of hundred bucks for software I found so buggy as to be unusable, and neither returned it nor complained. Why?

I suspect it's partly because there's a tendency to hope for developers to notice and fix their mistakes, as if they were with you, watching benevolently from inside your computer. This scenario, even with the best face put on it, is unrealistic; as Dave Winer pointed out in TidBITS-280, the notion that a corporation is going to generate good software is irrational. Another problem is that it's surprisingly hard to describe - objectively and helpfully - a problem or shortcoming with a piece of software. And yet another problem is that most software companies have no clear ingoing communications channel: the folks who mind the phones or the email, I find, are usually not responsible for the program itself, and are either there to act as a buffer between you and those who are, or else, if they actively try to advocate your view, are just one more voice apt to be lost in the corporate storm.

I have no solution. There must be give on all sides. Developers must break out of their present isolationism and genuinely respect their users, actively seeking and facilitating cooperation with them. They ought to especially pay attention to those who show expertise, which is usually accompanied by a visionary commitment that the developers ought to value, not marginalize. Users must stop expecting either that software problems won't exist or that they'll just go away, do less flaming and less sitting on their hands, and make an effort to communicate cogently and persuasively with developers. And, I suppose, software prices should become more realistic. Think how different your loyalties, feelings, expectations, patience, and response would have been if the upgrade from Word 5.1 to 6.0 had cost $20.


PDFpen and PDFpenPro 9 add 100+ enhancements to improve your PDF
editing experience, with annotations, Tables of Contents, and more
export options. For PDF reviewing, editing, signing, redacting and
exporting, PDFpen has you covered. <>