wu :: forums (http://www.ocf.berkeley.edu/~wwu/cgi-bin/yabb/YaBB.cgi)
riddles >> general problem-solving / chatting / whatever >> How do you convince users to lie to their friends?
(Message started by: amichail on Nov 25th, 2007, 9:50am)

Title: How do you convince users to lie to their friends?
Post by amichail on Nov 25th, 2007, 9:50am
I'm working on a Facebook app to generate Facebook app ideas along with estimates of their popularity if they were to be built.

However, the app depends on users being willing to lie to their friends.

Besides avoiding the word "lie" in the app and mentioning the utility of the output, what else can I do to convince users to lie to their friends?

Title: Re: How do you convince users to lie to their frie
Post by towr on Nov 25th, 2007, 10:13am
Bribe them?
Convince them it's not really a lie?
Start a cult?

I'm not sure what to suggest without the faintest clue of what you're talking about, to be honest..

Title: Re: How do you convince users to lie to their frie
Post by amichail on Nov 25th, 2007, 10:17am

on 11/25/07 at 10:13:33, towr wrote:
Bribe them?
Convince them it's not really a lie?
Start a cult?

I'm not sure what to suggest without the faintest clue of what you're talking about, to be honest..


The app is a game with a purpose, much like the ESP Game.

The game involves tricking your friends into adding the app by lying to them about what the app does. The more friends that you trick in this way, the higher you will score.

For example, if you have a friend who is into virtual gifts, you might send him a notification that says "X has sent you a virtual gift".

So he would add the app expecting to find out what this gift is.  But then he would find out the true nature of the app.

The idea is to determine what sorts of notification messages are most successful in getting people to add the app.

These successful notification messages provide ideas for future apps that are likely to be popular.

Title: Re: How do you convince users to lie to their frie
Post by Grimbal on Nov 25th, 2007, 10:36am
There are these chain letters that say: "forward this mail to as many people as you can, and you will get to see a funny video".

I hate these.

In your game, are you supposed to lie to people who have agreed to play the game, and know you might lie to them, or are you supposed to lie to unsuspecting friends?

Title: Re: How do you convince users to lie to their frie
Post by towr on Nov 25th, 2007, 10:49am
I think he's suggesting the latter.
It does sound like a recipe for annoyance, doesn't it.. Never mind that if people are playing it for points, it'd be very easy to cheat. Just start a pool with friends and sends zillions of retarded app suggestions to each other. Or spam everyone on facebook, while your at it.

Title: Re: How do you convince users to lie to their frie
Post by amichail on Nov 25th, 2007, 12:13pm

on 11/25/07 at 10:49:49, towr wrote:
I think he's suggesting the latter.
It does sound like a recipe for annoyance, doesn't it.. Never mind that if people are playing it for points, it'd be very easy to cheat. Just start a pool with friends and sends zillions of retarded app suggestions to each other. Or spam everyone on facebook, while your at it.

Points are awarded only when new people add the app.

Why would anyone bother to cheat in any case?

Don't you think that novel ideas might emerge in the output, thus resulting in better apps being produced?

Title: Re: How do you convince users to lie to their frie
Post by amichail on Nov 25th, 2007, 12:15pm

on 11/25/07 at 10:36:48, Grimbal wrote:
There are these chain letters that say: "forward this mail to as many people as you can, and you will get to see a funny video".

I hate these.

In your game, are you supposed to lie to people who have agreed to play the game, and know you might lie to them, or are you supposed to lie to unsuspecting friends?

The deception is an important part of the approach -- just like deception is an important part of psychology experiments.

It's important that the lies are made to unsuspecting friends so that you can get reliable app popularity predictions.

Title: Re: How do you convince users to lie to their frie
Post by towr on Nov 25th, 2007, 12:58pm

on 11/25/07 at 12:13:57, amichail wrote:
Points are awarded only when new people add the app.
Define new.
Is there any point 'recommending' it to friends, if only new people count?


Quote:
Why would anyone bother to cheat in any case?
The same reason they would bother to lie. To get points.


Quote:
Don't you think that novel ideas might emerge in the output, thus resulting in better apps being produced?
I doubt that any good ideas that might come out of this will be noticeable amongst the crud it will produce.

Title: Re: How do you convince users to lie to their frie
Post by amichail on Nov 25th, 2007, 1:04pm

on 11/25/07 at 12:58:52, towr wrote:
Define new.
Is there any point 'recommending' it to friends, if only new people count?

New refers to people who have not added the app yet.


Quote:
The same reason they would bother to lie. To get points.

Maybe, who knows.  People may not bother doing anything that requires much effort.


Quote:
I doubt that any good ideas that might come out of this will be noticeable amongst the crud it will produce.

The ideas are ranked by how many people they have tricked.  

Title: Re: How do you convince users to lie to their frie
Post by towr on Nov 25th, 2007, 1:26pm

on 11/25/07 at 13:04:18, amichail wrote:
The ideas are ranked by how many people they have tricked.  
Yes, and as the internet has shown time and time again, crud often ranks higher than actual content.

And any crummy idea suggested by someone with a large network of devotees is an instant win.


Quote:
New refers to people who have not added the app yet.
So any good suggestion that aren't first in a network of friends are already dead at the start.

Title: Re: How do you convince users to lie to their frie
Post by amichail on Nov 25th, 2007, 1:29pm

on 11/25/07 at 13:26:58, towr wrote:
So any good suggestion that aren't first in a network of friends are already dead at the start.

I'm actually going to let people get points in two ways. You can get one point by tricking a friend into adding the app.  You could also get a point if you were the first to suggest a particular idea and someone was tricked with that idea into adding the app.

So even if you have few friends, if you were the first to suggest an idea, you would get points when others are successful in tricking their friends into adding the app using that idea.

Like the social news site reddit, you will see hot/top/new rankings of submissions.

Title: Re: How do you convince users to lie to their frie
Post by Grimbal on Nov 26th, 2007, 12:49am

on 11/25/07 at 12:13:57, amichail wrote:
Don't you think that novel ideas might emerge in the output, thus resulting in better apps being produced?
No, it will result in existing apps being distributed more widely to people who didn't want it.  You know, like spam.
Besides, it take its toll on friendship.  The true value of friends is not that you can lie to them to sell them apps.


on 11/25/07 at 13:04:18, amichail wrote:
Maybe, who knows.  People may not bother doing anything that requires much effort.
Like for instance spending time on writing apps that are actually useful and don't need deception to be successful.

Title: Re: How do you convince users to lie to their frie
Post by JiNbOtAk on Nov 26th, 2007, 1:13am

on 11/26/07 at 00:49:16, Grimbal wrote:
You know, like spam.


Funny, first time I read amichail's post, it seemed he is talking about spam. Deceiving others ( friends or otherwise ) in order to get reliable app popularity predictions is an oxymoron, how do you get reliable predictions from lying to others ?

Title: Re: How do you convince users to lie to their frie
Post by amichail on Nov 26th, 2007, 8:03am

on 11/26/07 at 00:49:16, Grimbal wrote:
No, it will result in existing apps being distributed more widely to people who didn't want it.  You know, like spam.

Yes, if you are referring to this app.  There's an incentive to make it spread (as is often the case with Facebook apps).


Quote:
Besides, it take its toll on friendship.  The true value of friends is not that you can lie to them to sell them apps.

Once you add the app, the lies stop. I think Facebook apps are blurring the distinction between truth and lies. If someone gave you a beer, but it turns out to be only a virtual beer, then is that a lie?


Quote:
Like for instance spending time on writing apps that are actually useful and don't need deception to be successful.

That's what this app would help do by telling developers which apps are worth building.  

Title: Re: How do you convince users to lie to their frie
Post by amichail on Nov 26th, 2007, 8:06am

on 11/26/07 at 01:13:36, JiNbOtAk wrote:
Funny, first time I read amichail's post, it seemed he is talking about spam. Deceiving others ( friends or otherwise ) in order to get reliable app popularity predictions is an oxymoron, how do you get reliable predictions from lying to others ?

Psychology experiments use deception in a major way to get more accurate results.  Also any use of placebo in medical experiments is deception.

Title: Re: How do you convince users to lie to their frie
Post by ima1trkpny on Nov 26th, 2007, 8:36am

on 11/26/07 at 08:06:26, amichail wrote:
Psychology experiments use deception in a major way to get more accurate results.  Also any use of placebo in medical experiments is deception.


Yes, but you do realize that if successful, you will unethically greatly increase the "pest potential" of advertisers? Not to mention, lying about what the app does is no different from the way people sending viruses, etc will try and conceal it is a virus. If you can't trust your friends to be honest, what mail or internet communication can you safely open? To say it quite frankly, I think this is a very poor idea and you should drop it.

Title: Re: How do you convince users to lie to their frie
Post by Grimbal on Nov 26th, 2007, 8:41am
OK, I reread your initial post, and I think I jumped to conclusions a bit fast.

So what you want to do is, in other words, to design an app that is an empty box, just to see how many people would actually pick up the box and look inside?

So you need people to actually recommend that empty box to their friends, and you want to know how to convince people to do that?

The Milgram experiment would suggest that you should wear a white blouse, use a doctor title and tell the people that within the experiment, they can lie, you take all the responsibility.

More seriously, you must design your app in such a way that the one installing it doesn't feel hurt of cheated in the end.  For instance have it show a prototype and tell "this app is not available yet, but if you like it, le me know, if there is enough interest, I will write it".

But I would prefer the approach to have a "box of ideas" where people can look at the descriptions of potential new apps and vote for their favorite ones.

Title: Re: How do you convince users to lie to their frie
Post by amichail on Nov 26th, 2007, 8:45am

on 11/26/07 at 08:41:59, Grimbal wrote:
But I would prefer the approach to have a "box of ideas" where people can look at the descriptions of potential new apps and vote for their favorite ones.

Yes, that's an alternative.  But a vote is different from adding the app because you think it's already made. The latter is what would happen if the app were already made and thus may indicate a more accurate prediction of popularity.

Also, the latter has the advantage of making the app spread, so you can get more data.

Title: Re: How do you convince users to lie to their frie
Post by Three Hands on Nov 26th, 2007, 11:46am
Largely on the basis of the initial poat, maybe calling it something like "Fool your friends!" would work.

Makes it seem a bit more of a game/simple prank. However, there's the interesting issue of the app having to be seen as popular enough to be passed around, and then get the limited data coming back of whch ideas proved popular for fooling people. Hence, it seems that the statistics are liable to be skewed...

Title: Re: How do you convince users to lie to their frie
Post by Grimbal on Nov 26th, 2007, 2:16pm
I don't know how Facebook apps work and how they get distributed.  I imagine it uses some kind of viral marketing.

It seems to me that people will first try out the app and then maybe recommend it.  That means that they will decide to recommend it or not based on the real content, not the app you pretend it to be.  So, the speed at which it spreads wouldn't measure the popularity of the to-be-written application, but the popularity of the process of tricking the people.

There is a joke e-mail that pretends it can use your screen as a camera and make a picture of you.  When you try it, it shows a picture of an ape.  It was popular, but that doesn't measure how much people would like to convert their screen into a webcam.

Title: Re: How do you convince users to lie to their frie
Post by amichail on Nov 26th, 2007, 2:28pm

on 11/26/07 at 14:16:33, Grimbal wrote:
I don't know how Facebook apps work and how they get distributed.  I imagine it uses some kind of viral marketing.

It seems to me that people will first try out the app and then maybe recommend it.  That means that they will decide to recommend it or not based on the real content, not the app you pretend it to be.  So, the speed at which it spreads wouldn't measure the popularity of the to-be-written application, but the popularity of the process of tricking the people.

Yes, many facebook apps provide an "invite" feature where you can recommend the app to your friends.

But that's not the only way to make an app spread.

What you can do is send a notification to a friend such as:

"X has given you a {gift}."  

Where gift is a link to add the application, which you might want to do to find out what the gift is.

But yes, I agree that there can be a skew since not everyone is willing to trick their friends for points.

Title: Re: How do you convince users to lie to their frie
Post by FiBsTeR on Nov 26th, 2007, 2:34pm
A person on Facebook is more likely to use an app that they have seen work. I personally don't bother to install an app I've never seen used, let alone an app I have never heard of.


on 11/26/07 at 14:16:33, Grimbal wrote:
It seems to me that people will first try out the app and then maybe recommend it.


I know that apps are more effectively spread when you see that your friends use an app that you find particularly interesting, and you then go and get it for yourself. I tend to ignore invitations to use an app because, quite frankly, people just press the "Send invitation to all friends" button (not actually called that, but you get the point).



Powered by YaBB 1 Gold - SP 1.4!
Forum software copyright © 2000-2004 Yet another Bulletin Board