?

Log in

No account? Create an account
Bad Design at Apple - Input Junkie
November 17th, 2015
10:04 am

[Link]

Previous Entry Share Next Entry
Bad Design at Apple
It fell apart after Steve Jobs.

I'd wondered whether people who care about good design (a combination of of beauty and usability) were that rare. The answer is yes. The link is the sorry story of a company which was built on good design, which had codified simple principles of good design.... and then forgot it all. Apple is making a partial effort to make its devices easy to use again, but it doesn't seem to have a solid grasp of the idea.

I'm seeing this as a failure of empathy, and not specifically a problem at Apple. It's apparently really hard to think about whether a product is easy to use, and to do the checking to find out whether real people are a good match for your idea of what will work.

This being said, how could anyone have thought it was a good idea to take away the undo button, and then substitute shaking the device to undo? You can't tell what sort of shaking is needed, and it isn't available for all applications. Aside from the unreliability of shaking, it seems to me that shaking would interrupt your train of thought.

I'm bewildered by the folks who believe you can get people to do what you want by applying the right incentives, both positive and negative. The truth is, people can be distracted by incentives so that they don't think as well. They may decide to resist your incentives. And people's capacities vary, so you, you all-dominant incentive supplier, might be trying to make them do something they can't do.

I think modern corporations and possibly some theories of economics are built on the premise that people respond in a simple and reliable way to incentives. This isn't just the belief in everyone being "rational", it's also a belief in people being interchangeable. It's wrong.

The Lottery of Fascinations-- aside from innate capacities, what people are interested in matters.

Link thanks to supergee.

This entry was posted at http://nancylebov.dreamwidth.org/1075032.html. Comments are welcome here or there. comment count unavailable comments so far on that entry.

(15 comments | Leave a comment)

Comments
 
[User Picture]
From:supergee
Date:November 17th, 2015 03:56 pm (UTC)
(Link)
Excellent! I shall add a link to this in my post.
[User Picture]
From:kalimac
Date:November 17th, 2015 04:00 pm (UTC)
(Link)
If you have not read it, I would recommend The Psychology of Everyday Things (aka The Design of Everyday Things by Donald Norman, for its discussion of how design influences, often in unexpected ways, how people use devices.

I am strongly of the opinion that Apple never had good design. Certainly the Mac did not. Designing a computer so that "delete file" and "eject disc" were the same command was off on the wrong track from the start.
[User Picture]
From:nancylebov
Date:November 17th, 2015 04:43 pm (UTC)
(Link)
Do you think Apple's design has gotten worse after Jobs' death?

Edited at 2015-11-17 04:44 pm (UTC)
[User Picture]
From:kalimac
Date:November 17th, 2015 08:30 pm (UTC)
(Link)
I have no idea. I've never bought an Apple product and I hope to go to my grave in that state. And it's been a while since I've had to use one of the fricking things. But my point was that it concerns me less if it's gone downhill if it was never uphill in the first place.
[User Picture]
From:dr_zrfq
Date:November 19th, 2015 02:59 am (UTC)
(Link)
Speaking as someone who knew people with original Apples (and got to play with same) back around the time of Jimmy Carter's presidential inauguration, those early Apples, up through the //e, were IMO the best designed micros of their day, and *both* Steves were responsible for much of that excellence. I was less impressed with the Apple III, it felt like they were coasting when they still needed to have the foot on the gas at least a *little*.

The Mac, in my opinion, was half-baked: the *basic* concepts were spot on, but the design fell apart in the details (wherein reside both God *and* the Devil) and then the implementation failed some sanity checks that had me scratching my head in great puzzlement. Not that anything else in the general consumer market at the time did any better, but the Mac team, having gotten so far ahead of most others right then, should have had the smarts to do, at least, a less sucky job.
[User Picture]
From:mneme
Date:November 17th, 2015 06:11 pm (UTC)
(Link)
I find it ironic/fascinating that you, as a self-identified liberetarian, are sceptical on the effectiveness of incentives as a mechanism for getting people to do what you want.

I do think that people over-estimate how easy it is to build up a solid and effective incentive structure, but also think that analysing by incentive (i.e., cost/benefit) is a solid model to looking at human motivation, even if many costs and benefits are individual and invisible. Game design is very relevant here--both in that you want the "game" to motivate and reward the behaviour you want--and in that you want people to be able to internalize the game rather than dealing with a thousand micro-transactions that in effect distract from the actual intended behaviour and increase the overall wasted costs.

So for instance, tips are effective in terms of getting more attention and service, and in getting better staff better compensation--in the short term. In the long term, of course, they normalize, so the staff don't get better income and the tipping custom don't get better service, and once a sizable population tips (or all tips at more or less the same level), the extra work and load involved in collecting and looking at tips--not to mention the ways tips create micro-economies in a service industry itself, when the service that is being tipped for includes suppliers who do not have direct access to tips, which again don't help anyone in the long term.

In many ways, I think this comes back to the "good UX" core question. Incentives are fundamentally part of a business's UX; the interface between them and their employees, and between them and the customers. There need to be good incentives or they can't hire or retain employees, and nobody will buy from them or they can't stay in business becasue they're not drawing in enough to service their underlying capital or pay their costs. And those incentives can be shaped somewhat to fit how they want to fit into the business and make sure they get paid for what they want to be doing. But it's still incredibly important to maintain a good UX between those two interfaces (three, really; suppliers, employees, customers), because otherwise you're driving a wedge between the price paid and that collected to no good purpose.

[User Picture]
From:nancylebov
Date:November 17th, 2015 06:17 pm (UTC)
(Link)
I'll probably do a longer answer later, but I think people have excessive trust in incentives, not that incentives don't matter at all.

Governments are bad at incentives, too.
[User Picture]
From:nancylebov
Date:November 18th, 2015 02:49 pm (UTC)
(Link)
"I find it ironic/fascinating that you, as a self-identified liberetarian, are sceptical on the effectiveness of incentives as a mechanism for getting people to do what you want."

Keep being fascinated. There are two angles of attack. One is that you don't understand libertarians as well as you think, and the other is that I'm a weird libertarian. Both are probably true.

Something which might be leftish or might just be common sense, is that I believe you can't control people very effectively. You can affect people with incentives, but that's not the same thing as getting them to do exactly what you want, and the problem is considerably harder if you want people to come up with something that will satisfy you than you don't have fully specified. I think part of what you're saying about game design is that you have to build on intrinsic motivation.

Punishment gets even more complex since the person you're trying to punish may decide to de-incentivize you by not complying.

Over at dreamwidth, siderea pointed out that it can be quite hard to figure out what rewards will motivate people to change what they do.

It's not a bad idea to look at incentives to explain what people do, but there may be a problem with non-falsifiability there.

In re tips: a lot of people just tip as much as they usually do, without fine-tuning it for quality of service, so that's another way tipping fails as an incentive system.

It might be interesting that even though incentives are theoretically similar for all businesses, business competence varies quite a bit.
[User Picture]
From:mneme
Date:November 18th, 2015 04:17 pm (UTC)
(Link)
I don't think it's either in this case--but it's still interesting that libertarian thought in general is very suspicious of artificial, deliberate incentive structures, while relying significantly on the market, which is, of course, an emergent incentive structure (among other things, but that's centrally why it works at all).

Punishment is tough as an incentive structure in the first place, because while rewards are, by and large, additive -- the person receiving the reward will help you give it to them and recognize reward-earning behavior, a punishment will generally involve having to detect undesired behavior, recognize it and identify it, and then apply the negative incentive to it. In effect, this means that there are two major sources of loss involved in punishments--first that you have to spend resources on whatever process detects punishable behavior, resources that could be spent on something else -- and second that unless the process is foolproof, you will sometimes fail to detect punishable behavior, and it will thus go unpunished. So the full force of your punishment isn't guarunteed to balance every infraction, and on top of that, you have to not only pay of the punishment but also for detection and enforcement.

It may still be necessary (will), but it's obviously much better to provide positive incentives where possible (or have "punishments" be emergent failures to reward rather than applied incentives).

[User Picture]
From:agrumer
Date:November 17th, 2015 07:44 pm (UTC)
(Link)
Tog’s been complaining about Apple losing its user interface clue since well before Jobs died.
[User Picture]
From:madfilkentist
Date:November 17th, 2015 08:26 pm (UTC)
(Link)
I think Apple is largely driven by the effectiveness of the demo more than usability. If something makes a good impression when shown, they think that means it's good. A lot of people buy on the same basis, so that strategy doesn't lead to bankruptcy.

A bedrock rule with the first Mac was "No cursor keys." Bad design decisions, resulting from Apple's being too enamored of its own ideas of elegance, have been around from its beginning.
[User Picture]
From:kalimac
Date:November 17th, 2015 08:28 pm (UTC)
(Link)
I made the same point as your second paragraph here in a similar discussion elsewhere, and my reward was pitiful looks because I didn't understand the Religion of Steve. (Of course they denied that was what they meant, but it was.)
[User Picture]
From:matgb
Date:November 18th, 2015 11:44 am (UTC)
(Link)
The tone and timing of the article implies it fell apart when Jobs went back. He was front and centre of the iPhone launch and it's iOS generally that's the biggest problem.
[User Picture]
From:nancylebov
Date:November 18th, 2015 02:51 pm (UTC)
(Link)
I presumably need a better/more detailed timeline of what happened with Apple's interfaces. I'll post again if I get one.
[User Picture]
From:dr_zrfq
Date:November 19th, 2015 02:49 am (UTC)
(Link)
My own opinion, based on long observation: I don't think that early iOS was as much of a problem as it has become since then. And in all honesty, I think Apple was in serious trouble before they asked Jobs to come back (via effectively a merger with NeXT) and their products improved dramatically in design afterwards. But Jobs didn't bat 1.000, and whatever plan he may have had for "the design guru succession" -- if he had a plan at all -- hasn't worked, and it was obvious once he got sick enough.
nancybuttons.com Powered by LiveJournal.com