Add option to hide total receiving from others

[juliepagano]

Is it possible to add the option to hide the total you receive from others, similar to the option for hiding what you give?

This came up with some folks I was trying to encourage to sign up for gittip today and they voiced some valid concerns around it. For some, there can can be negative consequences from having that information being so public.

It would be nice to add this option to encourage more people to feel safe using gittip. :)

The $15 bounty on this issue has been claimed at Bountysource.

[seanlinsley]

I may just be oblivious, but when wouldn't it be safe to share that information?

[chadwhitacre]

@seanlinsley Presumably when you're an outspoken woman in tech, and a steady stream of trolling and abuse has led you to guard your privacy.

@juliepagano Good call. I can't promise how soon it will happen but I'm +1.

[juliepagano]

Woops. I just accidentally clicked the wrong button. Good job, me. :(

[chadwhitacre]

:-)

[juliepagano]

@seanlinsley In this case, these folks are women bloggers who are often harassed about asking for support for their work. They've been discussing some issues related to harassment around getting too much/little, being accused of "begging", privacy issues, people policing their spending, etc.

I think adding this as a feature would make gittip much more appealing to them and others like them.

[timothyfcook]

+1, in general there's a lot of potential for Gittip to be customized based on the context and interest of whoever is being supported.

However, there may be some extreme cases that test this allowance in terms of the "Open Company" model. If Gittip gets really big, certain people could start earning some really serious money. In those cases, it might be truer to the Gittip model to allow the community to see just how much people are earning so they aren't over-compensated. Obviously, this isn't an issue yet.

Though, certain people might also want their total receiving amount hidden just for reasons of humility, etc. if they aren't interested in broadcasting a "I earn more than you" message.

Just want to note this in case, in two years from now, someone's earning $20,000/week and hiding it from the public. Potentially their supporters wouldn't support as much if they knew they were earning so much? It loses that valuable social control that is so valued in the "Teams" function.

Maybe some sort of system could be implemented where Receivers could switch visibility on/off, but a critical mass of their supporters could force-switch it back on? Or maybe, if they switch it off, their total receiving could only be visible by supporters who have contributed a certain amount of $? Maybe all supporters who have contributed 10 consecutive Gittips and have totaled >$50 in support.

Seems like trollers/haters wouldn't donate $50 over ten weeks just to find out someone's total receiving...

[heidigardner]

+1

[timothyfcook]

Been meaning to play around with BountySource, here goes nothin': https://www.bountysource.com/issues/1329420-add-option-to-hide-total-receiving-from-others

[seanlinsley]

Though, certain people might also want their total receiving amount hidden just for reasons of humility, etc. if they aren't interested in broadcasting a "I earn more than you" message.

I'm interested to see this topic discussed on Chad's call with DHH on December 3rd 🐱

On the subject of implementing a system to manage visibility, I think the if we're going to support income hiding then we should respect that person's wishes 100% of the time. That's both to garner a sense of trust, and because no one is forcing you to give to someone who hides their income. Subbable does just fine while hiding the amount raised.

[timothyfcook]

Yeah, it might not cause any issues. Just wanted to raise the possibility. Part of the goal of Gittip (at least the "Open Companies" aspect) seems to be about increasing transparency and working against corporate anonymity. Hiding tips possibly works against this.

It seems reasonable to implement income-hiding for everyone except committed supporters. Maybe there is a better possibility.

[seanlinsley]

This feature could be quite a deep rabbit hole to jump into :-)

[MikeFair]

Users can/should absolutely be able to keep private the amount they are
receiving from the network if they'd like. I don't see it as an open
company dilemma at all. There's a distinction between the project's
data/actions and the users data/actions.

An open company can absolutely keep their client's data private.

Something that can help here is recognizing/deciding who exactly owns what
"data" in the database? I say the users themselves each own their own data.

Basically most of the data kept in the database can be classified as
"personal" information about a user. Tip allocations, giving history, and
how much they are receiving.

Until there's something more formal, an easy way to get started could be
something like "it can be kept private as long as the total within the year
is below the US Federal annual exclusion limits for the gifts (currently
$14,000/year)". There's really no technical or legal reason for dong it
that way, but it seemed like a number that was kind of related in the given
context.

Another number might be whatever the US considers it's poverty threshold.
It's a US specific answer, so other countries might have different rules;
but anything happening below the poverty thresholds is really not worth
hassling over in my opinion. We can keep the aggregated statistics public,
and perhaps the flag saying whether or not you're receiving you're target
and the end user can control the rest.

It's about giving the people doing the giving the information they need to
select whether or not they individually wish to support that recipient.
And just like jumping off bridges; personally selecting to support someone
isn't something that needs to be predicated on whether or not other people
are supporting that person too.

The system can take care of handling redirecting any over allocations to a
particular individual because of mass giving to the same person. That's
not something that the users are going to need to control.

On Wed, Nov 20, 2013 at 3:30 PM, Sean Linsley [email protected]:

This feature could be quite a deep rabbit hole to jump into :-)

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/1683#issuecomment-28943307
.

[seanlinsley]

personally selecting to support someone isn't something that needs to be predicated on whether or not other people are supporting that person too.

I consciously give more to under-supported people / organizations on Gittip because their relative need is higher. While you're right that it doesn't need to be predicated on their current income, it certainly helps to know how much more support they deserve in relation to the rest of this micro economy.

I'm currently +0 on this feature because I'm not sure how it might negatively effect our gifting community.

[duckinator]

+1

[MikeFair]

"I consciously give more to under-supported people / organizations on
Gittip because their relative need is higher."

Right, what I say you're attempting to do is maximize the number of people
who, as I like to say it "get done"; i.e. are meeting their weekly goals.
And you're starting with those "furthest away" from their goal (aka the
most need).

What if the system could do that for you? Would that change your opinion
here?

So here's the "algorithm/payout strategy" in a procedural terms:
Note: the model here is that each donor has multiple funds they can tip
through which are processed in order of their assigned priorities (assigned
to by the donor).

1. Take the donor's money and allocate it to their targets starting with
   the highest priority fund.
2. For everyone who has received over 13% of their target (when you
   cumulatively aggregate the total from everyone) or has reached the donor
   assigned maximum; refund the money back to the donor's gifts and prorata
   reallocate it excluding those people who have "got done" (or reach their
   limit).
3. Repeat until the donor's money is gone or everyone in that fund has "got
   done"
4. If everyone in the fund has "got done" the money overflows to the next
   highest priority fund from the donor.

Note 2: Most people think of funds as being "hand built" but the idea is in
addition to hand building funds they can be built automatically by
different algorithms.

We will need to work out the formulaic expression details but I see no
reason we couldn't find a way for you to express

"Pay out X to those in the system that are asking for the least amount and
have the least amount of money allocated to them after all other monies
have been paid out". The system would then tell you who you gave money to
after payday. Like a "surprise" to you to find out who you gave to that
week.

The ability to express these kinds of statements would make fund
construction much more fun.

Note 3: I'm looking at the scientific/mathematical languages/systems to run
these pay out algorithms. I just found out about "Julia" (julia.org) that
looks like a very promising way to distribute the payday/payout algorithm.

If you didn't have to personally select who to give money to, but instead
of have the system give you the results of a search it ran to find people
who met that criteria (without telling you the exact details of how much
they give or receive) would that work for you in your case?

On Wed, Nov 20, 2013 at 9:49 PM, Sean Linsley [email protected]:

personally selecting to support someone isn't something that needs to be
predicated on whether or not other people are supporting that person too.

I consciously give more to under-supported people / organizations on
Gittip because their relative need is higher. While you're right that it
doesn't need to be predicated on their current income, it certainly helps
to know how much more support they deserve in relation to the rest of this
micro economy.

I'm currently +0 on this feature because I'm not sure how it might
negatively effect our gifting community.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/1683#issuecomment-28959760
.

[duckinator]

@MikeFair while a fascinating idea, and something I would personally love to start playing around with, I don't think it's worth delving into for this. Unless I'm missing something huge, the only negative impact is that it may skew what people who don't show total receiving actually get -- they may not get as much (or may get more) because people can't as easily compare. Given that it's opt-in and only those who opt-in could be affected, I don't consider it a significant problem.

[tshepang]

Perhaps I'm just too used to it, but I really love the idea of seeing what each person is getting, so I am -1. I love the transparency. Do you guys really think Gittip is getting hurt by being this transparent? What happened to @juliepagano?

As a sidenote, was an idea of reducing this transparency (i.e. the title of this Issue) discussed somewhere?

[juliepagano]

@tshepang Nothing happened to me. I'm totally fine with leaving my amounts public for now. However, I like the idea of people being able to decide.

I got the feedback from some folks I was trying to get interested in gittip, so I could donate to them. The lack of privacy immediately jumped out at them as a barrier to them joining. We had a good conversation about why that convinced me it was a good feature to add.

If not seeing how much one receives bothers someone, they can always choose not to donate to people who keep that information anonymous. I think a system like this leaves plenty of room for people to use it different ways.

[bruceadams]

I'm a solid +1 for this. Supporting personal privacy is important. The only downside I can see is that it might confuse users a little, which seems like a small problem.

[sigmavirus24]

+1

[chadwhitacre]

To be continued on #1721 ...

[chadwhitacre]

Process derp, sorry. IRC

[chadwhitacre]

There seems to be less than consensus on this ticket. See also IRC. What I'm hearing is that from the personal privacy standpoint, the feature makes sense, but that it might have unintended negative consequences on Gittip as a whole. Shall we unpack that a bit?

I guess I've been expecting that this feature would be seldom-used and would basically create two Gittips, in the same way that protected Twitter accounts seem to create two Twitters: the public one and the private one. My sense is that the private Twitter is much smaller (anyone have data here?). Certainly protected accounts are a very, very small part of my own Twitter experience. I follow maybe ... two or three protected accounts, out of 952? I'd be fine for Gittip to end up like that: a "public" Gittip with an overwhelming majority of users, and a much smaller "private" Gittip. However, if Gittip were 80% hidden or even 50%, I think I would feel quite differently about Gittip than I do now. Anyone else? What can we say about the desirability or likelihood of a "mostly private Gittip" scenario?

[rummik]

It seems like funding goals will keep the majority of it public, plus most people probably aren't going to tip private accounts. So I'm guessing we'll wind up with the "mostly public Gittip" scenario

Edit: Also, +1 from me :)

[juliepagano]

My expectation is that this use case would make gittip an appealing platform to a new demographic of users. That demographic is largely comprised of underrepresented groups, so that is likely to keep it to the desired minority. I also think the nature of the people involved in the system will keep this to a minority. As others mentioned earlier in the thread, some people will be less inclined to donate to someone who has their amount received anonymous.

The impact on the ecosystem seems similar to the impact of setting your giving to anonymous, which is already a feature. My impression of gittip is that it is for giving small, no strings attached, weekly gifts to people when you appreciate their work. I don’t see anonymous values (either for giving or receiving) as counter to that goal. Absolutely requiring that you share these values seems a bit counter to the “no strings attached” part.

[tshepang]

In response to #1683 (comment) (by @duckinator), why shouldn't Gittip promote transparency amongst its users as well? Without users, Gittip is nothing. Imagine a transparent Gittip, but not-transparent users. What would be the point? You are not going to redeem the economy if you do not strongly encourage your users to take part in that movement, Gittip being the tool towards that goal. Many users are not going to care about that movement (nor will they be opposed), but being forced to be transparent would be win for the movement regardless.

[duckinator]

If they never have to think about it, they won't. I'd say offering the option would probably make them think about it more. What the end result of that thinking would be is rather unpredictable, however.

[tshepang]

The popularity vs. idealism argument is a challenging one. Part of the reason is that it's not possible to predict what the future holds, and it doesn't help when we don't have data. Taking the example of this feature request, Gittip has been growing well even with this forced transparency, yet @whit537 is concerned that we might lose users. Would it really grow faster? By how much? We can't predict that. We can take a chance though, and I think if we don't have data, the ideal should be preferred. What if we'd gain, say 10% more active users (by enabling this feature), but slowing progress of a more open economy by 20%? We can only speculate here.

This reminds of the Free Software movement, where adherents would work hard to ensure their software works on closed platforms (e.g. Windows). I think a common reason is that these people feel that popularity trumps idealism, at least in the (selfish) case regarding their creations. It's an understandable thing, given that creators want their work to be experienced by as many people as possible, but it's opposed to the idea of killer (platform-specific) apps, which would encourage people to move off the closed platforms. There are pros/cons, and there is no easy way to tell which is greater than the other. It's just speculation really, even in the case of superstars like Firefox (or so I think).

[tshepang]

@whit537 if you decide to enable this feature, please keep transparency as a default option. That will help in cases where users don't care either way.

[zbynekwinkler]

To see the power of defaults (most people do not care to change them) witness the post from Dan Ariely about consent with organ donations: http://danariely.com/2008/05/05/3-main-lessons-of-psychology/. Also the growth of flickr is being attributed to the default option to share all pictures (I can't find the link right now).

[chadwhitacre]

@tshepang

You are not going to redeem the economy if you do not strongly encourage your users to take part in that movement, Gittip being the tool towards that goal. Many users are not going to care about that movement (nor will they be opposed), [...]

Yes!

[...] but being forced to be transparent would be win for the movement regardless.

Well, ...

Would it really grow faster? By how much? We can't predict that. [...] There are pros/cons, and there is no easy way to tell which is greater than the other. It's just speculation really, even in the case of superstars like Firefox (or so I think).

Yes! Well said. I've decided to +1 this feature, but ultimately for me that's a decision based on speculation (or as I phrased it, "a subjective art").

[chadwhitacre]

[P]lease keep transparency as a default option. That will help in cases where users don't care either way.
To see the power of defaults (most people do not care to change them) [...]

Accepted. I just want to be careful not to do what Facebook comes across as doing: using confusion regarding privacy settings as a tactic to increase sharing. Hopefully we're helped by a parallel commitment to corporate transparency.

[chadwhitacre]

What the end result of that thinking would be is rather unpredictable, however.

Yes! This is the fun part, in my view. Let's demonstrate that economic trust and openness is possible by practicing it ourselves. As @/rtomayko has said, "Lead by example as loud as possible." Let's make Gittip easy for everyone to join, and also let's make it plain what the heart of Gittip is. Not as a bait-and-switch, mind you—no coercion! Anyway, we can't coerce people to trust, we have to draw trust out of people by creating a community and a culture worth trusting.

[chadwhitacre]

Okay! Let's land #1721 and move on with life. :-)

[seanlinsley]

@whit537 how do you see the UI working, especially if plural users aren't allowed this feature? What happens when an account goes back-and-forth between singular and plural?

[chadwhitacre]

@seanlinsley Sorry, forgot to mention it: You're right, distinguishing between singular and plural users here is a bad idea. The question of transparency for larger givers/receivers is still on the table, but I don't see that we need to solve that as part of this ticket.

[jslegers]

IMO a good reason not to show how much you're earning on Gittip, is being EITHER among those who receive little to no tips or those who receive most tips.

If you're among those who receive little to no tips, many people are likely to conclude that they shouldn't tip you either... because surely others would have already beaten them to it if your efforts were worth tipping.

If you're among those who receive far more tips than most, many people are likely to concluse that that they shouldn't give you any more tips... because others need them more than you do.

The end result is that not being able to hide the amount of tips you receive is likely to have a negative impact on both extremes of the Gauss curve... with mean and median members probably benefiting most from NOT hiding their amount. Or am I missing something?

[zbynekwinkler]

Deployed 💃

[juliepagano]

Thanks so much for implementing this! I really appreciate it. ❤️

[seanlinsley]

Why is the user description style different for those who hide both?

[timothyfcook]

Awesome! Stoked to pay out my first bounty on Bountysource.

[clone1018]

@seanlinsley #1847 (comment)