Prescriptive Design and the Decline of Manuals

Instruction manuals, and instructions in general, are incredibly important. I could be biased, since part of my job involves writing instructions for systems, but really, they’re important!

As this look into the historical importance of manuals makes clear, manuals (and instructions) make accessible professions, tools, and devices to anyone that can read them (which, admittedly, could be a hurdle of its own):

“With no established guild system in place for many of these new professions (printer, navigator, and so on), readers could, with the help of a manual, circumvent years of apprenticeship and change the course of their lives, at least in theory.”

However, as the economy and labor system shifted, manuals did too:

“in the 1980s, the manual began to change. Instead of growing, it began to shrink and even disappear. Instead of mastery, it promised competence.”

And nowadays, manuals are very rarely separate from the devices or systems they seek to explain:

“the help we once sought from a manual is now mostly embedded into the apps we use every day. It could also be crowdsourced, with users contributing Q&As or uploading how-to videos to YouTube, or it could programmed into a weak artificial intelligence such as Siri or Cortana.”

By attempting to make things more seamless, integrated, and easy-to-use, devices not only technologically lock out users with copyright and DRM and pre-soldered hardware, they also provide short, “quick start guides” in place of a true manual before you can begin using a machine.

“Furnished only with a manual of one or two pages, users soon reach a comfort zone, a knowledge plateau from which they tend not to wander. The aggregate effect, culturally, may be that less is less. The less we’re inclined to know about our devices, the more beholden we are to the manufacturers that make them, and the more we offer control to those who, for good or for ill, know more than we do. If manuals began as great equalizers, then their disappearance should at least give us pause. By dispensing with them, we could, consciously or no, be setting the stage for something few would relish: a society divided.”

This is certainly preferable by a large portion of people (judging by the types of calls and help requests I would get during my years of customer support). People want devices to just work, so they can move on to what they need to do–buy groceries, file their taxes, and yes, write documentation.

The trouble is that we’ve built a near-incomprehensible technological system, and now plan on using that technology in nearly every device on the market today. The internet of things is here, it doesn’t always work right (video), and the design isn’t usually easy to understand.

Design, when algorithmically-controlled to the point of being a total black box to a person, isn’t exactly helpful. In those cases we’re stuck wondering, Why is my house this temperature? Why is Facebook showing me this? Why am I being shown an ad about asphalt mixtures?

The rise of the algorithmically-sorted life isn’t exactly welcome, even if it purports to be helpful and make things easier to use (okay maybe the asphalt mixture ad is under no such illusions). However, that doesn’t necessarily make all prescriptive design bad (or do you think it does?). Some prescriptive design is basic and helpful–make it easy and clear what the person should do next.

Google has started using insistent, prescriptive design when it comes to various kinds of security warnings.

  1. Gmail warns for suspicious email.
  2. Search blocks “crapware” and malicious software download sites from appearing in search results.
  3. Chrome blocks malicious software from being successfully downloaded at all.
  4. Chrome improved its warnings for SSL (secure sockets layer). Oddly, Wikipedia no longer has a dedicated page about SSL, instead directing people to the page for TLS (the encryption method that superseded it). The page now has a warning because it’s too long.

Google is taking two different steps with these design choices.

  • In the case of the spam emails, when they are identified as malicious, Google tells you exactly why you should be suspicious of the message–perhaps recognizing that more information in this case could mean more obedience to best practices, e.g., don’t click on a phishing email.
  • In the case of the malicious software downloads and the SSL warnings, Google takes the step of making it hard for you to accomplish something (click through a warning, download a file, click on a search result) with the end goal of preventing you from doing something harmful (putting your data at risk in a man-in-the-middle attack, having your computer infected with malware or a virus).

This isn’t so bad, as long as the design choices are made with strict user benefit in mind (and to that end, who gets to decide what is beneficial for the user?) As the presentation on the SSL warnings makes clear, the goal is to change behavior, since changing comprehension alone didn’t warrant enough of a change.

Google may be parenting the Internet with their design choices, but in this case, the Internet could use a little bit of growing up (and so could its users).

Writing error messages is an art in and of itself. A guide on how to write error messages identifies that:

“A good error message has three parts:

The problem – explains that an error has happened;

The cause – explains what caused the problem;

The solution – explains how to overcome the problem.”

This is even tougher to do when you need to keep them as short as possible.

Error messages today can be overly generic, and even purposely unhelpful in the hopes of making a system more secure. Two types of errors are especially guilty of this:

With a little bit of work, these workflows can be improved. System security can be strengthened without providing vague, unhelpful error messages to a person. I’ve attempted to log in 10 times in a row, getting the “username or password is incorrect” message and re-entering my password (since I can’t see my password when I enter it, it just shows up as little black dots). After 10 tries I finally realized that I had mistyped my email address on the first try, but never thought to correct it (it’s the shorter of the two, after all).