Easy entrance is how September happened, both on LessWrong and on Usenet.
My personal bias here is that I see little hope for most of the application level network protocols built in the 80s and 90s, but have high hope for future federated protocols. Urbit in particular since a certain subtribe of the LW diaspora will already be moving there as soon as it’s ready.
I’m pretty sure the problem isn’t primarily technical—it’s not that Usenet mechanisms or protocols stopped working, it’s that the interesting conversations moved elsewhere. Sure, a woeful security model (trivial forgery, unauthenticated moderation headers) helped it along, but the fundamental community tension (it’s not possible to be inclusive and high quality for very long) is what killed it.
LessWrong is actually pretty good in terms of keeping the noise down. There are a few trolls, and a fair number of not-well-thought-out comments (case in point: what you’re reading now), but they’re not enough to drown out quality if it were still here. Where we’re failing is in attracting interesting deep thoughts from people willing to expand and discuss those thoughts here.
My analysis saw the fundamental problem as the yearning for consensus. What was signal? What was noise? Who was trolling? Designers of forum software go wrong when they believe that these are good, one place questions with actual one place answers. The software is designed in the hope that its operation will yield these answers.
My suggestion, Outer Circle got discussed on Hacker News under the title Saving forums from themselves with shared hierarchical white lists and I managed to flesh out the ideas a little.
Then my frail health got even worse and I never did anything more :-(
Not sure about details, but the general idea seems right to me. My thoughts on the topic are usually something like: “How is it possible that in real life we can filter the good stuff much easier than online? I guess because in real life we can use strategies X, Y, Z, but there are not digital equivalents of them in online systems. We cannot use our usual strategies online, because the corresponding button is simply not there.” In real life:
different people see different content, because they use different sources of content
people show interesting stuff to their friends
people have different personas for different friends
sometimes a friend sees more than one persona; sometimes we hide a persona from some people
sometimes we agree to talk only about a specific topic for a while
Okay, I probably missed a few important things. But this is already difficult to do on many websites.
For example, I miss the “persona” feature on Facebook. Having multiple accounts is discouraged. There is an option to post something that only a selected group of friends can read, but that is not what I want. Sometimes I want to post an article that anyone can read, but which only appears by default only on walls of some of my friends.
The most obvious example: different languages. There is no point to spam my English-speaking friends’ walls with comments written in Slovak. On the other hand, if they decide to view them and use google translate, why not? It’s not like I want to keep something secret; I just predict that with high enough probablity they won’t care, so I don’t want to bother them. Also, I want to keep those comments accessible to Slovak-speaking people who are not in my contacts.
If I understand it correctly, Facebook only gives me two options: public (which will push the message on everyone’s wall) or private (which will hide the message from everyone except a few hand-picked people), and neither is what I want. This would be easy if I could just have two personas, one for each language, and anyone in my contact list would have an option to follow just one of them, if they want.
Similarly, I could have personas for “private life”, “politics”, “rationality”. My relatives probably want to see the photos of my baby, but don’t care about my opinions on Bayes Theorem. For other contacts, it may be the other way round. Sometimes the personas intersect (a post could be about politics and in Slovak language; or perhaps a political comment on local affairs that are uninteresting for a foreigner). Sometimes they don’t apply (a photo of a baby is language-independed).
So perhaps these “personas” could be just some predefined flags, applied to any content I make, in any combination. And my friends could specify that they are interested in some personas and uninterested in others. Access to some personas could be limited.
...but this is obviously far from the complete proposal.
Also, the whole interface must be very simple, especially for people who don’t give a fuck about the sophisticated features. There must always be a “default” setting that the Average Joe can use; otherwise the Average Joe will complain about the difficult software and will not use it, which hurts the value of the whole network.
Not protocols. High-level structure of a BBS/mailing list/forum/Twitter/etc. Protocols (in the technical sense) provide some constraints on what kind of structures can be built on their basis, but there are enough degrees of freedom to construct very different things on top of the same protocols.
Otherwise, Eliezer could have posted his Sequences on 4chan.
So the difference between LW and 4chan is protocols..? X-)
And now that I actually write it down and compare it to previous online communities (including a few mixed online/offline) I’ve been part of and loved, and which have universally followed the same pattern of growth, overgrowth, loss of some driving valuable members without obvious replacement, slow decay into irrelevance (to me; at least 2 of them are going strong, just with a different feel than when I was involved)), I’m pretty pessimistic.
I’m going to put some effort into being OK with LW as it is, enjoying the parts I enjoy and being willing to follow those parts I’m missing to their new homes.
This fits my own prior experience of the life cycle of a community—but when my previous community failed, a fragment of it broke off and rebuilt itself in a few form. That fragment still exists as a coherent tribe more than a decade later, and I still love it even if I disagree with certain, uh, technical decisions surrounding the splintering process.
Oh, indeed—fragments or even whole (slightly altered) communities live on. Two of my prior identity-tied groups are still meeting and going strong, they’re just not producing original research or even super-deep discussions on their topics. I still have fond feelings toward them, but I don’t participate enough to consider them part of my identity.
This is primarily a reminder to myself that this is okay. I can enjoy LW for what it is rather than lamenting what it was.
Easy entrance is how September happened, both on LessWrong and on Usenet.
My personal bias here is that I see little hope for most of the application level network protocols built in the 80s and 90s, but have high hope for future federated protocols. Urbit in particular since a certain subtribe of the LW diaspora will already be moving there as soon as it’s ready.
I’m pretty sure the problem isn’t primarily technical—it’s not that Usenet mechanisms or protocols stopped working, it’s that the interesting conversations moved elsewhere. Sure, a woeful security model (trivial forgery, unauthenticated moderation headers) helped it along, but the fundamental community tension (it’s not possible to be inclusive and high quality for very long) is what killed it.
LessWrong is actually pretty good in terms of keeping the noise down. There are a few trolls, and a fair number of not-well-thought-out comments (case in point: what you’re reading now), but they’re not enough to drown out quality if it were still here. Where we’re failing is in attracting interesting deep thoughts from people willing to expand and discuss those thoughts here.
My analysis saw the fundamental problem as the yearning for consensus. What was signal? What was noise? Who was trolling? Designers of forum software go wrong when they believe that these are good, one place questions with actual one place answers. The software is designed in the hope that its operation will yield these answers.
My suggestion, Outer Circle got discussed on Hacker News under the title Saving forums from themselves with shared hierarchical white lists and I managed to flesh out the ideas a little.
Then my frail health got even worse and I never did anything more :-(
Not sure about details, but the general idea seems right to me. My thoughts on the topic are usually something like: “How is it possible that in real life we can filter the good stuff much easier than online? I guess because in real life we can use strategies X, Y, Z, but there are not digital equivalents of them in online systems. We cannot use our usual strategies online, because the corresponding button is simply not there.” In real life:
different people see different content, because they use different sources of content
people show interesting stuff to their friends
people have different personas for different friends
sometimes a friend sees more than one persona; sometimes we hide a persona from some people
sometimes we agree to talk only about a specific topic for a while
Okay, I probably missed a few important things. But this is already difficult to do on many websites.
For example, I miss the “persona” feature on Facebook. Having multiple accounts is discouraged. There is an option to post something that only a selected group of friends can read, but that is not what I want. Sometimes I want to post an article that anyone can read, but which only appears by default only on walls of some of my friends.
The most obvious example: different languages. There is no point to spam my English-speaking friends’ walls with comments written in Slovak. On the other hand, if they decide to view them and use google translate, why not? It’s not like I want to keep something secret; I just predict that with high enough probablity they won’t care, so I don’t want to bother them. Also, I want to keep those comments accessible to Slovak-speaking people who are not in my contacts.
If I understand it correctly, Facebook only gives me two options: public (which will push the message on everyone’s wall) or private (which will hide the message from everyone except a few hand-picked people), and neither is what I want. This would be easy if I could just have two personas, one for each language, and anyone in my contact list would have an option to follow just one of them, if they want.
Similarly, I could have personas for “private life”, “politics”, “rationality”. My relatives probably want to see the photos of my baby, but don’t care about my opinions on Bayes Theorem. For other contacts, it may be the other way round. Sometimes the personas intersect (a post could be about politics and in Slovak language; or perhaps a political comment on local affairs that are uninteresting for a foreigner). Sometimes they don’t apply (a photo of a baby is language-independed).
So perhaps these “personas” could be just some predefined flags, applied to any content I make, in any combination. And my friends could specify that they are interested in some personas and uninterested in others. Access to some personas could be limited.
...but this is obviously far from the complete proposal.
Also, the whole interface must be very simple, especially for people who don’t give a fuck about the sophisticated features. There must always be a “default” setting that the Average Joe can use; otherwise the Average Joe will complain about the difficult software and will not use it, which hurts the value of the whole network.
That is an excellent and thought-provoking essay, and a novel approach.
...I actually don’t have more to say about it, but I thought you’d like to know that someone read it.
And I second this. Short, readable and intriguing.
Yep. That is THE problem that LW has to solve.
Notice how it doesn’t care about which protocols are used to shuffle which bits back and forth.
Protocols have an impact on discussion, and discussion has an impact on what articles people write.
Otherwise, Eliezer could have posted his Sequences on 4chan.
Not protocols. High-level structure of a BBS/mailing list/forum/Twitter/etc. Protocols (in the technical sense) provide some constraints on what kind of structures can be built on their basis, but there are enough degrees of freedom to construct very different things on top of the same protocols.
So the difference between LW and 4chan is protocols..? X-)
And now that I actually write it down and compare it to previous online communities (including a few mixed online/offline) I’ve been part of and loved, and which have universally followed the same pattern of growth, overgrowth, loss of some driving valuable members without obvious replacement, slow decay into irrelevance (to me; at least 2 of them are going strong, just with a different feel than when I was involved)), I’m pretty pessimistic.
I’m going to put some effort into being OK with LW as it is, enjoying the parts I enjoy and being willing to follow those parts I’m missing to their new homes.
This fits my own prior experience of the life cycle of a community—but when my previous community failed, a fragment of it broke off and rebuilt itself in a few form. That fragment still exists as a coherent tribe more than a decade later, and I still love it even if I disagree with certain, uh, technical decisions surrounding the splintering process.
So it’s not impossible.
Oh, indeed—fragments or even whole (slightly altered) communities live on. Two of my prior identity-tied groups are still meeting and going strong, they’re just not producing original research or even super-deep discussions on their topics. I still have fond feelings toward them, but I don’t participate enough to consider them part of my identity.
This is primarily a reminder to myself that this is okay. I can enjoy LW for what it is rather than lamenting what it was.