Used and abused

I recently went to a training day where a very earnest presenter tried to tell us that his software could do amazing things. And there is no doubt we were amazed, but perhaps not quite in the way he was hoping.

That the software was glitchy became apparent very early on, as it failed to do a number of the things he was trying to show us. He clearly had a set of carefully scripted demos that he knew would work, but even those produced somewhat unreliable and unpredictable outcomes.

What really had me going for his throat by the end of the day, though, was the number of times one of us would have a problem with the software and he would greet us with “oh, that’s because you’re doing it wrong.”

And no, I wasn’t grumpy because I am a grammar nazi (although I confess I do have tendencies in that direction). It was the overwhelming implication that we should be changing our way of working to suit the software. We should not expect to be able to do things the way we have always done them. We should alter our entire workflow in order to do precisely what the software wants, the way the software wants it – whether that meets our needs or not.

There, in a nutshell, is the sheer arrogance, ignorance and, yes, malpractice of the computer industry, all wrapped up in one neat little package.

It’s not our fault. You’re just doing it wrong.

I have a PhD in Computer Science (in usability, in fact), and I have lost count of the number of technical support people who have tried to talk to me as though I am a 3 year old standing near a wall covered in texta, feigning ignorance. It’s always “What did you do??” in weary, exasperated tones.

Even with my relatively high level of technical understanding, it can take forever to persuade them that it wasn’t some dumb thing that I did, but that there really is a problem with the technology. How does the average user cope with that? Usually by assuming that they did, in fact, do something wrong.

If I only had a dollar for everyone who has ever told me “I’m no good with computers”. I always give the same answer: “No, computers are no good with you!”  The best tools do exactly what we need them to do, without interfering with our workflow at all. We don’t need to compensate for them or understand them. They just work. Software all too rarely fits into this category.

For a simple example, picture a door in a public building. It can only be pushed, not pulled, and it has no door handle. Just a flat plate where the doorhandle would normally be. The whole “pull/push” quandary is missing with this door. You can only push it. The very design of the door says “push”.

Of course, software is intangible and conceptually complex, so it’s not easy to make it as obvious as a pushable door. But think of the number of doors you come across in your day that can only be pushed, yet have pull-able handles on them. They need labels to tell you what to do – and who stops to read labels when walking through doors? As a simple rule, if it needs instructions to tell you how to use it, then the design isn’t good enough. Good design is distressingly rare.

In the software industry, it is all but absent. In a way I don’t argue with that – as users, it is up to us to do our homework and choose the most effective and usable piece of software we can get. In general we don’t – we buy the most effectively marketed, the cheapest, or the most famous, for which we mostly have ourselves to blame. Caveat emptor and all that. But what I do find incredibly objectionable is that culture of blaming the user when things go wrong. A little respect would be a fine thing.


7 thoughts on “Used and abused

  1. Daniel May

    Oh, you just need to read the printed training manuals and go on the training course. The affordances of training materials are clear: open, turn page, keep turning, keep turning …

      1. Daniel May

        It’s in the manual. Page 232. Page *not* left intentionally blank. Read it. Remember it. Memorise it. Be perfect: software demands it. That is all.

  2. Joe

    I agree entirely that the need for a bigger instruction manual for a similar level of function tends to indicate a poorer system. A system should work within the most common terms of its target audience, and present a consistent set of visual guides. Given a user who has the assumed set of requisite knowledge, the things in front of them should “make sense”, doing something “sensible” should work, and trying to do something “not right” shouldn’t *break* anything.

    The system being demonstrated sounds like it didn’t meet these basic criteria, and the presenter sounds either desperate or blind or too far buried in a love of his own product.

    There’s a “but”. A system can only encapsulate an agreement between user and builder. First they need a common understanding of a frame of reference. A system you are looking at may simply be too far from your target frame of reference, so of course trying to “do what you wanted” is outside their conceptual scope. But also to the extent that you or your business have a way of doing things that is “different from anyone else’s” it’s entirely *un*reasonable to expect that a non-custom system can cater to everyone’s foibles. As a software builder, if we can agree to achieve X *this* way, I don’t need to cater for 500 *other* ways. Of course, if I only got to speak to 50 other businesses somewhat like yours, I only have their ideas to get to “X”.

    But maybe your peer businesses managed to get “X” anyway, even if it wasn’t quite your way.

    I *am* a proponent of adjusting business workflow to suit the best-match software you can find. If you *must* be highly system dependent my advice after I.T. contracting through about 35 businesses is: put your ego away. Research to find a 90%+ fit for your business, buy it, and yes adjust your business to it rather than trying to adjust it to your business. Assuming you did your homework, the new workflow will achieve your purposes even if it’s not like the old workflow, and the capability you bought resulted from the common experience of a group of businesses like yours who have each contributed to build the “90% fit”. The last 10% you wanted wasn’t valuable enough to “businesses like you” that the history of your predecessors combined ever made it happen. The last 10% will cost more than the first 90% combined (for good reason) but yield less benefit, and you’ll be wearing the cost yourself instead of sharing it with everyone else who bought the system.

    Unless, of course, you are a business with no peers.

    While a door has a very simple function that we all grew up with, tools of even moderate complexity always require *training* … a calculator, a drill, a telephone, a vacuum cleaner… my cupboard is full of lengthy instruction manuals for “simple” tools. By the time you reach the level of a tool that is trying to replace even mundane parts of a professional job you need a common (and restricted) set of assumptions or the system state scope becomes unwieldy.

    1. Joe

      ps “I’m no good with a paintbrush.” But it’s nothing that a combination of interest training and practice wouldn’t rectify. (Though, it’s pretty hard to get started without someone to guide me.)

      Or maybe the paintbrush is no good with me?


  3. Daniel May

    Further riffing off Joe’s post:

    Purchasing software (rather than developing it bespoke) should bring cost savings. Unfortunately, no tool fits exactly what we need so we either bend the business or bend the software. How much you do this is the tradeoff: you may bend the business to achieve tremendous savings (but this could break the business). Or you may bend the software to such an extent that you might as well develop bespoke software. Pick your poison! A tension that businesses need to navigate, but they can often get trapped in one or the other valley.

    On another point: it is good to not assume people will read a manual before playing with something or using it. But sometimes you just need some instruction bits to go with your artefact. That’s life. But yeah … how come there are instruction sheets that are just really easy to use? how come there are tools that you do need to customise, but it’s just so much nicer to customise those tools and use their supplementary material?

    MacPaint never made anyone into artist who wasn’t, but gee, it sure was pretty awesomely easy to use cf whatever else there was then!

    1. lindamciver

      Hey Joe,

      you’re quite right. Sometimes it is worth compromising your working style to get the benefits of a good bit of software. But it does have to be worth it – and all too often it’s not.

      And of course, the more complex a bit of software gets, the more likely you are to need some help. Seriously useful stuff is never going to be 100% walk-up-and-usable. Nothing wrong with providing good documentation. It just makes me cross that so often the documentation is provided as a substitute for usable design. AND the documentation is usually hopeless as well. of the: Menu Item X does X. variety. Gee. Couldn’t have guessed that!!!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s