Hi there, this is Keith from Factland. You probably haven’t heard of us but we’re a small team that’s been working on decentralized crowdsourced fact-checking for the past two years.
We’ve been lucky enough to get a grant from DFINITY to integrate as much of our protocol as possible to the IC. Admittedly we’ve spent more time in EVM communities so we’re hoping to make some connections here and be more transparent about our progress and roadmap.
We think we can be one of the flagship projects for the IC and DFINITY, but we can’t do it without community support and engagement. Please say hi here or on our socials! Also feel free to ask any questions.
Thanks! We’ve been slow to launch a token since we’re in the US, but that will be our focus for the next few months. We’ll be posting our ideas for tokenomics here and hope to have some good brainstorming with the community.
I saw this project a few months back and found it extremely interesting. Not sure who all is on Twitter here but it is safe to say that Community Notes on Twitter has changed the game on disinformation and context needed.
The project sounded awesome until I went on and the very first claim I saw that was deemed “True” by the “3rd party fact checkers”
In no way am I trying to tie politics into this. But this is clearly a “false” claim. It takes just about 0 time to check independent facts and see that this is indeed false.
What steps are being implemented to make sure this platform is not simply analyzed by people of all the same mindset? Going through the claims now there is a very clear bias on the answers to the posted claims.
Like I said- great project and awesome idea! Thank you for your time.
Great question. We’re working right now to implement completely random jury selection in the IC Canister itself. Our other engineer is working on the random mechanism and I forget how it works exactly. It’ll all be open source though.
We don’t have a wide enough user base to have it be meaningful at the moment, but we’re excited to start experimenting. If you want to know more about the thinking behind this, sortition is a good keyword to look up. Also the Irish Citizen’s Assembly is a great example of how this method can deliver results that are more representative of a group’s true preferences.
Really cool project - do you have plans to tie this onto other projects where the Factland voting system could be present inside threads on a site like DSCVR for example?
Yeah, we still get really weird results sometimes just based on timing and being a completely open platform. This is partly why we’re calling it a ‘demo’ and trying to get more community engagement. The model doesn’t work with a small or disengaged user base, which is what we have now, so we’re not surprised at this point by some rulings that are clearly wrong. In most cases it only takes three jurors to rule a certain way. If we expand that number, rulings never happen, but at this number they have biased outcomes. We’re excited to get more people on the platform so we can increase the number or required juror rulings.
Let us know if you think we need to be stronger about telling people why they might see some weird results. We’re working on appeals as well so that when you come across something like this there’s a clear path to overturn it.
Yes! I’ve been researching and planning how to make an embed code for a claim so that all the action can take place on other platforms. Kind of like embedded tweets. But interactive.
That sounds great - you could have real time checking, and the results for new readers, as it happens. Would be pretty amazing. Good luck and hope the IC can really help out with the project.
Using the Factland Dapp, people can upload claims for community review and comment, and stake FACT tokens to vote on whether they are true or false. Cases are decided by anonymous juries drawn randomly from the community to weigh the evidence
and issue a verdict.
I’d say that the focus is on information which can ideally be truthfully verified using a slice of the ‘wisdom of the crowd’. If being used to combat what should be called ‘timely’ disinformation (things like news stories which possess a shorter information lifetime), then I think sounds like an excellent delivery mechanism.
Hi everyone thanks for weighing in and the words of support, I’m also on the Factland team as a nontechnical contributor, my background is in journalism, former News dot com department editor, WIRED dot com EIC, Medium editorial director, Periscope EIC, Twitter Curation Director (very familiar with Notes! fun fact 90+ % of Notes never get published due to the requirements - resulting in high quality but low coverage.) IRT “bad” takes by the community, we envision an appeals process where once a stake has been decided it can be re-opened by people who disagree with the result. Would that resolve some tensions around this or just open another can of worms? Curious to hear what you think. Our goal has been MVP all the way and keeping the door open for substantial community input vs pushing prescription solutions on the group.
