Slashdot Mirror
Are More Choices Really Better?
A. Bosch writes to mention that Joel Spolsky of Fog Creek software has a commentary that examines the need for choices in software. From the article: "This highlights a style of software design shared by Microsoft and the open source movement, in both cases driven by a desire for consensus and for 'Making Everybody Happy,' but it's based on the misconceived notion that lots of choices make people happy, which we really need to rethink." With software steadily becoming more sophisticated, are more choices really necessarily better?
Next question?
Seriously though, yes, more choices are always better. However, the additional choices don't have to be easy to get to.
For example, practically everything in Windows is configurable. However, in most cases the configuration is not exposed via a GUI. It's set to some default and you need to tweak the registry.
The same is true of Unix, of course; you often need to go to the config file directly to accomplish something, even where a GUI is available. You can accomplish all kinds of wacky things editing Xresources files.
But in both of these cases the full complexity is not directly exposed, so the user doesn't have to deal with it. On one hand this makes the software more complex and typically leads to bloat. On the other hand, this lets one tool accomplish many tasks without bothering people who don't use the functionality with its presence.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
That said, let's take the average American. Their head would explode if you started explaining all they could do with Linux. They'd probably rather be trapped in the movie Deliverance than be faced with building and configuring Linux from scratch.
So don't ask me if more options are better because it depends on the case. I don't want my text editor to have all the bells and whistles known to man although I expect my process management suite that I use at my company to be able to interface with web services. Even though I prefer Emacs over MS Word, the next person my prefer them flipped.
To recap, ask your customer. Ask your end user. Ask your mother if she'd be able to user your software (provided it's meant for the general public). But the last people you should be asking are members of the Slashdot community.
My work here is dung.
That said, the KDE and GNOME guys can return to ranting at each other...
What I'm listening to now on Pandora...
...click...whirrr...whirrr..."I'm sure glad I don't have to solve all those hard problems like alphas and betas do..."
The problem - if any really exists - is not the number of choices, it is the manner in which the choices are presented to the user. ( For an example of good presentation, look at the average browser's bookmark function. You can have a well organized database of thousands of URLs, all of which are easy to find. Yet if they were one long list, it would be incomprehensible. )
The solution is not to obsess about the number of choices, but to think about the best way of presenting choices.
Seriously though, yes, more choices are always better
False. It has been shown in numerous studies that more choices often cause information overload, and result in poor choices being made. I will cite two examples:
1) Gov't Health Care - During the Clinton years, the idea of nationalized health care was bandied about. A majority of Americans agreed with the notion. How did the Republicans get it mired down and defeat. Besides Hillary leading the effort, the way it got shot down was brining three or four different models into the picture. Americans got overwhelmed, and opted for (f) None of the above.
2) 401(k) plans. Want to reduce your participation rates? Add more investment options. Sure, your sophisticated investors might like it, but Joe Sixpack gets eyes like saucers when he sees forty-five options that he must pick from. Study after study has shown more options = lower participation.
See my journal for slashdot ID's by year. Mine created in 2005. http://slashdot.org/journal/289875/slashdot-ids-by-year
I'm still waiting for the computer with one button: "Do What I Mean"
Everything else is an abject design failure.
So... people should be given a choice when it comes to the question of how much choice they should get.
My brain hurts now.
dragonhawk@iname.microsoft.com
I do not like Microsoft. Remove them from my email address.
Actually, more choice isn't always better. Sheena S. Iyengar is a professor at Columbia University who studies choice and in particular, challenges the notion that more choice is always better. A list of her publications is available on her site. For those who believe more choice is always better, I recommend you read a few. In fact, I recommend you start here (pdf).
Ironic, given that the screenshot he is showing has exactly two easily accessible options (lock and power down) and hides the rest away. Most users may or may not take a look at the other tucked away options in the drop-down/pop-up-box, and probably not worry about it again if they feel scared. So, for users who want less complexity there is already a very reduced choice of options. Is he then suggesting taking away options from power users? Really? B the same logic shouldn't notepad bet better than any IDE for doing programming, because it has less choice? And we probably want to do away with the command line for good, because there's clearly waaay to many options there. And the large majority of people already favoured the one-button mouse from apple very strongle, so much in fact, that apple never was asked for mice with more buttons, and most pc-users today buy and use one-button mouses.
Now, clearly Joel (and me here) have oversimplified the topic so much, that the dogma "less is more" has led to absurd suggestions. The key for successfully applying "less is more", is to properly look at the context. For a computer that is used as an internetkiosk, "log off" is the only button you need, there reducing choice is helpful. For a laptop user it would be extremely annoying not being able to choose sleep or hibernate, because it is going to waste energy and reduce the time I'll be able to work on it. Automatic powermanagement is not an option, because it can't read my mind. The computer will always be in hibernate when I just don't have the time to wait for it to power back on, or it will waste energy in sleep, when I know I'll be away.
I like to compare those options with my clothing options as a human. How would you like it, if somebody wanted to simplify things for you, and you only had two choices: naked (for sleeping), and fully dressed (for work). Want to take of the sweater because you have a shirt underneath? Tough luck, it was "optimized" away so you wouldn't have to worry about choices. Want to take off your shoes on the plane? Nope, either naked or fully dressed are your only options. Pretty silly - for most people, now of course there will be some people (those you are stressed out by clothing choices) that may feel a binary choice is progress, and good for them, yet that doesn't justify taking away the options from those who feel very comfortable partially taking some clothes off.
The funny thing is, that Joel even acknowledges tht there are good reasons why people who are comfortable with the choices, and why they are necessary for some, yet he somehow implicitly argues that those people are overridden by the ones that get scared by the options. He never explains why, though. Which IMHO makes his argument/position look very weak.
This is an unanswerable question, like "is more production really better?" Like every other rational question, it becomes a matter of marginal costs and benefits.
Additional options are always better until the marginal cost (in researching/comprehending the option) becomes greater than the marginal benefit provided by the option. Thus, options with low marginal benefits and/or high research costs are not better, and other options are. For example:
This is an example where more options are probably bad, because each additional option has huge research costs associated with them - that is, it takes a lot of effort to find out exactly why a person would prefer one or the other.Here, more options is probably still not better because while the research costs are low, the marginal benefit to being able to choose a slightly different shade of gray are so tiny as to be outweighed by the effort of having to even answer the question.Here is an example or more options are better. It's clear what the differences are, making research costs low, and the benefits to choosing the correct keyboard layout are huge.If you don't know where you are going, you will wind up somewhere else.
The question is, "Who here prefers a manual transmission car to an automatic?" I have been in probably a half dozen classes of programmers when he did this, and every time he gets about 50% of the audience to raise their hands. Privately he tells me that it's almost always 50%, give or take a couple of percentage points.
After he gets the count of hands and shows that it's about half of the audience, he points out that the public as a whole (at least in the U.S.) prefers automatics to standards by a margin of at least 9:1.
His point in doing this is to show that the kinds of interfaces that programmers like (lots of knobs for extra control) are not necessarily the kinds of interfaces that most people -- which is to say "the people who buy your software" -- want. The vast majority would prefer simplicity; in fact, they will pay extra for simplicity.
Building in a lot of options makes about one tenth of the audience happy, but annoys or confuses the heck out of the other ninety percent. It is not good software design; it makes for more difficult training and much more difficult technical support. If you feel you must do it, it's best to hide these knobs in an expert mode ... but by and large you're better off by not providing a lot of knobs in the first place. Spend your time carefully designing your software so that you make the right choices so that your users don't have to figure out how to fix what you did wrong.
jim frost
jimf@frostbytes.com
The whole point of software is to automate tedious things, so if choice becomes tedious we should really try to automate that as well. Joel's approach is entirely correct: figure out what your users are trying to do, and try to offer options which relate to them. Exporting all functionality and leaving the user to figure out what all the options mean exemplifies design failure. It is essentially "no design". Slightly better is to provide sensible defaults, but the best designs focus on things the user thinks a computer ought to do and things the user wants to do (disregarding, sometimes, what the computer really does do). The whole situation reminds me of command line interfaces, where the user had to memorize a bunch of commands and options and put them all together in order to accomplish a goal. We have somehow recreated that mess in graphical form, with the added inefficiencies of navigating menus using a mouse. It's almost like a cruel joke has been played on the public. Just when computers look like they might become simple enough for anyone to use and are widely adopted - BAM! Take that suckers! Whenever this topic comes up I think of my Grandma. She is a smart lady and she uses computers, but she is also old. She has been around for a while, and she's not about to tolerate any more nonsense. Contrary to the stereotype she is not intimidated by technology, but she avoids a lot of it because her impression is that most of this stuff is designed by idiots who don't have a clue what they are doing. I hate to say it, but she's right. I think that when we talk about software that grandmas can use, it should be in this sense. Old people are not going to put up with this crazy crap the way younger people seem to do. It has to make sense, and it has to work.
In Soviet America the banks rob you!