IRC - feem and I discuss decentralization/p2p/distributed/centralized glitch art

12:07 ednapiranha so i am of the belief that centralized systems work with distributed services (like in revisit.link) because you are able to maintain a community, usability and discoverability in a consistent manner

12:07 feem discoverability not just of disparate nodes but also of users

12:07 feem i mean on a personal level

12:08 feem i think a lot of p2p software forgets there are people

12:08 feem peered machines with things to share but without the idea that there are people behind those things and that the people guide the things

12:08 ednapiranha i just think the current models of p2p and decentralized systems are for certain users and can’t break that wall for new people who just want to get something working right away

12:08 feem one of the great emergent bits of napster et al was curation

12:08 ednapiranha same as the crypto problem

12:08 feem well yeah

12:08 feem sharing a hash isn’t exactly approachable

12:09 feem plug in this magic string and all your dreams come true

12:09 feem it’s not difficult but there’s also no understanding of what’s happening or why

12:09 ednapiranha in the current model of revisit.link, i see it as a shift in how we have people learn and experiment with interactions dependent on each other

12:10 ednapiranha but as a completely decentralized model, i can’t imagine how you could easily find others in a user-friendly manner, nor have consistent interactions

12:10 ednapiranha i would in theory love the idea of getting to that point, but endpoints aren’t always reliable. hence why some services are ‘offline’ on the central hub

12:10 feem and that’s what makes it cool; because it’s a serialized stack of transforms with each bit of logic done by someone else and multiple users involved: the person submitting and controlling the stream, the people making the haphazardly-interlocking services, and the group of people receiving the message

12:11 feem the other problem with transitory endpoints and intermediaries with streamed data and inline logic is that you lose the ability to order that process as an end user

12:11 feem either the intermediary or the user control gets lost

12:12 feem centralization, in this particular instance, actually increases the amount of interaction in the system

12:12 ednapiranha but do you think it could be fully p2p if we redesigned elements of how that model works?

12:12 ednapiranha it seems like p2p / decentralization works really well for certain things but not necessarily for others

12:12 feem i’m not sure; we would have to overcome the fact that decentralization necessarily removes one viewpoint from the conversation

12:13 feem i mean, speaking just in terms of glitch, let’s think of it like group photoshops or chain stories

12:13 feem there has to be some method of serialization even if that serialization is random

12:13 feem and there has to be a trackback

12:13 feem a way for the person who got the ball rolling to be able to see what happens next

12:13 feem or else there isn’t a payoff for being involved

12:13 feem so if we go to bitcoin there, we have transaction chains

12:14 feem i’ve said before i don’t think that blockchain is as much a good method of handling currency as it is as handling data

12:14 feem especially if we’re talking about a multicast peer to peer network of logic blocks and transforms

12:14 feem multicast in terms of methodology, not necessarily networking

12:15 feem so if we’re willing to accept that control is lost, i think a blockchain approach would allow trackbacks and spidering endpoints

12:15 feem but obviously that isn’t desirable for everyone

12:17 ednapiranha i feel like the creation of revisit.link is a way for us to open that conversation. maybe to figure out a way to make the ideal better

12:17 ednapiranha because right now the ideal is so steampunk

12:17 ednapiranha it’s just a mess and i don’t honestly think people are content with its current state, but don’t know what to do about a better state

12:17 feem years and years and years ago, one of the first massive collaborative content generators i knew of was something called addventure

12:18 feem it was a choose your own adventure game where any unlinked page was created by whomever clicked into it

12:18 feem and they had their own set of branching pages

12:18 feem you could freely decide whether your was an ‘end’ or if it continued on to something else

12:18 feem and then you could either write the next pages yourself or have someone else do it

12:19 feem i think the strongest thing about revisit.link is that despite being centralized, it does open that conversation: it gives you a spec and an exchange format

12:19 ednapiranha there is also another part of this project that is interesting. the lack of authentication and identity

12:19 feem that was interesting in meatspace, too

12:20 ednapiranha and how that affects and/or enables certain motivations and behaviors

12:20 ednapiranha but it doesn’t end up being just a stream of penises

12:20 feem there’s a big difference between being anonymous in a stream of consciousness and being anonymous in a fixed system

12:20 ednapiranha producing an image on revisit is a silo for the user - it does get published on the real time feed but it’s not enough of a marker for any real attention

12:20 feem 4chan et al are, try as they might, fixed systems

12:21 feem so despite having that factor of being anonymous, it’s a meaningless thing because it’s still feasible – even if not necessarily possible – to track back that data

12:22 feem so while trackbacks are important for payoff in a shared system, they’re inherently tied to ego

12:22 feem revisit is cool because like you said

12:22 feem it is a silo

12:22 feem it’s one person’s silo of other peoples’ transforms

12:22 feem it’s a single container of ego

12:23 feem and again this goes back to meatspace

12:23 feem it becomes a transaction

12:23 feem whose capital is based upon the end user to assemble

12:23 feem rather than a website to structure

12:24 feem i think an important question is: what kind of transaction do we care about? do we care about a transaction that anyone can bank, or a transaction that only one person can bank that everyone can see, or a transaction that anyone can see but no one can bank?

12:25 feem there’s no equity in a tweet

12:25 feem you can retweet or you can reply

12:25 feem but there’s very little investment and favoriting it is a cop-out

12:25 feem i think the action of clicking ‘like’ or ‘favorite’ actually reduces the enjoyment and incorporation of an event into the self

12:25 feem like spending your whole vacation doing nothing but taking photographs