One of the more interesting conversations I had at Hivefest in Mexico was with Dan from @blocktrades about PeerVerity, the trust and knowledge system that they are gradually building behind the scenes. More and more people have been asking me or talking about the kind of system that Dan is creating here, so I thought to take a moment to share some of what I know about it so far...
I found the PeerVerity website by chance shortly before travelling to Mexico and Dan was surprised that I had seen it, since they haven't advertised it or shared it yet - the content on it is very basic at this point, so that's understandable. Dan did leaves some breadcrumbs for us to follow, in the form of a few posts on the topic on Hive a couple of years ago - This first one is an introduction to the concepts and motivations and this second post talks more about data modelling. He did say then that there would be later posts on the topic, but I'm not seeing any in their post history yet.
What is PeerVerity?
In brief summary, PeerVerity is an open source protocol (and probably an app) for sharing/creating data that is used when rating knowledge and people, enabling us to build a network of people and to note who we ourselves trust on which topics - also allowing evidence to be stored to demonstrate people's expertise. All of this information can then be voluntarily shared with others, while maintaining appropriate data privacy.
Dan himself said that some people have said this sounds like a 'social credit scoring system' of the kind that the Chinese Communist state has introduced, but while a few of the features cover similar areas of life as would be found in a social credit system, PeerVerity absolutely doesn't operate in the same way. PeerVerity is a system that anyone can run on their own computing device in a decentralised way, that enables them to create their own trust system that can remain totally private if they like. Perhaps we can think of it as our own personal trust scoring system, but how we use it is up to us alone.
With that out of the way, the details of this system seem to offer us a way to improve on and even solve some of the main problems we face today online and in life generally. The mass of information that is created daily on every topic imaginable is difficult to navigate, simply due to it's scale. On top of this, we have many 'vested interests' trying to direct our experience of it, whether that be through 'editing' the content before it gets to us (up to and including outright censorship), or trying to bias/influence us through advertising, buying out 'influencers' to 'whisper in our ears' or other such nefarious methods. The 'influence' market is only growing and yet we have little in the way of new tools to be able to deploy in our attempt to remain confident that the information we receive/use is accurate/neutral.
In a world where the number of possible topics, their scope and also their content is vast and ever expanding, we simply do not have the time or energy to study them all. As a result, society 'suggests' (oh so subtly) that we rely on 'trusted experts' who we are assured have put in the long hours to study a topic in depth and who are somehow magically unbiased and have our best interests at heart.
Obviously, by now, most of us have had experiences where we know that this is not always the case. The 'experts' are often either misleading, speaking half truths, are corrupt, are in denial of key points or not even remotely experts at all! One answer to all of this is for us to use tools which score and track people's expertise, allowing us to share the benefits of our own experiences - easily recommending to others when we have found someone who we ourselves class as a true expert.
For example, I might have a plumber in my house to fix a difficult problem and he might fail badly. I might mark that in my trust app. I might then find another plumber that my friend on the trust network recommends and find that he solves the problem easily, so I give him a good score too. Over time, we will have a complex network and lots of data regarding how each of us relates to others in the community - plus be able to track expertise too. I'm sure you can see how this could turn out to be very useful.
Google My Business, Trustpilot and other such apps attempt something similar to this in order to generate 'social proof', which is a marketing phrase that means 'the community likes this service, so you should too'. However, these solutions are centralised and open to exploitation. There are many people in the world whose job it is to leave false negative reviews on these sites in order to falsely boost a competitor who is paying them! Decentralisation can solve this.
Although a 'trustless trust' system sounds like the kind of paradoxic zen riddle that definitely should be turned into an app, I think Peer Verity is not exactly going to be trustless, in the sense that connections between app instances will need to be authorised by each user based on their existing trust for the other user they share information with. None the less, I'm sure there will be ample need for inventing new terminology to deal with some of the idiosyncrasies involved!
Ultimately, how this system turns out will depend on who develops with it and what they make from the code, plus how end users integrate it into their lives. There are many possibilities here. I can easily imagine a world where all news stories are automatically run through a system that informs me of the insights about them that others that I trust have written online.. but then that also highlights other people that I don't know who disagree, why they disagree, their credentials and relevant sources of information. While this might sound like 'fact checking' on steroids, it is really quite different.
Fact checking 'services' are centralised groups who are paid to 'discern truth' and then to proclaim what truth is. In contrast, the PeerVerity system is a tool that anyone can make use of to build their own fact checking mechanisms, similar to the ones we are already creating internally in our own minds - but faster and using more information, automatically!
My regular readers might know that I am opposed to most forms of AI because I know they will detract away from our own internal development, we will rely on them and while we may get stronger in some areas, we will get weaker in others - which may then become too damaging to recover from. This principle does apply to all areas of life where we can use computers to replicate our own internal faculties and processes - however, there is still a sweet spot where technology can be helpful, without being inevitably harmful. I imagine PeerVerity to have the potential to be just such a system.
Improving on X Community Notes
In my last post on Hive I pointed out that the 'community notes' feature on X/Twitter is now being used to demonetise users (who already find it very hard to earn a penny anyway!). It's a no brainer that Hive is infinitely better than X when it comes to monetisation and even to discernment of truth, but this new peer to peer system could really take things to a new level.
Now, instead of just blindly trusting that 'the community' has created a 'fact check' of a post for you - you could perhaps instead do a deep dive into the background and knowledge base of every commenter, find out who trusts them and why - then reach your own conclusions about why any given comment is right or wrong. You could then be rewarded for adding your comment and explaining why you hold your position, referencing the data stored in Peer Verity along the way.
Summation
I can only offer a pencil outline of what this system is and will become, we'll need to look out for future posts from @blocktrades on the topic, but since their website states the end of 2023 as a possible launch date - we might be seeing just that materialise on Hive very soon.
I don't yet know just how well this system will work out, but I can easily see how it can be yet another example of why decentralisation is so necessary and how it often destroys centralised versions of solutions for the same problem. While PeerVerity is not designed to be part of Hive and is it's own standalone project, it can absolutely be integrated in to existing Hive projects in whichever way dApp creators choose. So expect to see some interesting and cutting edge implementations in the Hive ecosystem in the nearish future.. the most stunning in the whole of the 'sooniverse'! lol
Final point: The only way that centralised corporate interests can prevent the world from making use of truly decentralised solutions is through censorship, preventing the ideas from spreading - so it's no surprise that censorship is at an all time high and even now, few people have even heard of Hive on web 2 sites! It's time to change that now!
Wishing you well,
Ura Soul
Read My User Guide for Hive Here
Powerful insights into the Hive blockchain are available at my website, Hive Alive.
Including the only way to track downvotes on Hive - The Untrending report
I recall having a conversation about Dan's intention to build this at HiveFest in Thailand. Very cool to see some progress. There are definite advantages to a decentralised approach. Review brigading - whether positive or negative is one. The disconnect between audience and critic scores for movies.... Even standards of evidence for fact checking.
Across publishing, standards of evidence have become practically random, IMHO. I was agog with disbelief when I realized staid and respectable scientific journals were blatantly misstating false information as factual, destroying in some cases a century of hard won integrity in a single paper. Being able to track adherence, or failure to do so, to standards of evidence may go far to enable trust in certain resources that have seen their integrity degraded merely because they are in fields where others have abandoned it. I'm thinking of BMJ, for example, that has not abandoned it's integrity where JAMA and NEJM did, which most people may not be aware of because they have not read these journals and discerned these lapses and adherence to standards, but just discount the entire edifice of science publishing, and even science itself.
It's interesting you bring up academic journals - that level of academia operates on peer review! However, anonymity in the peer review process makes auditing and transparency difficult. The other thing about academia is that different disciplines have different standards for evidence. Each community of discourse establishes their own and I think there's value in those differences that a p2p system could express.
This is true. There are also means of assigning reviewers to papers, and other mechanisms that reduce the integrity of the peer review process.
Thanks!
Reviews of #movies are already very popular on Hive. At one point, we considered creating a combination of object types, such as movies, actors, studios, etc., to lay the groundwork for an IMDB-like data structure on Hive.
However, Waivio has currently prioritized products for social shopping, though interest in movies remains high.
When reviews and user ranking votes are published on the open blockchain, it becomes possible to find like-minded users and discover new movies based on their recommendations.
Hi! This reminds me of what taskmaster is building. Have you seen this post?
https://peakd.com/hive-167922/@taskmaster4450/bringing-rotton-tomatoes-to-hive
Thanks for the link. I agree; building an open, decentralized, wiki-like database of movies has a lot of potential. The transparency of voting and ranking on the open social blockchain can bring about profound changes in the movie industry, improving the process of movie discovery along the way.
Hey @grampo! I'll look into it. Does waivio have a peer-trust-reputation system for movie reviews?
Each object type has its own set of rankings. For example, restaurants are rated based on Ambiance, Service, Food, and Value:
For movies, there could be its own set of rankings, such as Acting, Direction, Plot, etc.
Each vote is recorded on the blockchain and signed by the user. As a result, it is easy to trace the votes of people I follow (trust) and compile a list of recommendations for discovering movies to watch.
Alternatively, it is possible to find like-minded individuals if their votes tend to match your own, and you can compile a list of movie suggestions based on their votes for movies which you have not yet watched (ranked).
With each vote a user casts, the system will be able to refine suggestions even more effectively. We are talking not just about finding similar movies, but rather about identifying films that have been positively rated by like-minded individuals.
Thanks for the detailed response. That's your solid structured data layer ;)
I foresee some of the resulting networks, if public enough, being a mirror of linkedin, with people in certain industrial/academic circles just parroting each other and rejecting everyone else. lol.. At least to begin with, though that might change over time.
I imagine the majority of data will be kept private though, no-one wants gangs of unknown thugs turning up at their doorstep to convince them to change their stated public position on particular topics!
A big difference in a p2p system is that you can eventually tune out the sources you don't trust.
Absolutely, yes, though it can get complicated quickly!
Great post, thank you for explaining Peer Verity. I heard about it at HiveFest but am much closer to "getting it" now.
Thank you for posting about this! I had this idea when I was trying to figure out how to vote in my local election. I knew some people who I respect their opinion and some I opposed, so I'd like to know who they would endorse. We do this with newspaper endorsement, but I wanted to have this more fine-tune. I imagined various pundits posting their reviews of candidates where I can start out with a weighted score on who I'd probably want to vote for, and then be able to drill down into why those I respected endorsed or why those I disagreed with endorsed.
I fumed this all into an email to a local radio host at one time. I had no answer but this has been on my mind for a while since then. Great to have all these avenues for sifting through reviews of various things via your confidence of users reviewing them! Now, how does this actually work? PeerVerity's website does not explain. I see those first two posts you linked (Thank you again!) and I agree that is how it would work. I hope we get that third post about how to do it. Otherwise, just start tinkering ourselves, I suppose.
You are welcome! At this point @blocktrades is the best source on this, we will have to wait for comments from them to update on the shape this is taking. I imagine it will evolve significantly as more people use the new protocol. Ultimately, the form it takes will to some extent depend on the users and the apps that use the protocol, similar to the way that web browsers have evolved on top of http.
This would be very interesting to use to rate crypto "influencers" in a decentralize db so ppl can know who to avoid and the score system can be verified, just a use case I can think of
How about adding a ‘favorites’ section to Hive profiles? This way, everyone can 'heart' businesses they like, just as they can with products today.
And when I browse maps or business directories on Hive, my friends' recommendations (users I follow) will be highlighted. It’s like getting an insider’s tip on the coolest spots in town!
Peakd already has private favourites and I think Ecency does too now.
I used to like Youtube's feature of letting me promote specific channels on my profile page, which is similar to what you are asking for I think.
As I understand it, the PeerVerity protocol will make it possible to not only favourite a profile but to add detailed feedback and scoring of different aspects of the person and their profile too.
For example, one of my favorite Japanese restaurants is Ichiro Japanese Restaurant. Information about this business is already available on the Hive blockchain, and I can reference it in my posts. I can even follow them and receive Telegram notifications when other users post reviews about it.
But imagine if there were a heart button that would allow me to feature this restaurant (or any other business) in my profile, just as you mentioned regarding promoting channels on YouTube.
I can foresee an overlay over Google Maps or other such business maps that filters the restaurants according to whatever parameters you like, drawn from your network of friends.
"Show me the Japanese restaurants that John Smith likes in this City.. Or would probably like best, even though he hasn't been here" :)
Exactly. In fact, Dining.Gifts does exactly that - it displays all restaurants whose information is published on the Hive blockchain on a map.
Yes, you can filter restaurants by the type of food they serve and the amenities they offer. You can even search the map to find out which nearby restaurant serves a particular dish.
All of the data, including user reviews and sponsored rewards offered by restaurants in exchange for sharing two or more photos of their dishes, is publicly available on the Hive blockchain.
Thank you for the link to PeerV. I shall be taking a while to understand it and see how I can contribute and maybe build something for myself and others. Appreciated.
Not only With Hive but integration of rss, too please
Combination of these open protocols can be fucking unstoppable
This I will find useful, as I am often flying on gut checks in the hurly burly of comment sections, based on prior knowledge that may be decades out of date. I have had to retract my comments, and I HATE being wrong, so this causes me distress. While this is humbling, and therefore good for me, I am suspicious I am too comfortable with that distress, and I would very much appreciate means of better certifying what I have confidence in, so that distress remains extremely uncomfortable, and I never, ever become complacent about being wrong about important things.
It is also very beneficial to enable understanding of specialized knowledge. For example Scott Ritter is a former US Marine, Weapons Inspector, and military analyst, but I have no reason to expect his remarks on butter manufacture to be expert. However, his proven veracity and panache of authority in fields where he has expertise could easily lead someone to expect anything he said about butter could be taken as gospel, which doesn't necessarily follow. Being able to note such expertise, and propensity for some folks to comment outside their realms of experience, can help greatly to appropriately view such interests and expectations.
Thanks!
I can foresee that an advanced dApp running this protocol could essentially replace the default view of Wikipedia, Youtube and other such sites - with them now only being databases for the main app to make use of in it's profiles and network models. This is another example of a true web 3 concept coming to life.
Seems like the "Rep" value has taking a life on it's own.