The slippery slope | Hacker News Reader

The slippery slope

90percentofeverything.com

520 pointsjustinph135 comments

I'm going to nitpick a bit here, mostly because it's entertaining to play devil's advocate on a topic that's always met with such one-sided responses on HN (in this case: dark patterns).

My question is: Why should a company make it easy for people to do exactly what it doesn't want them to do?

Sure, outright trickery and lies are almost unarguably immoral. And the author definitely identifies many examples of that. But there are other instances where I just can't agree with classification of the UX as a dark pattern.

Let's take The Ladders, for example. In order to cancel, you have to fill out a form with a few required fields. What a shady, underhanded, and downright spiteful trick! Or is it? Most cancellation forms online will take you through an intermediary confirmation step, anyway. What's so evil about requiring a small form to be filled out? The questions are directly relevant to the act of canceling, and if you truly don't give a shit, you can fill it in with gibberish. Even if you give honest answers, you could be done with this form in literally 5 seconds. It almost certainly took you longer to fire up your browser and navigate to their site, anyway.

My point is: The company is not attempting to trick you or obfuscate the truth. They know that the presence of the form will likely lower their cancellation rates by a marginal amount, yes, but there's nothing immoral about lowering cancellation rates. And the form isn't nearly long enough to be considered difficult or unreasonable, so it's not like they're attempting to frustrate people into staying. In the end, it convinces some people to change their minds, but nobody gets hurt.

Compare that to, say, The Ladder's decision to auto-renew monthly subscriptions while hiding that fact at the bottom of the page in small gray font. Or to email subscriptions with no unsubscribe link. Or to burying opt-out checkboxes within gigantic forms. In these cases, companies are outright tricking people.

To give an example: In line with Google's mission statement of "Don't be evil", I figure that it could extend to "Don't be annoying".

> My question is: Why should a company make it easy for people to do exactly what it doesn't want them to do?

Because its just plain annoying to users. Running a business and designing your application / system to trick people into behaving in a certain way is one thing, being transparent is another.

I think users these days are becoming a lot more educated and wiser to these techniques, as a result their loyalty is beginning to side with companies that offer a more honest level of transparency.

Would you like to send that extension to the google integration team? This is the 6th time they've asked me to merge my account with google plus.

I'm starting to think the only way to get them to stop is with violence. I wonder if those nag screens come up in a standardized way...

Acting shady hurts your brand.

Harry Brignull(the author) discusses this in a talk[1] that I found on the dark patterns website(I checked out the website after loving the article).

He makes the argument that sure, many of these tactics might increase your conversion rates or make you some money, at least in the short run. But it also causes you to come off as seedy, to come off as a back alley discount store instead of as a premium brand.

He points out that if you're doing this and you're aware of that and you're ok with it, that's one thing. But plenty of sites might not have realized what they're doing and how it comes off.

I know I personally resent it when sites leave boxes like "sign up for our newsletter" checked or when they have fake download box adds. Making it hard for me to do what I want to do, no matter how you do it, bothers me. It feels like you're the adversary of the site, not a customer or friend.

I believe he talks about it around 12 minutes in but the entire presentation is easily worth watching. He has another one aimed at designers that I plan on watching as soon as I find my headphones(the sound is too low for me to hear it on my MBP).

http://www.90percentofeverything.com/2010/12/17/dark-pattern...

.... actual premium services are usually under contract. Even when they're not they still want you to call into a real live human being to cancel because finding out about your concerns that lead you to cancel a $10,000+ contract is certainly worth the minor possibility that you'll be offended that cancellation is not a push button process.

...finding out about your concerns...

That's the benefit to the service provider. What's the benefit to the service consumer?

The point at which a user cancels a service is the last chance a service provider has to end the relationship on a positive note. The provider might earn a little more goodwill, and the consumer has a frustration-free experience.

Match.com wanted a whole herd of information when I cancelled. The information I typed in was complete fabrication. As such tactics gain popularity I suspect more people will balk at it, making it a net loss to the service provider.

Being nice to your customers is called win-win.

I was countering the notion that premium services don't do this.

Also, Match.com isn't a premium service. Real personalized match-making and concierge service (they'll take care of everything involving the first few dates) is a premium service and the pricing starts at $10,000 and only goes up from there. If you want to join up with a premium matchmaker, you need to have a phone call and at least an hour consult. If you want to quit, you need to call in---there's no automation at all.

For another example....

When you hire a corporate lawyer, it takes a phone call to engage their services, or cancel their services. Their goal isn't to make the threshold to complete an action easy, their goal is to be the best representation your money can buy.

If you go to LegalZoom.com for your incorporation services, it ought to be expected that you can cancel with a click of a button.

I understand what you're saying, but disagree (up-voted your clarification anyway).

If I'm paying $10k for a service, I'd especially expect a one-click opt-out or cancellation (assuming that's possible contractually). So the point here is that more I pay the more I expect to interact with a provider on my terms. Think about it. Rich people don't have time to fuck about. If they've an assistant who does this for them then that assistant shouldn't have to divulge personal reasons on his or her employer's behalf.

> If they've an assistant who does this for them then that assistant shouldn't have to divulge personal reasons on his or her employer's behalf.

Well in my experience, you're not forced to divulge anything. You're just asked a couple of questions, and they do try to change their service (within reason) so that you won't leave. But if your heart is set on leaving, and you don't really want to interact... you can just say so, and they'll let you go albeit with a sad sound to their voice.

> Acting shady hurts your brand.

These companies are not silly, they understand the damage to their brand can be quantified as $x and that the benefit to their bottom line can be quantified as $xx .

Sometimes you do things you know have a negative, because you believe that the positives outweigh it.

Not that I agree with these dark patterns, but I understand why companies choose to implement them.

These companies are not silly, they understand the damage to their brand can be quantified as $x and that the benefit to their bottom line can be quantified as $xx.

Actually - they often are silly. It might be $x right now, but it turns into $10x later on as the consequences of that initial damage roll on and affect future interactions, recommendations, etc.

It's very easy to quantify the short-term gains being sneaky get you. It's much harder to quantify the longer term consequences.

By doing this they survived first and took place of some other honest companies. Still very large benefits.

Do you have examples of that? Because he has several examples of the opposite in his talk.

If you scroll right to the end, you'll find a great example of a site who thought they were being smart by being shady, and instead annoyed their users enough that people just went elsewhere (experts exchange), even though the measures were easily worked around.

If, as you acknowledge, these are user-hostile patterns, using them is going to lose the trust of your users, and in the end you will flounder, even if in the short term they provide economic benefits. That's the premise of the article, and yes the companies are silly for employing these dark patterns because they're trading long-term trust in their brand (and thus long-term profits) for short-term economic advantage.

These companies are not silly

What evidence do you have for that? Even assuming perfect knowledge, humans are what run these companies, and we all know how irrational those can be. Sure, sure, "the market" will kill them off, sooner or later (maybe; that's also assuming perfect knowledge of consumers, and we know how true that is), but until then, this shadiness should be called out and shunned whenever possible.

>Acting shady hurts your brand.

The Ladders has a long history of actually being shady, viz.

"Liars at the Ladders" http://www.asktheheadhunter.com/newsletter/oe20090120.htm

He explained this at the end. Companies that do this succeed in the short term, but in the long run they fail.

yes, all bad practices... I think of this like when you buy something at the store at the advertised price and it comes up at a different price at the register (almost always higher)... this is wrong.

Interfaces communicate with their users (at least, if you ask me), and I sometimes find it helpful to explicitly put into words what I think an interface is saying. I interpret The Ladder's cancellation screen as saying, "You cannot cancel your membership without giving us a reason."

Imagine if you were trying to break up with a significant other, and they told you that you were not allowed to end the relationship unless you gave a reason first. That would be horribly controlling, and totally unacceptable.

Of course, a romantic relationship is rather more significant than a subscription, but I think The Ladder's exit process is controlling in the same way, and the only difference is that it occurs in a less important relationship. It still gives me the same feeling that my wishes and agency are being subjugated, and I believe that's a bad thing to do to a person, regardless of the particular situation.

Again, its a slippery slope. Making the user hover over for some more time over his/her decision is not very far from a dark pattern. In case you need a feedback the best you should be going for are optional fields.

We're seeing 'one-click unsubscribe' link in many emails these days. I think that is one good indication for a spam filter to let through. This is a proof that only the right approach prevail eventually.

I think it isn't moral to lower cancellation rates by making the cancel action difficult.

The best way to grow is to have users that love you. I believe the inverse is also true, having users that hate you will limit your growth.

I once had a terrifically hard time cancelling a gym membership that turned from a free trial into a 2 year, $500 membership automatically. I still express strong feelings any time I hear the gym's name.

> My question is: Why should a company make it easy for people to do exactly what it doesn't want them to do?

The problem isn't when a company makes it hard for people to do exactly what it doesn't want them to do. Apple probably doesn't want its users to electrocute themselves, so they make sure the charger is properly isolated and all that.

The problem is when a company makes it hard for people to do what they want to do. That sucks, because it's usually why people are paying money for.

In this case, some people will want to make sure they aren't tracked (the other people don't care about it, so the option's placement is irrelevant to them). Hiding that option so well is simply useless: it achieves nothing for the people who don't care about it, and annoys those who do care about it (who tend to be tech-savvy enough that they'll find it no matter how well it's hidden). It's simply a defect in design: a misfeature that was grafted there for the glorification of someone who needed a contract signed. It provides zero advantages for the contracting parties, and huge disadvantages for the users.

The distinction between the two groups is not that clear. I suspect many users don't care too much, but if they stumble upon a "don't track me" option, they'll select it.

> Why should a company make it easy for people to do exactly what it doesn't want them to do?

I'm reminded of Star Wars, light side versus dark side. Neither side is stronger, and which side is "better" really depends on your POV. We do know that the dark side is "easier". Companies similarly can choose to go light side or dark side and end up with the same bottom line, but light side takes more up-front effort.

Now, let me apply the same question: if the dark side is easier and is as likely to get you results, why isn't it a perfectly valid choice?

>What's so evil about requiring a small form to be filled out?

What's so evil about requiring a 20 minute form to be filled out in order to cancel?

>The questions are directly relevant to the act of canceling

For The Ladders, and not for the user. The user is done with questions and just wants to cancel. The website is demanding that the consumer do free labor for it in return for the privilege of canceling. It's not tricking you, it's simply extorting you. Combined with...

>[...]say, The Ladder's decision to auto-renew monthly subscriptions while hiding that fact at the bottom of the page in small gray font[...]

...it's a direct statement that if you don't participate in our consumer research program, we will withdraw $25 from your account every month until you do. I don't even think that's legal.

Easy solution: send a certified letter with your cancellation notice; any charges after that can be legally reported as CC fraud. Enough people do that, and maybe they won't be pulling that bullshit anymore.

If you can sign up on the web, you should be able to cancel on the web.

Factoring in the cost of my time, sending a certified letter costs me $200.

Eight months of billing at $25 a pop, and they won't stop until you fill out their stupid form; screwing them over with a CC fraud charge would be worth $200 to me, if they're going to be that shady.

A much better pattern for the cancelation that I've seen is to cancel your account and then present you with the feedback form, they will get a much higher rate of usefull feedback this way.

Attempting to change people's minds by frustrating them rather than by actually presenting them with reasons to stay is, in a word, evil.

I'd have no problems with a page that says, "We're sorry to see you go, won't you reconsider? Here are some reasons why we think you should stay." I have a big problem with a page that says, "We're sorry to see you go, so we're going to make your life a little bit harder in the hopes that you change your mind."

Whenever two entities transact, there are four basic outcomes. The best outcome is when both parties benefit. That's what we should aim for in general. Another good outcome is when one party willingly takes a hit to benefit the other party, which we could call charity. Then there are evil outcomes where one party harms another party for their own benefit, and finally stupid outcomes where both parties come out worse off.

Trying to convince me not to cancel by wasting my time is clearly in the "evil" category. There's no attempt at mutual benefit, they're just trying to increase my cost in order to increase their benefit. An honest company tries to increase my benefit to also increase their benefit.

> I'd have no problems with a page that says, "We're sorry to see you go, won't you reconsider? Here are some reasons why we think you should stay." I have a big problem with a page that says, "We're sorry to see you go, so we're going to make your life a little bit harder in the hopes that you change your mind.

See also the evil AOL dialer dialogue that used weird negatives to confuse people into not disconnecting. That was so evil I think they last some law cases over it.

It's disappointing that so much effort goes into evil tweaks. But it's worse that customers reward that evil behaviour. Candy Crush is huge at the moment. I'd pay £5 for something like Candy Crush without the evil; heck, I'd buy it for all my family too to avoid the constant FB spam from it.

The freemium app craze drives me nuts. Used to be, the way to get more revenue from your game was to make it better. Now game creators are getting more revenue from their games by making them worse. That's a great example of win-win versus win-lose transactions.

Note that I have nothing against game creators charging for their products, it's just that the freemium model is pretty much inherently built such that creators make their money by making the game less fun to play.

I addressed this pretty explicitly in my post.

I agree with everything you're saying (intentionally frustrating users is bad), but I don't agree that a simple, straightforward form that can be filled out in under 10 seconds is an attempt to frustrate anyone. I don't think it's reasonable to suggest that a significant number of people, upon seeing this form, decided not to cancel because filling it out was too much a waste of time. It's just not a plausible argument.

Another commenter mentioned going through hell attempting to cancel a gym membership. I've had the same thing happen to me: took multiple phone calls over the span of a few hours to cancel my Bally's membership. That was a deliberate attempt to frustrate. This form is several orders of magnitude better.

I don't think it's reasonable to suggest that a significant number of people, upon seeing this form, decided not to cancel because filling it out was too much a waste of time. It's just not a plausible argument.

This isn't really something that is subject to argument; it's an empirical question. It's measurable; a known quantity to people who have implemented it in an A/B fashion.

I'd be very surprised if additional friction to a cancel process had zero impact. That would be sufficiently surprising to be worthy of an article or paper.

Not true. A simple A/B testing between having a form and not having one will tell you the "what" but not the "why". If I decide to cancel my account and end up on that form page, then change my mind and navigate away, you have no idea why. You cannot accurately claim that I did so because I was frustrated or because I thought it was a waste of time.

If filling out a form is not a big deal, then why isn't it optional?

I'd have no problem with the form if you could just skip it and click Cancel at the bottom. But the moment you force users to fill out anything is the moment you turn evil.

I agree that it's not anywhere near the same magnitude as it could be. But that doesn't make it OK.

To disable Ad Tracking under IOS, go to Settings > General > About > Advertising > Limit Ad Tracking to On

This is a worthwhile read (with excellent examples) on the ubiquitous "Click here to not receive third-party messages" pattern.

This fall onwards it'll be: Settings > Privacy > Advertising.

Source?

I can confirm this. The language is "limit ad tracking" instead of "disable," which worries me a bit.

I think a better word would be "refuse", as it more accurately portrays the situation. It too is not perfect though, as your "refusal" can still be ignored.

Ugh, I hate these. Instead of disabling it at the OS level, where they should do it; they create a "let's give unethical companies another bit of information" option.

Actually, in this case "Limit" is perfectly accurate. The only result of checking this is that the application must call advertisingTrackingEnabled before using the identifier.

If enabled is NO, then the developer can use the identifier for "frequency capping, conversion events, estimating the number of unique users, security and fraud detection, and debugging". Notice that compliance with these restrictions is almost impossible to determine.

http://developer.apple.com/library/ios/#documentation/AdSupp...

I can confirm. I have iOS7 running on my iPhone 5 at the moment.

Its actually clearer now since there is no "on" or "off" label. Its just off and white or on and green.

http://cl.ly/image/2U45141L3E3N / http://cl.ly/image/081l2v2W3E3H

Let's not forget about disabling "Interest-based ads," for which there is no path to get to... other than going to a link in Safari. The link is oo.apple.com

This KB article explains the process. http://support.apple.com/kb/ht4228

Interestingly, this launches some sort of stealth app, not a hidden page in the "Settings" app.

Also, there is another setting for including your location with iAd calls. That one is Settings > Privacy > Location Services > (Scroll all the way to the bottom, yes, past every single app that uses your location, about 35 in my case) > System Services > Location-Based iAds

Wow. You didn't make that up. That's past ridiculous ...

And ofcourse following the link in the KB article just produces a timeout from my ipad (though i do get a proper message that i'm not on an iOS device when accessing it from my mac). You wonder why they even bother? If you have no intention of opting people out, why even put up the page?

I suppose there is a legal requirement in some jurisdiction that the functionality exists, but how accessible to make the feature was left unspecified. Therefore, dark patterns prevail.

>Interestingly, this launches some sort of stealth app, not a hidden page in the "Settings" app.

If you go directly in Safari, it'll prompt to open in "iAdOptOut". I wonder how it works.

I also liked the idea of "Dark Pattern" that he linked and introduced.

I like how it's called "limit" and not "disable". Compared to infinity, anything with a finite value can be considered to have 'limits'. Even Apple's $100B in cash has a limit, despite the staggering size. Weasel-words are an interesting phenomenon.

Spotify came up with a sneaky one.

I recently signed up and it only asked for some basic info. By default, of course, all email notifications are turned on. When you try to turn them off and save changes - it won't let you save because one "mandatory" field is incomplete [1].

So you're forced to add more info if you want to turn off email notifications.

It's not a big deal, but it falls in the same category. I like the company less now.

[1] http://img20.imageshack.us/img20/4800/jfgt.png

That smells like a bug to me...

[I've encountered similar issues, e.g. where I had entered my home address on Amazon using a format which at some point started being considered "invalid" by their input-validation code. The old address remained in the system (and worked fine when ordering), but I couldn't change any other personal info without also re-entering that address, because the input-validation wouldn't accept the old one (it apparently still validated all the old data at that point, even stuff I hadn't changed). Unfortunately the new system was much more restrictive about what it allowed in addresses, so I didn't want to... >< ]

Spotify is known for its shady tricks. Try figuring out that the client consumes a very large portion of your upload bandwidth. Another one is when they enabled the "local files" feature via auto update a few years ago, this feature not only allowed you to play mp3 files via the client but also phoned home to mama with a list of all your local music. There was no way to opt out of this, the client just auto updated and started importing everything from windows default music directory. It seems as though they update their TOS every time i restart the client also, who would read that when you are in a rush to start your music.

The way they remove songs from the library is also a bit disturbing, some songs become gray and some just silently disappear. The frequency of removed songs is also quite high, for a free service i don't mind but if i pay premium and tons of my favorite songs disapear for each day i'm not going to be happy.

When they introduced the option to play local music, it was a pretty big launch and you were notified. And it has always been easy to select where to grab local files from, or disable it. I think your point here is invalid.

The fact that they remove songs is not unethical, so not really relevant.

Long, but worth reading — a great overview of shady techniques that many startups may feel are worth employing to bump their growth stats.

It was even more interesting to me as a set of 'canaries in a coal mine'. If you are working for a company and start to see many, or one, of these patterns, it should serve as awake up call. Move on or don't, but be aware of the corporate behavior.

I wish there was a browser plugin or something that displays all the blackhat manipulation a site employs as you visit it, and give alternative options.

Very well written. When you turn against your customers' welfare, you're almost always going down a path of failure in the long run.

Banks seem to be doing ok.

For now. It's only a matter of time before a startup or two breaks that industry down just like they have for many others.

I'm a happy customer of both Simple.com and USAA. Both are very customer focused banks. The industry is being broken down, just really really slowly.

Just curious, what benefits do you get from simple.com that USAA does not provide? I looked at it in the beginning but it didn't seem to add anything, perhaps that's changed.

That's a great question. I like the experience of simple. I get push notifications when I swipe my card. I enjoy the "Safe to spend" amount that takes into account my goals. Photo check deposit is great too. It's early days for them, but I like backing that horse.

USAA is a great company. They give you ATM fee reimbursements which is a huge benefit over simple. I feel guilty getting them since they have to eat that cost. In dealing with them, they're always pleasant and friendly, but they have policies like photo check deposit is only available if you're ex-military.

Hope that helps!

Serving customers who aren't military or from military families?

You can open a bank account with USAA if you're non-military. However, some of their other services are not available to you.

Like mobile check deposit, which is a big deal if you don't live near a branch.

I think there's an analogous problem with encryption and GPG: like with credit unions (USAA in particular) there's an easy solution out there, but relatively few people use it.

Things like regulatory capture make it hard for small players to get started in entrenched industries. Some examples: telecommunications (getting leases on existing pipes to sell your services), meat processing (look at the regs that small producers like Polyface Farms have to comply with), car manufacturing (look at the opposition Tesla is facing trying to sell direct to the consumer) -- and yes, banking.

How long until a Silicon Valley startup cracks the hard problem of low-margin heroin smuggling?

Would you trust your savings to a startup? Maybe if there was FDIC insurance. Would you expect the federal government to trust everyone's money to startups?

No way.

While banks may be horrible to their retail customers, the regulations that protect them exist for a reason. The current state of affairs is far, far better than having hundreds of thousands of banks which could fail and drop off the map with all of their customers' assets at any time.

FDIC insurance has changed that, but a prudent insurance company would not insure a startup bank, or if it did, it would need to collect so much for that risk that the startup bank couldn't offer anything competitive. The FDIC is almost certainly not going to do this.

Simple is a UI on top of a traditional bank.

Generally, it's illegal not to use banks beyond a certain point.

That is, transporting large amounts of cash is strong evidence of criminal behaviour, is subject to seizure, etc.

It is illegal to transport more than a certain amount of money over the (US) border unless you declare it.

It's funny. I'm on the phone trying to reach a student loan representative right now. I keep pressing 1 for English, 4 for a student loan representative, 1 again for the type of loan, then I'm back to pressing 4 for a loan representative. It's an infinite loop!

I try hitting 0 for an operator but then I'm prompted to give a 4 digit extension for the party I want to reach.

It could be that the pool of operators for your type of loan is empty (i.e. they're all offline or engaged), and the app is set to automatically let you try again with some other department.

Just going with Occam here, but I've seen a couple of callcentre call-routing systems and the potential for misconfiguration is fairly high.

Hanlon, actually. Just for the sake of terminology: http://en.wikipedia.org/wiki/Hanlon's_razor . It keeps me sane.

Most automated phone systems will connect you to an operator if you press 0 repeatedly (as in 10 to 20 times in a row). There's almost always a way to get around automated phone systems.

Or when you call don't press anything. The system may assume you have an old rotary phone that can't make the correct tones and automatically forward you to a person.

You can also try http://gethuman.com/

Student loans by themselves are an example of a business anti-pattern in my opinion. Schools and governments should figure out funding models that do not charge the students huge sums of money because it's perfectly possible to organize an education system without that (as many countries demonstrate). It is immoral to drive students into debt.

Or, if you want a more rational reasoning: nowadays higher education is essential to gain access to knowledge worker jobs which are becoming the foundation of the post-industrial economy. Nations should want the best quality of students in those jobs, not just those who can afford the entry ticket. Therefore it makes sense from an economic perspective to amortize the cost of entry (higher education) across the entire taxpayer base and leave education as a purely merit-driven system where the best students get the most economically significant jobs.

Great article. This reinforces the "Data informed, not data driven" mantra - you can optimise a feature on your site for a single metric but always take the human effects into account.

It also shows the importance of talking to your users in person. An A/B test may prove that ticking a box by default will get so and so many more users opting in, but it can't show you the look on their face when they realize what you just did to them.

One large claim made here is that while Dark Patterns may increase subscription rates, they hurt brand image. Are there any scientific studies measuring the magnitude of such damage? Perhaps the occasional irritation is worth 5 others "duped" into accepting more advertising.

Enjoyed the article, but I had to read it using the Clearly extension because the native web page is extremely laggy for me. I'm on a Win8 machine running current Chrome. It seems like such a simple design but hot damn does it chug when I scroll.

Works fine with safari on my 2008 macbook.

Same on Windows 7 with Chrome (IGP with a Mobile 1st gen i3 processor however)

That's probably because it's a 15MB page including all the assets -- Somebody doesn't believe in jpegs.

The page still downloaded for me in 2.5s though, so I have to give the guy some credit for having a beefy host.

I think it may have to do with the fixed 'border' the entire page has. It is interesting though how Chrome chokes and Safari glides, though with choppy repositioning of the fixed border on scroll completion.

Don't wanna start an off-topic conversation but in spite of being somewhat of a Google fanboy I'm actively using (and have set as default) Safari instead of Chrome on my 2010 macbook pro for any page or site that is "heavy". My usage of Chrome has now shrunk to scenarios that involve heavy dependence on omni-search.

I'm glad I wasn't the only person who had that issue. Strangely it works perfectly on the mobile version of Chrome for Android, but it grinds to a halt on the tablet version. This is especially strange as I have a Padfone 2 (phone connects to a tablet dock) , so the hardware is the same, and Chrome just gets relaunched in tablet mode when it's docked. Maybe due to the size of images being served being varied on screen size.

That would be the 300px radius inner box-shadow set to 5% opacity on the body tag, which happens to be over 40,000px tall.

Very nice article. ethics! this is the problem of our society. the next article should be about making people sensible. How to reward them for been responsible. The thing is people don't care about their live. Managers change jobs, people accept NSA... Might be intresting to compare Customer-Live-Time-Value und Business-Model-LiveTime-Value. What if they knew that the bomb will be nightmare? I would love to see numbers? How about calculating Update-Costs of the SW to "how about having a new business model". The aren't these "dark pattern" businesses VC driven?

This reminds me of the android ui for turning of WiFi when the screen is off. It used to be on one of the settings pages. Now you have to go to settings, WiFi, menu, advanced. There are numerous articles in the Google index of people having trouble finding this option. It makes a WiFi only tablet last for a very long time without recharging if you don't use it too much. Why would they want to hide that option?

Kind of prefer it is hidden myself. As a developer with apps that sometimes upload and the like in the background, I don't want countless emails from users who have flipped that switch and can't get their scores to post or whatever.

If you want all these patterns collected in one page, try booking a flight with Ryanair.

in iOS7 Advertising is in Privacy. Maybe apple are taking notes.

Stole the words right out of my keyboard. This is fixed, just not deployed widely yet.

Not to defend but for context most of this stuff isn't new and just looks like the direct marketing industry applying their well-honed scammy techniques. If you look at the old record and book clubs that used to advertise in magazines it's the same story.

I actually had to design one of these dialogs for LinkedIn for the new who viewed my profile when we flipped it to a tit-for-tat system (you show me your name and I'll show you mine).

We spent weeks changing the language to be as clear as possible without being overly verbose. I admit that the first (unreleased) rev probably was a bit shady, but in the end we ended up with something that's primary goal was to inform.

I'm super glad we went that way too. Being upfront and clear turned out to be the way to go. Sure we didn't get instant benefit like you'd get with an opt-out, but the conversion rates were very good even when we spelled out what we were doing.

What happens if you keep the setting off most of the time, and only switch it on for a few minutes to check out your profile visitors, then switch it back off? Would you still stay anonymous (unless others happen to check their visitor lists in those few minutes?)

It only starts showing you views since the time you switch from being invisible to prevent gaming the system that way.

Also, changing your profile visibility setting only affects your visibility on future visits to people's profile pages, not past visits.

All I'm going to say is that even though it's awful The Ladders have all those tricks, they're successful.

Not saying the ends justify the means, but I wonder how their business would look otherwise.

Well that's the kinda the point of Dark Patterns? Companies use them because they work. They are unethical, but they achieve success (in the short term at least, sometimes long tem). This is why there is law, to level to play in field so the ethical groups can be not disadvantaged.

Except, law is invariably perverted to provide an advantage to a select group.

Sure, bad law. I'm not saying law is perfect or flawless. There's a ton of bad law. We should fix the bad law. Unfortunately making law is tied up to politics and PR and minefield of stupidity that is.

Did you happen to read the section on Experts Exchange?

If you're making an empirical argument that these techniques provide more harm than good, you need evidence. If you're making an argument that they are just evil and if you do them you should have trouble sleeping at night, that's another argument altogether. I'm not sure which of these two arguments the article is trying to make, but it's very short on the empirical side.

It's an ethical and aesthetic argument about what makes a good design, brand, or business.

>"Brand relationships are like human relationships."

No. Brands, corporations, and other legal fictions are not human. They're tools explicitly designed to generate profit. This is their fundamental purpose, and as such, they can never bear any significant resemblance to human relationships.

Aside from that, it's certainly an excellent write up. And I like the term too: Dark Patterns.

I found this statement confusing at first too, but perhaps the way to look at it is that most customers do treat their relationship with brands by analogy to human relationships, whether they should or not.

We get mad at brands, vow we'll never use them again, break up with them, come back to them, expect them to behave coherently, criticise the behaviour of one representative as if it were the firm as a whole, demonstrate our loyalty to them in the hope of reward, and generally treat them as human counterparties in a relationship. Of course none of this is realistic or justified, but if you're running a company it's not something you can ignore.

If you're a user/consumer/customer of course, it makes sense to step back and realise that your loyalty means nothing to a corporation, they are not a person but composed of people, that they have their own internal goals, and they are rewarded by shareholders for some behaviour we'd call sociopathic in a person.

Amazing post. I have a pitch for a startup: It's an NGO which regulates email-marketing. To receive a certification of fair practices you have to abide by certain rules.

Users can see a badge indicating that you are a member or not. Eventually gmail and others will have to add a setting on whether to allow emails from non-members or not and until then, browser extensions will do the work.

Like?

So what is the final state of the switch be if you want your self NOT to be tracked or reduced the amount of tracking : OFF or ON ?

You want it to be ON

I thought in the EU you're legally required to do opt-in. You're not permitted to do opt-out.

How can the UK Post Office do opt-out?

Is this some UK exemption from the EU law?

Or is it the typical government exemption from its own law (like e.g. how politicians exempted themselves from the US CAN-SPAM law)?

Third party must be opt-in, but first party can be opt-out, that's why they do the alternate "opt-out opt-in" thing.

In fact a lot of companies I've come across do it this way, you see it on physical forms too.

Speaking of Europe, lately I've been using iOS on my native tongue and the translation is "Limitar seguimiento..." I don't know what's behind that ellipsis, but the direct translation of what I see would be "Limit tracking", without double negative.

Depends on the country, I think it is up to all member states to implement and manage the laws internally.

My former employer did the following "trick" to boost the amount of newsletter subscriptions (and angry customers):

In Denmark you are allowed to create a competition where you write "By participating in this competition you will be subscribed to our newsletter".

And then they write in the legal text under the form that you agree to them (and their mother-company) contacting you on phone, email and SMS.

And you also agree that two separate companies owned by the same company is allowed to contact you.

You are also allowed to leave the following checkbox checked as default, making it opt-out:

[x] Subscribe me to the newsletter.

Why, it is opt-in.

Just so confusing and misleading that a lot of people will opt-in by mistake.

The directive does not say that you must find a literal opt-in button on a web site.

Or you could be like Facebook and just ignore request to unsubscribe by hitting their unsubscribe button.

Does this site really need to be over 15MB large?

Well, considering 97% percent of the page load is images (average size: 400 kB), and the page contains slides that explain and complement the article... yeah, I think it does!

The page renders really laggy for me. Seems like it'd benefit from dynamically loading in the images, or at least user more compression on the images. Also, the images have been shrunken to fit the page. They should just be that smaller size to begin with; you could probably get the page down to 8MB with minimal loss.

How intentional.

way too many images - whats the TLDR on this?

TL;DR: RTFA.

This article is complaining about the steps apple took to prevent guid tracking but framing it as being done explicitly to trick the user.

I would suggest that the article makes a very good case for it being done to explicitly trick the user. Apple particularly gives a huge amount of consideration for making an intuitive and obvious UI. It's not a big leap to assert that in this case, they've made it as difficult as they can to opt out while still offering the choice.

More than likely it was done to protect the users privacy while keeping the exact same user experience. The protections were made behind the scenes with additional options made available to the user, but disabled by default.

The transitions from guid had the potential to break many apps so encouraging users to disable it would have likely caused many unintended side effects in addition to cluttering the ui and annoying users with unknown options.

I've read this three times now and I'm still not sure I get it...

When on the shakier side of a heated discussion of anything Apple, you can always fall back upon the nebulous "user experience" argument.

>I've read this three times now and I'm still not sure I get it...

For about ten years now, every time I use Paypal for an eBay payment, I'm interrupted to choose a $10.00 off opt-in for setting up credit, I have to click-thru this every time. By now, I've done this several hundreds of times.

I'm just a fish trying to stay out of their double-trap dark pattern slippery slope trammel net.

I've been on google+ since the beginning (2nd half of 2011).

For the whole 18-20 months whenever I have gone to google+ on the web it has first had an obstacle "Find your friends" page asking me to spam all my contacts about google+.

It still does it even today. There is no opt-out or "stop asking me to do this" checkbox. Even though I am an active plus user with posts and comments, it still treats me like a first timer.

I'm trying to get into google+ but there is no way I'm going to spam my friends.

That it has been doing this to me hundreds of times for nearly two years is unbelievable.

Another one. HN subject titles often link to the New York Times. Not having a subscription and offended with their offensive dark screen blinding pop-up `dark pattern' `hook' --free, or otherwise-- I now only read the respective HN comments.

NYT coverage is usually good, I miss it, but now BBC, Reuters, Guardian, rt, Der Spiegel, HN, are my primary news feeds.

Near as I can tell, he's saying that the advertisement tracking toggle (located under General > About, down past the version numbers, MAC addresses, etc.), named "Limit Ad Tracking" and defaulting to "Off" is actually another brilliant example of Apple's premier, user-centric interface experience design.

But I've never really figured out the "double-negative, Newspeak is good for you!" thing, so I'm not sure.

And how does this solution achieve that better than putting the option in the Privacy menu, labelled "Disable unique identification (this may break some apps)"?

The unintuitive layering and double negative wording clearly tells me that Apple tried to have it both ways.

Reply on news.ycombinator.com