Bitcoin and the Future of Social Networks (Video and Script)

2019-02-22T13:59:13.000Z Honest Cash

Reddit sucks. It's supposed to be a place for online communities to curate a feed of user submitted content, and also curate the discussion of that content. The mechanism by which that curation occurs is the upvote-downvote system where, if you have an account on reddit and you see a post that you think is relevant and useful to the community (and so deserves to be higher up in the feed), you upvote it, and if you think the opposite then you downvote it. The same goes for comments on that content, which are arranged in a tree structure with highest upvoted replies at the top of each collapsible branch.

This curation system might seem like a pretty good idea until you realize that it's easy to make multiple accounts on a site like reddit, and then manipulate both the voting system and the discussions in the comments themselves. Because it's going to be in some people's self-interest to manipulate the system by either upvoting content to the top that doesn't belong there or preventing content that does belong at the top from getting there, you can be 100% sure that it's happening. Assholes with nothing better to do are doing it, companies that want free advertising are doing it, and governments that want to sway public opinion and suppress dissent are definitely doing it.

The root of the problem, however, is arguably not that there are bad people who want to manipulate others. That's more so a given of the human condition. The problem is that we are operating within a system in which the cost of performing that manipulation is worth the benefit to the manipulator. If it were impractically difficult or expensive to manipulate the system, people either wouldn't do it, or would try and fail. I believe that the friction-less, worldwide, low-fee nature of Peer to Peer electronic cash has the potential to catalyze a revolution in online social networking, leading to systems in which manipulating online interactions for personal gain becomes much less viable.

The foundation of such a system, and the part which requires a low fee, worldwide payment system, is micropayment upvoting and downvoting. If, on a site like reddit, it were to cost half a cent to upvote or downvote a post, most users would be unaffected, doing small enough amounts of upvoting or downvoting that the cost is negligible. A company or government running a large scale astroturfing campaign, however, would have a much higher burn rate, needing to consistently, repeatedly out-vote hundreds, thousands, or even millions of individual users.

While a paid voting system will cost manipulators significant amounts of money, it might also discourage regular users from voting, in spite of the negligible cost. Distributing the vote-payments to previous voters, similar to and, provides users with the incentive to vote (potentially making them a profit on popular posts). Also, as long as all vote payments are distributed to all previous voters, regardless of whether those votes were upvotes or downvotes, such a system distributes a well-funded vote manipulator’s funds to the actual members of the community, and puts the incentives in the right place for downvoters: they still get paid from later voters, but they are making the post less likely to be seen by downvoting it.

Of course, it’s entirely possible that some especially well-funded manipulator will still be willing and able to out-fund the general voting public, especially if their entire profit model is put at risk by the healthy function of an online forum. In order to combat these well-funded meddlers, what I think is needed is a user-specific vote weight system, like is described in an article of mine on, linked in the description. The basic idea, though, is for the vote system to weigh votes from users with whom you’ve have coinciding votes in the past more highly than those with whom you’ve haven’t, or have had contradictory voting history. Such a system (with some subtle tweaks and extensions) would cost even more money to subvert, potentially infinitely more, depending on the calibration of the vote weighing.

Finally, to make sure that there are no central points of failure, the community/topic differentiation should be 100% tag based. Any time a user posts, they’ll tag their post as being relevant to any number of topics and, when one votes for or against the post, they will be actually voting for or against a given tag on that post, and will have the opportunity to vote on all other tags associated with the post.

For example, say you search the site for the tag #BitcoinCash and one of the results at the top of that tag’s feed is a post about bch that’s tagged both #Bitcoin #BitcoinCash. You could then upvote the post’s #BitcoinCash tag if you think it’s a good post, and also its #Bitcoin tag if you think it’s relevant to people searching for the tag #Bitcoin. If others upvote it on both tags, you’ll make money on having voted for both tags. If the #Bitcoin tag ends up being controversial, you’ll still make money from the downvotes it got from you helping it get higher in the first place! A tag-based system like that would be much lower maintenance than Reddit’s current moderation based subreddit system, and would be invulnerable to manipulation via squatting of moderation positions for popular topics, since there are no topic specific moderation positions.

That tag-based system, in combination with the user-specific vote weight system, could lead to communities developing naturally and organically not as subsections of the website, but instead via users using of more and more specific tags to hone in on posts that interest them, and then their votes connecting them to other users who vote in agreement.

While I think that a social network with all those features would totally kick ass, I unfortunately do not have the skills, nor the funds to make it a reality. If you also think it would kick ass, please let me know if you’d be interested in adding to a bounty for its development. If there’s enough support, maybe we can make it happen.


RE: Bitcoin and the Future of Social Networks (Video and Script)

by @emergent_reasons

I don't think this would stand up to funded social (media) manipulation campaigns. Someone doing that has a focused agenda and money behind it. I think Bitcoin Cash is still the solution though - through a web of trust based on on-chain identities (not "real" identities which cannot be verified in a deterministic way). If that is in place, it gives weight and context to the up/down votes based on relationships instead of just money. Whoever gets that with a fantastic UX can be the next king of social media.

RE: RE: Bitcoin and the Future of Social Networks (Video and Script)

by @AD1AD

Why exactly don't you think the suggestions in this post would stand up to funded manipulation campaigns? The point of the suggestions is to both increase the costs of manipulation, and diminish the influence of those who still have enough funding that the increased costs are negligible.

Also, while the whole "web of trust" idea sounds great hypothetically, I think it will be much more feasible to use the coinciding vote system (and the extensions to it I suggest in my other article) to automatically create a "web of agreement", at least within a given platform.

RE: RE: RE: Bitcoin and the Future of Social Networks (Video and Script)

by @emergent_reasons

Voting alone is not enough because the best it can do is filter out petty trolls - who are not a serious risk anyway. Organized campaigns can fund more upvotes than any individual. But when those votes are viewed from my own web of trust / web of agreement, then they have no weight. I think we agree more than disagree. I should have phrased it in a more constructive way.

RE: RE: RE: RE: Bitcoin and the Future of Social Networks (Video and Script)

by @AD1AD

>Voting alone is not enough because the best it can do is filter out petty trolls - who are not a serious risk anyway. Organized campaigns can fund more upvotes than any individual.

This ignores the main original point of the article/video… that organized campaigns would have to do more than outvote an individual. They have to outvote a whole community. This certainly should reduce even funded manipulation, because it will cost much more to do.

But yes, again, there will of course still be those who are not just funded, but well enough funded to outvote entire communities. Which is why I suggest the vote weight mechanism that has to do with a "web of vote correlation" which has the potential to automatically make it even more expensive (potentially infinitely more) to manipulate the system (and to less effect), which should further reduce manipulation, without a user ever needing to actively establish trust via a "web of trust" system, which I think is a huge benefit.

Even for a particularly well-funded entity willing to throw everything they've got at manipulating a forum, increasing the cost to do so still decreases the amount of time they'll be able to afford to do so… so I'm not sure how you could argue that a system like this would only filter out petty trolls. If anything, it's more effective against large scale campaigns (that have to do lots and lots of voting, and provide tons of information to a vote weight system) than petty trolls.

RE: RE: RE: RE: RE: Bitcoin and the Future of Social Networks (Video and Script)

by @emergent_reasons

You don't have to out-vote a whole community. Just out-manipulate it. The cost of paying people to go and do the voting is greater than any voting that has reasonable cost. Especially if you have a decentralized social network, it is virtually free to create an enormous amount of accounts. The critical factor will be that each of us has a way to perceive that those accounts are worthless. The votes alone are not enough.