The plague of “smart refrigerators”

I think we’ve all by now heard of the mad, magical future in which your new refrigerator will have the intelligence to know when you’ve just run out of milk and will automatically order more for you. A perfect digital servant, that just happens to knew exactly which items in your fridge you need repeatedly, at a perfect frequency to match their consumption. But what about stuff I bought once and no longer want? What about the milk that went bad (even before the due date) and has to be poured out all at once? And what about all the commodities I keep on the shelf, and put in the fridge once I open them?

This so-called “smart fridge” is one of those nearly-generic, ubiquitous, almost brainless examples trotted out as a stand-in for for future tech, just as we see those stupid example apps show up on every new “extensible” piece of technology (phone, widgets framework, whatever) – the stocks, sports scores and weather apps. The apps that *no one* ever uses more than the first week of owning that tech (well, I’m sure there’s someone – like the dev – who must use them, but no one I know – and not like “I don’t know anyone who will admit to buying a Michael Jackson album while he was alive”).

Which reminds me of the foolish crapware that used to show up only on new PCs – but now ships with some Android phones and with all “smart TVs”. Ugh – I saw a report recently ( that most smart TV users just watch live, streamed or pre recorded content on their TVs, and almost none use the “smart” apps (generally less than 10% of smart TVs). In my experience they’re a resource of last resort – like when everything else has stopped working you’ll try them, but dog help you if you try willingly – hopes dashed, spirit mashed, ego crashed.

Which also reminds me of a great blog article by Scott Hanselman (My car ships with crapware about the terrible interface to the in-dash entertainment system in his new Prius. I’ve got the same one, and I fell victim to the same wow factor when considering the purchase. Once I actually tried to *use* the onboard apps, however, I quickly gave up – too slow, too many clicks, too many unintuitive choices, too few usages that weren’t much more efficient on my smartphone.

I happen to agree with Hanselman – not just about my in-car screen, but the in-TV “smarts” and the soon-to-be-everywhere “smart” appliances. I’d much prefer (at this stage in the “smarts” development) that these lesser apps be removed entirely in favour of just giving me a fully-integrated big screen on which to mirror my already-quite-handy pocket-sized computer. I understand the need for these industries to try to find ways to achieve bigger margins on the sales of these well-established markets. I just believe that these are poorly-executed, lesser-than bolt-ons that add nothing to the primary experience of the device to which they’re attached, and which will be in a few short years a supreme waste of space and an embarrassing relic. I fully expect that I’ll be unable to use *any* of the onboard capabilities of the Prius Entertainment system in three years’ time, and will have to add an aftermarket device or just sell the car to some rube.

I’d personally love to rip and replace the smart interface on my TV with something that was receiving active updates for more than six months from the manufacturer, and which provided me actually-helpful and complementary capabilities I can use right from my TV – and which aren’t just easier and more intuitive on my phone. How’s about a TV guide wired right into my TV? Or something that told me how much TV I’ve watched for the past month or year, and a breakdown of what kinds of shows I’ve watched? (Not that I’d find that info indispensable, but at least it would relate directly and more tightly with the device from which it derived.) How’s about a remote upload capability (push only, no pull – no need to freak out the privacy dudes) for all that data – and more, like power consumption and device health statistics, so I could do something useful and more permanent with that data?

And as for the fridge: how’s about a sensor that tells me how “empty” the fridge is, giving me a clue I should go shopping soon? This could be based on how much power it’s taking to cool the contents each day – or how much the fridge weighs (compared to an average of the last six max weight measures). Or what if the fridge could actually pinpoint where that foul smell is coming from – and better, could give you a warning when the crisper is getting more “moist” (i.e. more “rotty”) than it should be.

That would be a smart device I would actually appreciate.

A few notes on Don’t Make Me Think

Encountering Steve early in my transformation to a UX geek profoundly affected me. I’ve been re-reading his seminal book lately and captured a few thoughts I’ll share with you.

His book “Don’t Make Me Think” is one of the most shockingly plain and easy discussions of web usability I’ve seen, with lots of subtle lessons woven into a deceptively easy-to-read discussion. Easy to follow, both anecdotal and evidence-bound, and contains many dozens of insights distilled down to the very nut of the problem. No overlong self-pleasuring discussion of theory or why things should be different than they are – Krug recognizes first and foremost that the best systems give users the easiest path to success, and leaves the art of user experience design up to the definition of ‘success’ (to be struggled over by users, designers and stakeholders).

One of the great lessons of this book is typified by the following passage in chapter 8:

“The point is, it’s not productive to ask questions like ‘Do most people like pulldown menus?’ The right kind of question to ask is ‘Does this pulldown, with these items and this wording in this context on this page create a good experience for most people who are likely to use this site?’ And there’s really only one way to answer that kind of question: testing. You have to use the collective skill, experience, creativity and common sense of the team to build some version of the thing (even a crude version), then watch ordinary people carefully as they try to figure out what it is and how to use it. There’s no substitute for it.”

This lesson has echoed in my head for a year while I ruminated on it. I suppose I wanted to find a shortcut, see of there were other reasonably-equivalent ways to achieve the same outcome; you know how it goes, and you know how it turned out. Now I quote this idea at people all the time, and I keep hammering myself over the head with it every time I try to take a shortcut. Doesn’t mean I don’t take the shortcut sometimes, but I’m making sure I’m aware that I’ve optimised out the most effective way to determine the best design.

Another one that resonates every time I read it: “What testing can do is is provide you with invaluable input which, taken together with your experience, professional judgment and common sense, will make it easier for you to choose wisely – and with greater confidence – between ‘a’ and ‘b’.”. User testing gets presented by some practitioners (usually the ones fesh out of academia, or those who are terrified of the lack of rigor and thoroughness in the business world – or both) as a terribly important practice with lots of rules about statistical validity and ability to “prove hypotheses”. In my own experience even, it’s amazing to get that outside perspective – even if you throw some or much of the feedback and observations away.

I’m also permanent changed by Krug’s perspective and emphasis on navigation. Thinking of it like the navigation you need when walking into a department store is brilliant, invaluable advice. I think of this every time I design a web site (and in some cases even do a passable job of reaching a reasonable level of clarity).

Every UX/usability tome I pick up gets compared to this, and many predictably (and deservedly) fall short. It makes for much quicker reading sessions on the dry or artificially-academic doorstops.