Tuesday, February 2. 2010Micro-Options Reloaded - The Paradox of Choice A while ago I had written an article about the dangers of "micro-options" in software. Like many articles about usability, this one was discussed somewhat controversially (which is good in a way), but overall I did get a positive reception, even from a professional usability expert. While this article was perhaps interesting, there was no "proof in the blog pudding". Let's be honest here, anyone can claim the wildest things in blogs, which may be true, or not. I did however stumble upon an interesting connection between psychology and usability, which appears to back up my speculations. The connection is this: The Paradox Of ChoiceHave you ever been in the following situation: You're in a supermarket. You want to buy a salad dressing, and the timing is somewhat tight. Guests are waiting, the usual... Alright, found the "salad dressing" section (which can he hard in itself, in huge supermarkets). So, not much time, let's choose something that looks yummy, and preferably not too fatty! So, what do we have here? Vinaigrette, Blue cheese dressing, Caesar dressing, Honey Dijon, Hummus, Italian dressing, Louis dressing, Ranch dressing, Russian dressing, Tahini. Hmm. Nice choice. Let's go for some Italian Dressing, I like that one. Wait, there are three different brands of dressing, each of them offering further varieties of this product. Chef's Choice, Low Carb, Lite, Very Lite, and hey: the Special Edition! Here is someone who can explain this problem of choice much better than I could ever do. His name is Barry Schwartz, he's a professor of psychology, and he made one of the best TED talks ever (according to the founder of TED himself): TED Video: Barry Schwartz - Paradox of Choice I very much recommend watching this video. Few have been disappointed by it, and I suppose you might find it enlightening too. It does explain many phenomena in our modern life in a surprisingly simple way. One of them can be translated to usability: Too many choices can reduce the user's satisfaction with a software product. I'd be happy to read comments on this article, be it positive, or negative. Discussion is important, so I think we should do that. But please, do me one favor: Watch Barry's talk before commenting. Thanks  Trackbacks
Trackback specific URI for this entry
No Trackbacks
Comments
Display comments as
(Linear | Threaded)
Yes, the video is great.
Another one with the same guy I like even a bit more: http://www.ted.com/talks/dan_gilbert_researches_happiness.html
Yes indeed, Daniel Gilbert's talk (and his book) is highly interesting too.
It's not directly related to this topic, but in certain ways it might be. I was very impressed by his book "Stumbling on Happiness" (and it did make me a happier person, indeed).
Difference between choices in software and the supermarket: With software, you get some defaults. Kinda like the area in the shelf saying 'Buy this! Huge rebate and so on.' If you are new to the software or have no time, you just use them. However, if you have special needs, options allow you to adjust the software to your liking. Kinda like going to the grocery store, and finding all products in big packages, but you have a tiny bag (=screen). Then you ask a salesperson, 'do you have that in smaller versions?', and they grab under the counter and give you a small version.
With micro options, you can set reasonable defaults for most people (=result of the usability studies), but still allow people with special needs to adjust things. If they don't look for the options, they won't bother them. You want vertical tabs even though usability claims them to be bad, but you have a small screen? No problem. You don't like the context area in Amarok? No problem. Just about every flamewar/bko-wish-fight that I heard about in KDE recently could in my opinion have been solved with a micro option in about 5 minutes.
Unfortunately, "good defaults" are not a magic bullet.
You are correct that we should absolutely try hard to choose good defaults. This is crucial. However, even if you have these good defaults, and then you open the settings dialog, you will start to wonder if there might be better settings than the defaults you were given. So you might try to ignore that, but it keeps nagging and nagging. "Did I really make the right choices here? What if I made a bad choice?"
You try it and see if you like it. If you do, you keep it, if not, you change back. It is not like a store where you are forced to stick with your purchase. Even if you could exchange something at a store, it requires driving talking (possibly arguing) with a sales rep, correcting your credit card, and driving back. The cost of switching is very high. The cost of switching in a software program is negligible, perhaps seconds of lost time.
If there are two options where the results are almost identical, then I could see there being some trouble with indecisiveness. I agree in those situations there shouldn't be an option. But if there is a clear usability difference then, by definition, the two options are substantially different and that indecisiveness shouldn't be a problem. So for the vertical taskbar icons case, for instance, either you need the space, or you don't. If you don't need the space, then you choose the option that lets you actually read your tasks. If you do need the space, then you pick the option that lets you see the rest of your stuff. Sure, I guess there might be a small number of people with old devices with intermediate-size 4:3 screens, but I would suspect that number is much smaller then the number of people with small-sized screens who desperately need the space (please grant, just for the sake of the example, that this is the case). So you would be making life really hard for people with small-sized screens to save perhaps a little bit of confusion for a smaller group with medium-sized screens. That doesn't seem like a good solution to me. I think, rather than saying "options confuse people", you should ask "is it more likely that people will clearly want the alternate option, or more likely that people will be in some ambiguous situation where it isn't clear which is better?" Then you need to factor in how big of an inconvenience it is for people who want to be able to change it but can't. I think you can only say having the option is bad if the number of people for who the two options are basically the same is so much larger that it overrides the problems the lack of the option has for people who need it.
I think the supermarket analogy has to be seen a bit differently about software.
Suppose that you have an application that forces you to provide configuration choices in a dialog that only appears when the application first starts. I'm sure that will leave users in doubt of the right choice. That's one reason to have (good) defaults because most users don't even care about choice. The situation becomes slightly different when a user has a certain itch with some very specific setting. Then, the user knows exactly what choice he wants. Giving him this choice will certainly make him more happy since he doesn't need to change habits.
"Did I really make the right choices here? What if I made a bad choice?"
For me, in most cases, the answer is: I know I made the right choice for ME. I look at the default options, see where they disturb me, and look for a way to change it - either because I want to restore the behavior to the behavior of a previous version of a program or because the behavior is not intuitive for me. Of course, if you take Tab Mix Plus for Firefox as an example, the options for tab handling just blow you away if you were to try them all. But if you use Firefox, and say 'this I would do differently, let's see whether there is an option for it', it works. It is good that there are people working on usability and that developers try to create the 'perfect UI' that fits all. As long as I can tweak it here and there. Everybody has different ways to work, different screens, different memories, etc.
The problem is not choice, but habit. If you are in a supermarket, where only one kind of dressing is offered, you would take that. If a few years later, the same market offers hundred types of dressing, you will certainly search for the one you are accustomed to.
The same is with software. If you get a new version of a program, you usually do not care how many "options" it has, as long as you will be able to work with it the same way you did with the old version. Now imagine what happens if the supermarket would start offering hundred types of dressing, but NOT the one you always had. You are forced to adapt, to make a decision, to try something new. People do not like that. Compare with software, when the new version gets hundreds of new features/options, but does not allow the old way. This requires you to re-learn, start from scratch, try something new. When your user base grows (and you want that to grow), you will have to offer choice. Taking the supermarket as an example again, if you want customer B to also buy at your market, you have to offer the type of dressing that customer B bought at his previous store. Look at browsers. How often do you read something like "Why doesn't have browser A have this or that function that browser B had? I am used to it!" The only browser that will survive in the long end is the one who offers all options and all features of all browsers that had been available before. People never forgive regressions.
I would love to say I was successfully able to watch the video on my HP 2133 netbook running Kubuntu 9.10, but sadly the sound doesn't work. Without headphones there was total silence, and using headphones something like white noise. I tried to change the settings to KMix which has 6 channel sliders each with a 'split channel' option to allow you to set the right/left volume of each individual channel. Additionally, I couldn't work out how to use the exotically named HDA/VIA/VT82xx menu options - maybe they would have helped.
Maybe it is a problem with Flash, maybe the hardware on my netbook isn't quite right, I don't know. I don't have a problem with understanding bit rates, sound encodings, the difference between PCM and DSD and so on because I've been and audiophile for the past 40 years, and a computer programmer for more than 30 year. But can I get something as simple working like watching a video of an interesting talk? It seems not. People on Planet KDE have been a bit rude about the Apple iPad and how it doesn't allow software freedom, and is locked down. It probably doesn't have a mixer with 6 splittable channels. What if I actually just want to use my computer like a TV and have everything just work? If it is hackable too that is a bonus. I think Apple and Steve Jobs have the best understanding of the Paradox of Choice, and instead of being rude about the iPad I wish we could concentrate on making that way of thinking a basic part of the KDE community's culture. Because historically KDE has trying to be a 'better Windows than Windows', and Microsoft are unfortunately one of the least competent at understanding aesthetics, design and minimalism (the anti-Apple), I think we need to make some tough choices about how we can change.
Yep, I think that you have been raising two separate - and yet connected - issues here:
1) Multimedia on Linux is a mess. If you cannot get this freaking video to play (FYI: Richard Dale is a well known developer in KDE circles), then: Who can? I guess we've all been through this. And this does indeed touch on usability issues too, e.g. the "30 sliders and counting" issue in KMix. 2) While I dislike Apple products in many ways, and additionally I have also been stating that we need to find a balance with options (certainly not by removing all options!), KDE is going overboard with them in many cases. This is exactly what I'm trying to convey with these articles.
I see a significant difference between choice in life and using application: the reversibility.
If I'm not happy for a pair of jeans I have chosen, I could go back to the store and pheraps change it, but this waste time or money. When I'm using an application the cost of changing an option is really low and changing back to default is even cheaper. Also you're not forced to make a choice before starting using the application because defaults are the recommended choice. The more options != the better a program is still true but mainly because options increase very much the number of untested cases and so also of bugs. They make more diffioult finding a particular option.
I saw the video, and I though it was striking how all his examples were misguided.
for example, he complains about the excessive number of salad dressings. And it is true, one finds a really excessive amount of such things. Bu that it causes distress is purely a problem of education: in fact, none of the choices are any good! Because with if you cook, you have balsamic vinegar, olive oil, mustard, an all the spices necessary on hand. An even if you don't cook, making a salad dressing is a 20 s affair. The satisfaction comes from the production of the exact dressing you want. Not everything can be done by hand, but, but if you pick things for maximum flexibility, you will be able to o the exact optimal thing you need, when you need it. For software, this means that for small stuff, you just need to know a bit of scripting and have a text editor -- any one will do under linux. Not anyone can code, of course. But if you care about your tools, you can help in the development of your softs, submitting bug reports/ideas. In general, decisions are hard when you don't know enough. For example, his jeans story would not have happened if he knew about tailoring: he would have specified his ideal jeans, and if it had not been available, compromised or moved to the next store. So the solution to the problem is not to restrict choice, it is to help people be educated so they can perform informed decisions. There is no reason those of us who know should be hindered by those who don't. If people don't know, they should be educated. If they don't have time, for that, it is a societal problem which should be solved. If they don't want to know,it is a societal problem which must be solved. ignorance should never be accepted as a valid point of view
In a perfect world everybody knows everything,in the real world every person knows only a little. Also not everybody can understand every subject. Everybody learns at different rates witch can be really low.
I'd say that ignorance is necessary: when my car is broken I can't learn how to repair it, when I'm sick I can't learn medicine. If somebody can do something really well, I let him do it and then I use it. Should I recompile my kernel? It works fine now, why should I try? It has compiling options but if I just want to use my pc to write letters maybe I can let my distro chose. In software that's abstraction: I use a function, I know what it does, but I ignore the implementation.
That all is fine, not everyone knows everything. But in your example, the car, the options are all still in there! It just has sane defaults and anyone with knowledge can thinker with it.
Your example does not solve the problem of choice at all. It just shifts it.
Instead of having to choose from a huge amount of dressings, one now has to choose from an a smaller number of (olive) oils, mustards, vinegars, etc., making for a bigger number of possible combinations than the ready-made dressings added up to! I actually do agree that we need less choice, especially in some area's that you "just want to work". Take the health care system in the Netherlands. Competition and choice for the patients is being regarded as a holy grail. The number of insurance options has exploded. I have to try to find information on different treatments in different hospitals, see if my insurance company covers those (and where), and then make choices. Problem is: I do not WANT to make that choice! I want to be confident that if I have a medical problem, I will just be helped in the best way possible within a reasonable time. If possible, I'd like that to happen in a hospital not too far from where I live. I don't need choice there, I need to be taken care of professionally. The same goes for utilities. We now have to choose our suppliers for gas, electricity, phone, TV signal, etc. Why do I have to bother studying all the complicated plans? My quality of life would be better served with spending that time taking a stroll along the beach with my wife! Sure, average prices have dropped a little bit, but quality of service even more so. What did it gain us in the end?
This guy is an extremist... I actually like going to a supermarket and choosing salad dressings and coffee jars. It relaxes me, pulls me away from what I was doing for the whole day.
I also like options in the software I use... as long as they're deep hidden from the working panel and the defaults are well thought. Seems like this guy got a self-inflicted extreme simplicity syndrome. I say he's got a bigger problem than saving word from choice, he first need to catch up with evolution in order to chose next pair of jeans.
Talking about the paradox of choice in software configuration we mustn't forget that configuration is not *choice*! Often we are forced to choice and there is no other really pleasant option (buying no jeans is unpleasant as I like to have warm, vested legs).
But for doing something like changing an option you need enough motivation. This motivation may be that something doesn't work or something works in a annoying way, annoying enough to get someone to change the configuration. Configuration usually isn't something you like to do, in contrary, most people avoid configuring more or less (you too, even if you may do it more willingly). So I think good presets are very important. With good presets people aren't urged to care about configuration, even if it may result in a better configuration than they would have had otherwise. This is good as people still don't like bothering with options which means thinking, having decisions and making choices which could be wrong. I know people who are even afraid of changing some settings because they don't want to break anything. For them, but for us as well it would be nicer if everything works "out of the box".
Choices are fun. What is wrong with not knowing which to choose? It does not matter, even if there are only two options! Of course, anything in excess is just wrong.
I fear the do-not-think trend nowadays. Developers want computers to think for us, the goverment wants to think for us, the media wants to think for us. But I actually like learning things everywhere, everyday.
i agree with the above post.
I would say if you feel bad, uneasy, angry or whatever because there is 123 dressing choices and not one as you would have hoped, the issue is yourself. You can do talks, discuss, share you thought and find other people who feel the same, it wont change a thing. If you need someone to say 'that dressing is the one' for yourself, go get back to school you missed something. If you feel safer when everybody does the same as you, you scares me. Get up!
micro-options are bad because they usually make the ui ugly, but this can be avoided.
If every option most users would want was on the main configuration dialog and ordered and everything else would be hidden in an advanced button everyone would be happy. It is bad to tell user "the way you do things is wrong!, do it my way which has been tested by ui professionals." What do I care I want MY best way of doing things. If a user doesn't want to mess with confs he can ignore them because they would be in an advanced button going into the example of the dressing: It would be like, entering the supermarket and getting a recommended set of dressings,(basic conf) with one very recommended(cheap, light , and yummy)(default) also in a high shelf all other variations of the recommended dressing and some others weird dressings for people who would like to try something new or for people who came from other country and liked a dressing which was provided there.(advanced) If the ui is well designed they could all live in harmony.
The supermarket analogy doesn't work for me. In my case if I walk into the supermarket and there are 400 salad dressings to choose from I will simply choose one that seems interesting and go with that. If it turns out to be bad well it's just a salad dressing. I will simply choose differently next time. If I like it then I may purchase it again or I may choose a different flavour that might be interesting.
Choice of this nature in my mind only becomes a problem if you are focused on making the best possible choice all the time. That is focusing on the destination rather than the journey.
I'm afraid that being "focused on making the best possible choice all the time" is perfectly normal human behavior. If we didn't care at all about making good choices, where would this get us?
You could start breakfast with 3 burgers at McDonalds, then maybe buy a completely overpriced (and ugly) car, have sex with a random partner, and then drink a bottle of really bad Whiskey, combined with a bunch of Paracetamol pills. Then you'd be dead. Is that really typical of humans? This reminds me more of "Darwin Award" winners.
Whem I was a student I would just pick the cheapest salad dressing. Then I got a job and so I could pick a more a expensive dressing. Now I have a baby so I am back to the cheapest one
Heh, I can think of a few problems with Barry Schwartz's doctor analogy. If the options are losing an arm, or losing a leg, I sure as hell don't want the doctor making that decision without consulting me.
One tricky problem is making options discoverable, but without confusing the user. Some HI designers favor removing options entirely unless there's overwhelming need for the user to be able to configure it, which I think is going a little too far. The other example is having an "Advanced" tab with a million different things to configure, which is also confusing. The case I'm thinking of is where there's something in the UI that annoys the user, but the user is disinclined to dig through the preferences to tweak the behavior (and there may not even be an option, in most cases). They'll just live with the annoyance, which probably costs them a little bit of time each day, but at least they won't have to think about it. A good example is that stupid "the site you're visiting is unencrypted" warning that browsers show when you leave a secure site. Fortunately, you can check the box that says "don't warn me about this anymore" right then and there without digging through the preferences. That gives the user the opportunity to change the behavior of the program, without having to know that there's a tiny little box in the preferences somewhere that controls it. Of course, I know a number of people who just always click through the warning, because they'd rather click through the dialog every single time rather than read the text of it once (which also pretty much defeats the point of the security warning). Unfortunately, a lot of user interface elements don't really have a good way to spell out "I'm configurable!" -- although having a consistent right-click for options helps (I like how I can right click on a window title and choose "Configure window behavior" and jump to the window preferences, even if I rarely do it). Meanwhile, systems that try to auto-configure by watching the user's actions are frequently plagued by their own set of problems.
Just wanted to add that the browser warning is also an example of bad defaults. The security folks might think otherwise, but the number of people who actually want that warning is minuscule.
There's probably much better examples out there for things that are reasonable defaults, but annoying enough for some users that they cross the line between "nice if it behaved the way I want" and "detrimental to user experience". That's the case where telling the user "yes, there's a better way" might be worth the extra decision-making involved, if done politely enough. Not polite: tip of the day dialogs, which don't offer relevant tips 90% of the time. There ought to be a good way to offer (bring to the user's attention) useful configuration choices, without being annoying or confusing, but it sounds like a hard problem.
There's already a desktop (and a set of applications) for people who hate options, it's called GNOME. Why should we target the same userbase, leaving behind the many people (like me) who actually LIKE options? People are using KDE for a reason.
It seems like the philosophy of the new KDE is "The new KDE 4, for people dumber than people who use GNOME"
It seems that KDE's 4 says "KDE 4 is so great, it's better than ever cause we've taken all choices away from the users and prevented them from configuring anything cause we think our users are too stupid and they don't deserve to have the options" Ok don't take that too seriously. There seems to be a new trend in the EU and the US to take away choice from people. What if it was decreed and enforced by armed troops that people will be forced to now eat hamburgers and french fries for dinner every night because giving people a choice makes them unhappy? You are very wrong in thinking you have to take away choices from users. Your job as a KDE and Amarok developer is: 1. Remove ALL obstacles that prevent users from configuring everything the way THEY want. 2. You as a developer may "present" a simplified view that for users who don't want to spend the time with all the options. KDE 4's biggest mistake is not having 2 or 3 presentations levels for different users like beginner, intermediate, and advanced views of options. If you think taking choices away from users is what you should do and good then why does Linux have the "options" and give users source code so they can change things? |
Amarok LinksCalendar
QuicksearchCategoriesSyndicate This BlogBlog AdministrationWe Listen To... Powered by
|
|||||||||||||||||||||||||||||||||||||||||||||||||
