I'm not a big fan of this idea, though I probably can't express my reasonings with a lot of clarity at the moment. I'll certainly try. First, I'm all for a way for Hubski to raise some money. I'd be more in favor of either direct donations (which probably exist somewhere) or some sort of premium membership (no idea what that would look like, but I'm thinking of reddit gold) - but I'd support that. This is my first problem. Where and how? I have the bite in the back of my head that tells me this idea was probably jolted into existence by the opening of Circle - which I have no experience with, no confidence in, and no knowledge of how it works. If it's (edit: 'it' being the Hubski system, not Circle) just paypal operated (another site I have issues with, but mostly just because their customer service is terrible), I'd be a bit more okay with it. But really, the "Hubski Bank Account" isn't something I have a giant problem with, just an insecurity. This is a bigger problem for me. I don't like the idea of being charged to post content, even if it is one cent. Even if a ten dollar deposit would give me 1000 posts+shares, I just don't like the precedent. I'm sure it seems minuscule and insignificant, and there may very well be no logical reason for it. I simply don't like it. This is also exactly how badges work already right? Badges "transfer" from person to person when they're awarded. If this were discussed further, I'd want to see statistics on badge movement. Do they actually move around? I've got seven (a fact I didn't know until I turned off Zen mode a few moments ago to collect info), and they definitely don't go around. This is important because... I don't think Hubski's content should have an economy on social capital. It has an economy built in: interest and discussion. I think implementing something like this would have serious upward distribution. In fact, I don't think it would have anything but upward distribution. Power users with hundreds of followers will be able to make an absurd amount when compared to newer or simply sharing users. For instance, thenewgreen has 977 followers. Assume half of his followers share something, that's $4.89 he's made with a .01 cent investment. Great return for him. I have 15 followers. If I share something, and half of my followers share it, that's .075¢. Not a great return. If all of my followers share it, it's 15¢. Not a great return. That's obviously discounting the cascade effect a "share" actually has. Obviously if thenewgreen shared my post, it has a potential to reach another 977 people, which would open me up to his "market." Which brings me to... Trickle-down economics is a bad idea, and it's never worked. Though, with extremely microscopic transactions like 1¢, it probably has a much better potential here than anywhere else. Again, I want to see information on badge distribution and storage. thenewgreen has 73 badges (I see him badge things quite frequently). For users with a lot of badges, do they badge enough to sustain an economy? It's not something they should really be concerned with, I think. While I understand the desire to keep out spammers, I really wish the rhetoric on "undesirables" on Hubski would disappear. It comes off as elitist. That's an irrelevant nit-pick.
--- Despite all of that, I do find the idea interesting. My biggest problem is the eco-speak in the presentation. Hubski shouldn't be a marketplace. It should be a forum. I don't want people to pay me for my comments here, and I don't want to pay people for their comments. In the rare cases which I do, I'd give a badge. Here's my suggestion, which I think could test this idea and some of the theory behind it: Introduce a way to buy "marks", a strict one-time-per-user test run. $1 = 100 marks (making each mark worth 1¢, the proposed price). I'm imagining a clone of badges (the only reason I suggest using something other than badges is so that the integrity of the test isn't compromised by existing badges or badging behavior). Marks can be given in the same way that badges can, received in the same way, but also go away with each post. So for users who have invested the $1, a post now costs 1¢. Track the transfer of marks to and from users for the duration of... between one and three months. I think this would be enough time to track some of your theories and see if they work. It also makes the experience voluntary, which means there will be an existing control group of people who don't participate in buying marks. This would allow us to see how behavior changed when money was introduced into the equation (even an infinitesimally small amount). At the end of the test run, our subjects will be the evidence. Those who participated could only have initially purchased 100 marks. How many do they have now? How many did they give away? I'm sure a dozen other metrics could be taken from the information as well. I'm purposely ignoring shares here because they're a metric that I think should remain totally free of pay. I don't anticipate a great deal of difference between the shares and marks in this test, however. My theory is that if a post is worth badging or marking, it's worth sharing. That's a possible weakness to the test, but it's my suggestion that shares be completely free. This would also be an incredibly interesting study of extremely micro-economics which could be cool to watch and participate in. My testing suggestion probably has issues too, but I think a test run with something more focused (marks (or whatever), instead of shares) would be beneficial). ---- I realized about halfway through that I was being really negative, and that wasn't my intention. I get very worried when people start putting price tags on opinions, and when free discourse becomes a market. Things can go very bad in a very short amount of time, and it worries me (just look at the evening news). 1¢ is basically nothing, I get that. I worry about down the road. You create a Hubski bank account of sorts
Then, every time you comment or post, you are charged a fee (say, one or two cents), which goes to Hubski's coffers. Alternately, when you upvote or share, the same fee is transferred to the account of the user whose content you are appreciating. You would also have the ability to transfer credits to whomever you want without having to share or upvote.
Third (and this is the most interesting aspect, IMO), it would create an economy on the site based on social capital, as credits would be passed around, bought, sold, and given away based on appreciation from the members. The negative side, of course, is that this could represent an upward redistribution of wealth from the poor to the rich, so to speak, but it could also encourage development of ideas, since each comment would have a value.
It is my conjecture that members who may not have much or any money, but who are valuable would be supported by the richer among us.
I think this does a couple things. First, it will keep out spammers and undesirables.
Just a clarification - it seems from your post that you think a badge, once received, can then be given again. That is not the case - badges can only be given once. So I have 4 badges right now. If one is given to me for a post, that doesn't go into my badge count and I can't re-give that badge. Right now they're a one-stop shop. I was thinking about this proposal with cryptocurrency in mind (sadly it doesn't quite work out) and I think the Hubski bank account should be like a wallet - but then, either we would have to use a 3rd party provider or the work of the people behind Hubski would increase, which I think might counter the point of raising money - it could increase the cost of running Hubski as opposed to defraying it. I also like your idea of running experiments and looking at the data as opposed to just jumping in on this head-first.
I don't think so. I would envision a system where the credits you buy have no cash value. That is, they can't be redeemed. Otherwise, Hubski turns into a potential income source, and the accounting get real complicated, real quick, and I'm sure as shit not going to jail for not providing 1099s to people all over the world. Edit: I may have misread what you were saying, but I still stand by the above if anyone else is wondering....I think the Hubski bank account should be like a wallet...
If the income is under 20 dollars or something, it doesn't have to be reported. So I highly doubt there'd be any problems with redeeming currency. That said, I don't like the idea much either, I'm behind CashewGuy on the donations idea. I've got like 30,000 dogecoins you guys can have, plus I'd donate some cash. If there is a way to donate it isn't very apparent, so if you do accept donations you may want to post that somewhere obvious.
Well, that is one way of generating some cash :)
That's what I thought, thanks for correcting me! Call me a technophobe, but I think cryptocurrency (in pretty much every implementation I've seen up to this point) is a bad idea. If the system were built on cryptocurrency only, I definitely wouldn't participate. ironpotato is the polar opposite, he practically fetishizes dogecoin.Just a clarification - it seems from your post that you think a badge, once received, can then be given again. That is not the case - badges can only be given once. So I have 4 badges right now. If one is given to me for a post, that doesn't go into my badge count and I can't re-give that badge. Right now they're a one-stop shop.
I was thinking about this proposal with cryptocurrency in mind (sadly it doesn't quite work out)
I was thinking that we should make our own Hubski currency? And that posting would be like "mining"? It was mostly metaphorical. Then I realized that the cryptocurrency would need to be based on actual currency because the point of this would be to fund Hubski and that to me then seemed to defeat the basic purpose of cryptocurrencies (as they are currently imagined/practiced).
Interesting points, CG. Although b_b qualified it, I want to reiterate this is little more than the product of a conversation we had in the hallway a couple of hours ago. It's not even half-baked. Basically, b_b said "What if comments and posts cost a penny?", and we started riffing on it from there. The first problem we saw, was that it might stifle good conversation. For that reason, we came up with the approach that shares and comment upvotes might transfer a penny from the appreciator to the commenter. In that sense, good conversation is free. However, once we happened on that, we started to think of the mountain of credit that kleinbl00 would be sitting on. Although it might not be undeserved, what would he do with it? Our answer to that was that he should be able to give it away to anyone he wanted to. The 'marks' experiment is an interesting idea. TBH even if we thought this idea was a good one, it would probably be better to first convert dollars and cents into something like marks if for anything just to rinse off some of the connotations that come with cold hard cash. Also, if we found that 1.2 cents or 0.8 cents made for a better rate, we could do that. One aspect I do like about this idea is how someone could come to Hubski with $1.00, and because they were so valued, never pay anything beyond that. But, the question then becomes: Who is paying more, and why? I wouldn't say that this idea is a winner, but there are some interesting components to it.
To Hubskify this idea, the longer and more pinwheeled a comment is, the less it should cost.
That's far too simple. I want to see grammatical analysis applied server-side so long, complicated, information dense posts are forced into existence by economic pressure and low effort one-liners like this cost me a dollar!
An arbitrary value. If I were going to do it for real, I'd set a range of costs between $0 and $Something where $0 is the Gaussian analysis of the length*grammar score of highest voted, densest comments in the entire Hubski database and everything else falls away to $Something that would be enough to make one think twice about being glib and lazy. A Thoughtful Web Is Free; For Everything Else, There's Mastercard. This would still be costing me money.
I agree. I was just saying to b_b, that if there is a lesson here, it is that money should have nothing but an arbitrary relationship to interaction on the site. It seems once you create a link between the two, you must then work to correct for all the ills that it begets.
Based on your It's basically a 'rich' guy's upvoteI realized about halfway through that I was being really negative, and that wasn't my intention. I get very worried when people start putting price tags on opinions, and when free discourse becomes a market. Things can go very bad in a very short amount of time, and it worries me (just look at the evening news). 1¢ is basically nothing, I get that. I worry about down the road.
This is exactly my opinion. Commerzialized opinions are inherently bad and in social networks this can create a distribution worse than the power of law distribution. It discourages new users and encourages circlejerking.marks
idea I want to propose something: How about you can buy 1 mark for $1 and these you can use like badges or upvotes and show your appreciation and these are then added to the account of the user you gifted them. This user than can give them a different user as appreciation, though with every new 'transaction' the mark looses in value (like a coin, that gets more and more marks of destruction, up to a point (10 txs?), that it can't be sent anymore. If you buy marks, these don't get added to your account, so no hoarding ensues or 'bought appreciation' is created. Also you see how much marks a user had, like "10 marks currently (25 marks total)", so hoarding gets discouraged and they get passed along. To show on the profile a mark with less 'value' got send, either you display them next to each other or you can use decimal digits ({1,...,0.1} at the first tx the mark doesn't use value).