I’m intrigued, but this is a bit vague. What kind of thing are you looking to build, concretely? The gestalt I’m getting is this: A social network with a transparent feed / recommendation algorithm that each user can explicitly tune, which as a side effect tags any piece of information with a trust score. This score is implicitly filtered through a particular lens / prior depending on who you trust. Looking at things through different lenses for different purposes is encouraged.
Two thoughts on this:
It seems to put a lot of administrative burden on users to keep their edges up to date. Can the system be made mostly automatic, e.g. alerting you to changes, suggesting manual review of certain nodes, adjusting weights over time based on your usage, etc? Basically, can you give a user-centric description of the idea, rather than a system-centric one?
What about strategic incentives? Sure, very blatant voting rings and such can be easily rooted out, but this system would give malicious actors a lot of information to embed themselves in a web in more subtle, insidious manners, e.g. sleeper accounts.
It’s not vague, it’s general! =). But also yes the UX is a bit vague and just incomplete. We could probably characterize it a bit more clearly by listing the kinds of queries/views/activities it supports. For some users, a thin frontend for learning and running various types of queries will be all they really want.
polling, fetching news: Show everything in tag from my presence in a web, that has been posted within date range
replying: view the replies to an article (and some of the replies to the replies) that are qualified by a particular web (this should enable every conceivable demographic to find a commentariate that they like and want to engage with, which would be pretty darn neat)
notifications: View recent replies to my articles (or articles I’ve subscribed to) from presences in the web reachable from my presence, or submission requests. Other things, probably.
chatting: (I didn’t mention this at all) seeing a realtime view that is focused on one head of a reply chain and its most recent branches. Probably eventually offer an additional view that shows other active chats you might be interested based on their topic tags and the participants involved. There should not need to be a firm distinction between asynchronous discussion venues and chat venues and I believe we can open a channel between them!
processes and views for promoting the creation of new tags, the formation of new webs: Legitimately vague, I don’t know how this should be done, I’ve mainly been asking how to enable good networks to exist and sustain themselves over time (the answers are: specificity of purpose, and accountably inheriting trust), the question of how to accelerate the formation or reshaping of the network might be a whole other component of the problem.
There are reasons to think that administrative burden might turn out to be very light, for a lot of webs:
If you trust just three people, and if they trust just three people, and so on, that could conceivably get you a very large userset with very little effort on the parts of each person. The question is, how do each of those people discover the others? How do you get into a network? Do you form a portfolio of submissions, proof of humanity, then issue join requests to the curators who know you the best? Yeah, that should probably be common pattern!
You don’t need to maintain an active presence in a web in order to make use of it. Somewhere there is a small community of people who are obsessed with vacuum cleaner reviews. To get an extremely well curated set of vacuum reviews, all you need to do is find them and include one of their central figures in a query over them.
I propose governance communities (maybe funded by the project) like metacurators to aid people in finding those communities. As an experiment, I recently tried to find “train twitter”, knowing that such a community almost certainly exists. I couldn’t. There were actually lots of results for the search “train twitter”, but all that and a whole hour weren’t enough to find me to it. Measures will probably need to be taken.
adjusting weights over time based on your usage
I like that one, that could be one of the things the above mentioned, very simple like/dislike (“relevant”/”irrelevant”) actions cause to happen.
It would be interesting to see what happens if you fully automate the formation of webs based just on users’ liking/tagging interactions, whether we’d end up with a perfectly good automatic recommender system that’s also perfectly accountable/controllable. It’s not clear to me that we would need automation for the first version, or for the types of users who’d use a first version of anything.
I’m intrigued, but this is a bit vague. What kind of thing are you looking to build, concretely? The gestalt I’m getting is this: A social network with a transparent feed / recommendation algorithm that each user can explicitly tune, which as a side effect tags any piece of information with a trust score. This score is implicitly filtered through a particular lens / prior depending on who you trust. Looking at things through different lenses for different purposes is encouraged.
Two thoughts on this:
It seems to put a lot of administrative burden on users to keep their edges up to date. Can the system be made mostly automatic, e.g. alerting you to changes, suggesting manual review of certain nodes, adjusting weights over time based on your usage, etc? Basically, can you give a user-centric description of the idea, rather than a system-centric one?
What about strategic incentives? Sure, very blatant voting rings and such can be easily rooted out, but this system would give malicious actors a lot of information to embed themselves in a web in more subtle, insidious manners, e.g. sleeper accounts.
Regarding Vagueness:
It’s not vague, it’s general! =). But also yes the UX is a bit vague and just incomplete. We could probably characterize it a bit more clearly by listing the kinds of queries/views/activities it supports. For some users, a thin frontend for learning and running various types of queries will be all they really want.
polling, fetching news: Show everything in tag from my presence in a web, that has been posted within date range
replying: view the replies to an article (and some of the replies to the replies) that are qualified by a particular web (this should enable every conceivable demographic to find a commentariate that they like and want to engage with, which would be pretty darn neat)
notifications: View recent replies to my articles (or articles I’ve subscribed to) from presences in the web reachable from my presence, or submission requests. Other things, probably.
chatting: (I didn’t mention this at all) seeing a realtime view that is focused on one head of a reply chain and its most recent branches. Probably eventually offer an additional view that shows other active chats you might be interested based on their topic tags and the participants involved. There should not need to be a firm distinction between asynchronous discussion venues and chat venues and I believe we can open a channel between them!
processes and views for promoting the creation of new tags, the formation of new webs: Legitimately vague, I don’t know how this should be done, I’ve mainly been asking how to enable good networks to exist and sustain themselves over time (the answers are: specificity of purpose, and accountably inheriting trust), the question of how to accelerate the formation or reshaping of the network might be a whole other component of the problem.
Regarding Administrative Burden:
There are reasons to think that administrative burden might turn out to be very light, for a lot of webs:
If you trust just three people, and if they trust just three people, and so on, that could conceivably get you a very large userset with very little effort on the parts of each person. The question is, how do each of those people discover the others? How do you get into a network? Do you form a portfolio of submissions, proof of humanity, then issue join requests to the curators who know you the best? Yeah, that should probably be common pattern!
You don’t need to maintain an active presence in a web in order to make use of it. Somewhere there is a small community of people who are obsessed with vacuum cleaner reviews. To get an extremely well curated set of vacuum reviews, all you need to do is find them and include one of their central figures in a query over them.
I propose governance communities (maybe funded by the project) like metacurators to aid people in finding those communities. As an experiment, I recently tried to find “train twitter”, knowing that such a community almost certainly exists. I couldn’t. There were actually lots of results for the search “train twitter”, but all that and a whole hour weren’t enough to find me to it. Measures will probably need to be taken.
I like that one, that could be one of the things the above mentioned, very simple like/dislike (“relevant”/”irrelevant”) actions cause to happen.
It would be interesting to see what happens if you fully automate the formation of webs based just on users’ liking/tagging interactions, whether we’d end up with a perfectly good automatic recommender system that’s also perfectly accountable/controllable. It’s not clear to me that we would need automation for the first version, or for the types of users who’d use a first version of anything.