TRUSTLINES for social: Another wacky idea!

[thinlyspread]

In the spirit of fun recently and because we're all trying to revive interest in Ye Olde XRP Ledger DEX, here's another totally-not-at-all-thought-through idea I had once during my usual morning coffee thinkathon. I'm never going to develop it, so putting it out there because you never know...

Could XRPL trustlines be utilised somehow as part of a social network, either as a way to find mutual connections for specific things/skills, or, a way to filter out certain connections? 

For example:

• Twitter has a bots/spam/trolls problem. Just like an adblocker uses blocklists on a separate layer i.e. plugin for the browser, so too could XRPL be used as a kind of public database of good/bad filters for content and/or accounts? Whitelists/blacklists etc. 
• What about if you want to find someone to ask them something (kind of like Quora) and can even pay for expert opinion/advice etc. You can't always trust reviews as they can be gamed, or, you just don't trust the word of strangers... BUT... maybe if you had a shared trustline between one person (or a group) you trusted, and their trusted network, you could sort of "pathfind" someone... a bit like a reputation system, basically. "Recommend a friend" but all automated online.
• Time sharing. This has actually been discussed before. In fact I believe a few people created tokens ("issuances") years back on the XRPL, that you could buy on the DEX and exchange for their time. E.g. 1 THSP = 1hr of thinlyspread's time. Etc. 
• Could trustlines be used as a sort of alternative mechanism for admin rights on shared apps/databases/forums etc? Because you could have distributed public/private voting on rights, where the trust gets spread out among partipants and the more trust you build, the more access you get and vice versa. In this case, you not only extend a trustline, but add the amount on a scale you trust them, so you can actually set a quantity not just quality of trust.

GOING FOR MORE COFFEE.

[thinlyspread]

In fact, riffing off this last part a little more...

One problem I see with social database/content links like Twitter or LinkedIn, is that you can follow someone as a binary 1/0, but you can't really define that in detail, there's no granularity. 

Trust in general is like this. For example, let's say I trust @brianwalden and @KarmaCoverage quite a bit because they're Old Hats at the XRP game. That's not the same as trusting a good friend or family (personal), or my local plumber (skillset), or whatever. But in their field (as I perceive it), let's say I trust their judgments on certain things and their character generally (e.g. to not troll or be nasty). You could then extend this to social media content at large, for example. So instead of "I follow her, she follows him, and so I follow him by extension" methods (Tweet suggestions probably do this), that doesn't tell you much about the quality or extent of the relationships. But if you could "rank" these connections, that adds a new layer. 

I *believe* SM companies try to quantify relationships using smart algorithms based on mutual tags, shares, follows, content types, etc. But there's no free/open and accessible general trust system I know of... although actually I didn't really research it. And if there is, surely not on a blockchain! And if there is, surely not one as good as the XRPL!  

You get my general point. I might extend a trustline, like a "follow" on Twitter. But what if I set that to a scale of 0–5? Maybe you can have a social graph that has granularity. You can have certain fields/types etc, or just make it very general. And it's a public database, maybe it's only there to blacklist or greylist rather than whitelist. Or maybe it's a great new tool to EXPORT your trust relationships across networks. So the XRPL trustlines work as a kind of bridging tool between data networks.

I can go on and on with so many ideas here, but I'll leave it here because I doubt many will be interested. 

ONWARDS! 

Edited March 17, 2021 by thinlyspread

[thinlyspread]

OK quick extra bit – electoral voting systems today. They're archaic. Organisations like FairVote are trying to "upgrade" to systems like ranked choice, instant-runoff, and so forth. Is that a use case too? Trustlines can be thought of as living, evolving "votes", and trustline networks almost like delegations of voting power or confidence. Since ranked choice also required a quantitative relationship (the ranking of candidates), this works well with trustlines because you can set the trustline (qualitative) then set the trust amount (quantitative). 

[brianwalden]

I'm thinking of something like reddit, each subreddit will be identified by its own token. Let's say we're in the Decentralized Ledger Technology subreddit because its token can easily be abbreviated by DLT. When you upvote someone in the subforum, you increase your trustline to them by for DLT by one. Downvoting decreases the trustline by one. No coins are actually transferred, you're just adjusting trustlines.

I think this will allow you to do interesting things. For example, if @JoelKatz is in the forum, he will have recorded a ton of upvotes on his various comments. If he upvotes you, all the accumulated trust he's received from the subreddit could give his upvotes more weight.

The one downside of using trustlines is that reach one locks up 5 XRP. Honestly, I don't think you need a decentralized ledger for this unless you want to set up some type of cross platform trust system.

[KarmaCoverage]

5 hours ago, thinlyspread said:

Could XRPL trustlines be utilised somehow as part of a social network, either as a way to find mutual connections

This is "Rippling" 

5 hours ago, thinlyspread said:

for specific things/skills, or, a way to filter out certain connections? 

I think the IOUs per skill makes sense. Also Quality in/out on those IOUs could be a "measure" of the "real world quality" of that trustline.

4 hours ago, thinlyspread said:

SM companies try to quantify relationships using smart algorithms based on mutual tags, shares, follows, content types, etc.

Yes all that stuff and way more data points. Complexity is the type of math you use to manage a Network (like XRPL & RippleNet & an IoV | google, fbook, twit, yada) https://en.wikipedia.org/wiki/Social_complexity , https://en.wikipedia.org/wiki/Systems_theory  https://www.youtube.com/watch?v=HMBl0ttu-Ow

Your gut is pointing you in the correct direction. Trustlines are in fact the thing that makes the KarmaCoverage methods possible, and the whole purpose of my original interest was the "near zero marginal cost" of processing a transaction. The financial model behind KarmaCoverage depends on lots of small/micro transactions, hence why XRPL was an attention getter. Then the Trustlines make everything socially easy to stitch together into a "Trust Graph" which would be the actual asset that KarmaCoverage would build/assemble. Just like fbook assembled the "social graph", and linkedIn the "colleague graph", and google the "link graph", etc. 

I explained to one of Ripple's very early investors during a few phone calls the concept of a Trust Graph (asking him to keep it to himself, he didnt on at an investor conference in Europe a few weeks later), and I basically told him that what I saw Ripple building (what is now RippleNet) was an "intuitional trust graph", and that KarmaCoverage was designed to build the Trust Graph for individuals. At the time 2014ish the idea was that every user would have a wallet, and a UI would enable the end user to extend Trust to their friends they are willing to Help if a loss occurs.

Now it think it makes much more sense to have the UI sit on to of a Smart Contract, which has 1 wallet on XRPL per Risk Type, and keeps the Trust Graph either centralized to start, or in a side XRPL so as not to clog up the real XRPL with wallets and trustlines and ultimately end user costs (Reserve fees). I guess it could start on XRPL, then run in parallel a centralized KC sidechain (we run all the validators), then take that decentralized just like Ripple did. The KC ledger would always sit on top of the real XRPL for final money movement, and just the little stuff would occur on it's own Risk ledger (which would not need it's own XRP, seignorage is not the business model, but it could be one of them).

Anyway, the main point is that Trustlines and Orderbooks are what make XRPL so far superior to BTC. Bitcoind just simply does not and never will have the functionality necessary to build something like a Trust Graph... and if you are building a network company you need to understand the "graph" you are building. BTC is just a network of servers a "server/miner graph", I'd bet that 99% of the holders dont run a node, never did an on ledger TX, and run everything through a dreaded "CENTERALIZED" exchange  where the govt can freeze your account.  If any one remembers the up roar made when the Freeze feature was added to XRPL, youd remember how "awful" the idea was. Just "Awful enough" to be compatible with the real world, that is IMHO.

4 hours ago, thinlyspread said:

OK quick extra bit – electoral voting systems today.

This reminded me of a "use case contest" that Ripple had back in the very beginning. One of the dozen or so pitches was a Voting system. I think XRPL is way too powerful of a system than what voting needs. Obviously it needs to be on a federated & distributed system, but a voting network really dosent need to do 90% of what XRPL can do.

I went looking for the video on youtube without luck, but it may be buried in some of the old videos that searching this finds... "Ripple Developer Conference use 2013". I'm going to have to rewatch some of these for fun. I know most of the core concepts have never changed, but the nouns used have gone through about 5 up grades. 

 

 

 

Edited March 17, 2021 by KarmaCoverage

[RacerX]

I like the idea but the 5 XRP per trust line can get pretty expensive as the price continues to rise.  Even the minimum amount of 20 XRP will be too steep eventually for new wallet holders if the price reach even $100 per XRP.

 

Right now, I am trying to build a more user friendly XRP Tool kit that allows seamless XRP trading and other cryptos using the oeer to peer decentralizes trading.  Then expand to a system that buys real estate assets using the liquidity pools created through buying PIUS tokens more like a country club where you have free access to these assets, e.g. apartments, Hotels, BNBs or whatever assets are bought.  Prior to assets being purchased each member user has a vote on what next to buy.  The Oracle would be providing what real estate assets are available to buy.  Any thoughts?