Back in the stone ages, I believe the Extropian list had extensive configurable collaborative filtering mechanisms. I didn’t use them much, but that seems to me the actual solution. Let people trust who they want, and follow who they want. I see a Karma Score configured by me.
The failures of old mailing lists and Usenet were why social mediums universally abandoned killfiles and similar filtering mechanisms: the balance of costs was all wrong—a large number of people had to take affirmative action to ignore the small number of bad apples. It turned out to be better to actively curate the default than to thrust the burden of filtering signal from noise onto each and every user.
To give an Extropian-list-specific example: determined harassment was why Nick Szabo stopped posting there. The filters didn’t help there.
No; a lot of the materials are now private, I don’t think Nick wants to drag old stuff up, and if the harasser was the same Detweiler dude who did some later harassing, he may well have been mentally ill and not really responsible for his actions.
Thanks! I guess the main thing I wanted to check was that you meant Nick was the one being harassed rather than the other way round, which you have indeed answered.
The failures of old mailing lists and Usenet were why social mediums universally abandoned killfiles and similar filtering mechanisms:
Evidence? Aren’t such filters still available in Usenet readers? My theory is that such code was just never implemented in the shiny new web.
And with collaborative filtering, everyone doesn’t need to make every adjustment themselves. That’s the point. You delegate ratings to others, or combinations of others.
But is plopping someone in an ignore file supposed to be so difficult? Should be easier than ever. Have a plonk button on every post to add the guy to your kill file. “Hmmm, this guy is a dick. Plonk.” Couldn’t be easier. Just as easy as clicking a point of karma.
To give an Extropian-list-specific example: determined harassment was why Nick Szabo stopped posting there. The filters didn’t help there.
What was the nature of the harassment, and how would it be prevented in the current list software?
Usenet fails, therefore killfiles suck? I still don’t see evidence.
Usenet’s failure is often attributed to the defaulting to allowing everyone and expecting users to killfile their way to a good experience, which doesn’t work for keeping communities vibrant or dealing with spam. Hence, the decline of Usenet as alternatives opened up and Usenet failed to scale to Internet access getting wider.
Collaborative filtering is about the only way to scale.
Or tons of moderation and voting. Seems to work for Reddit.
No more strain or vigilance necessary than a click. I don’t find that so taxing.
Trivial inconvenience.
How is that an indictment of collaborative filtering or killfiles? Yeah, they can’t solve all problems.
The question is whether they solve any problems. If they’re so great, why are they so rare?
Usenet’s failure is often attributed to the defaulting to allowing everyone and expecting users to killfile their way to a good experience, which doesn’t work for keeping communities vibrant or dealing with spam. Hence, the decline of Usenet as alternatives opened up and Usenet failed to scale to Internet access getting wider.
Got a source? Having previously pretty much lived on Usenet and now not having fired up a newsreader in years—while frequenting reunions of two Usenet groups I used to be on, one on Facebook and one on G+ - I’m interested in anything written on the subject; I think it’s one there’s not enough well-written post-mortems of.
I don’t think killfiles were a significant factor myself, but I admit I’m basing that opinion just on “it sounds wrong”, not any actual data.
I’d have attributed the decline of Usenet and mailing lists to (1) not being on the Web (that’s the biggie) (2) barrier to entry to create a new discussion forum (even alt.* had process). Mostly (1) - the wine-users list (for Wine, the Windows compatibility layer for Linux) has a two-way gateway to a web forum, and immediately the forum was available the volume was 10x.
I also posted some hypothesising as to why there are no good Web-based Usenet readers—and why forums aren’t backed by NNTP—here, with a bunch of people I met on Usenet commenting. tl;dr that the unit of NNTP is the message, but the unit of forums is the thread. Same applies to mailing lists, which is why GMane seems weird considered as a “forum”.
Not really. This is my own lived experience comparing Usenet to Google Groups, Reddit, web forums, and Wikipedia, and noting the explosion of user-contribution in the shift from Overcoming Bias to LessWrong. You could easily prove Usenet is declined, but I’m not sure what research you could do to prove that the incentives were structured wrong or that features like killfiles fostered complacency & reluctance to change, other than to note how all of Usenet’s replacements were strikingly different from it in similar ways.
I don’t think killfiles were a significant factor myself, but I admit I’m basing that opinion just on “it sounds wrong”, not any actual data.
My read is that killfiles were a major aspect of systematically bad design of Usenet which made it uncompetitive and unscalable: it increased user costs it should not have, adding friction and trivial inconveniences. Killfiles express a fundamental contempt for user time: if there are 100 readers and 1 spammer, it should not take 100 reader actions to deal with the 1 spammers, as killfiles inherently tilt matters. What would be much better is if 10 readers take an action like downvoting and spare the remaining 90. Rinse and repeat. What is better, dealing with spam/trolls while using O(1) or O(n) in reader time?
I’d have attributed the decline of Usenet and mailing lists to (1) not being on the Web (that’s the biggie) (2) barrier to entry to create a new discussion forum (even alt.* had process). Mostly (1) - the wine-users list (for Wine, the Windows compatibility layer for Linux) has a two-way gateway to a web forum, and immediately the forum was available the volume was 10x.
The non-Web thing is another example of this. Yes, an uber-nerd (and buybuydandavis is exemplifying this attitude in this thread) may contemptuously look at it as an irrelevant problem: ‘what sort of person can’t maintain a good killfile? or figure out how to deal with NNTP servers and ports and local clients?’ But it’s a big deal when repeatedly incurred by millions of people who do not wish to become uber-nerds and to whom costs matter.
Of course, all of this could have been fixed. But they weren’t fixed in time, and so Usenet stagnated and died.
Another experience here from a long-time former user of Usenet, overlapping yours to some extent.
comp.sources.* was made obsolete by the web and cheap disc space. The binaries newsgroups also, except for legally questionable content that no-one wanted the exposure of personally hosting. (I understand the binaries groups still play this role to some extent.)
I dropped sci.logic and sci.math years before I dropped Usenet altogether, and for the same reason that if I was looking today for discussion on such topics, I wouldn’t look there. There’s only so long you can go on skipping past the same old arguments over whether 0.999… equals 1.
rec.arts.sf.* took a big hit when LiveJournal was invented. Many of its prominent posters left to start their own blogs. Rasf carried on for years after that, but it never really recovered to its earlier level, and slowly dwindled year by year. Some rasf stalwarts mocked those who left, accusing them of wanting their own little fiefdom where they could censor opposing viewpoints. They spoke as if this was a Bad Thing. It’s certainly a different thing from Usenet, but if you want a place on the net for pleasant conversation among friends, a blog under your own control is the way to have that. Rasf was that, for many of its members, for many years, but blogs do it better.
Usenet was never designed, it just grew. There were various bodies and people involved with managing it, but they generally played King Log, leaving it up to the users to manage the creation of newsgroups and stamping the resulting consensus. Kill files didn’t come from a design team, they were invented one day by Larry Wall), and taken up by everyone because they saw what a brilliant idea it was. That everyone had to manage their own kill file was, from the point of view of what Usenet was, a virtue, not a flaw. Everyone could speak, no matter what they had to say, but no-one had to listen. The libertarian ideal of free speech. I say this not particularly to defend it, but just to say that that is how people saw these things, that was the animating spirit of Usenet.
Then spam was invented, eternal September began, blogging developed, and mass public access arrived. Usenet managed to respond to all of those things, but it couldn’t change what it fundamentally was, because what it was was what those who loved Usenet wanted it to be.
Of course, all of this could have been fixed. But they weren’t fixed in time, and so Usenet stagnated and died.
Here I disagree. Usenet could not and cannot be fixed, any more than we could have brontosauruses roaming around the modern world. Usenet was a creature of the technology of its time and the spirit of its participants. There may be some lessons to learn from the history of Usenet, or some ideas worth taking up, but in the present world there is no place for Usenet.
As I recall, at least the parts of usenet where I hung out (rec.art.sf.written, fandom, and composition, and soc.support.fat-acceptance) weren’t that badly plagued by spam (there were volunteers dealing with spam for usenet), but trolls were a problem.
OB allowed users to send in emails and they would be posted, which is not a high bar (lower than, say, learning a Usenet reader) and a fair number of people contributed. It’s just that LW made it much easier and unsurprisingly got way more contributions. This apparently came as a big surprise to Eliezer (but not me, because of my long experience with Wikipedia; it was a bit of a Nupedia vs Wikipedia scenario to my eyes).
I like ignore buttons. Cleans out the crap very quickly. And provides useful feedback to people joining lists who want to talk to people. As grown up after grown up plonks you, those who might get the message do.
The failures of old mailing lists and Usenet were why social mediums universally abandoned killfiles and similar filtering mechanisms: the balance of costs was all wrong—a large number of people had to take affirmative action to ignore the small number of bad apples.
No, I don’t think that’s true. You’re arguing that internet user interfaces become better at hosting debates over time. If I believed that, I’d also believe that the user interfaces for holding rational discussion have gradually improved, from Usenet, to bulletin boards, to Facebook and Wordpress, to Twitter and Tumblr.
You’re arguing that internet user interfaces become better at hosting debates over time.
No, I’m not. I’m saying the interfaces got better at certain features of UX, like dealing with spam and trolls. Usenet could be intrinsically better at debate (in the hypothetical universe where it had a restricted userbase and wasn’t dying of spam and other issues).
eg. imagine a forum where all comments had to be accompanied by an argument map but the forum didn’t have any way of banning/deleting accounts. I have little doubt that the debates would be of higher quality, since argument maps have been shown repeatedly to help, but would anyone use that forum for very long? I have much doubt.
The failures of old mailing lists and Usenet were why social mediums universally abandoned killfiles and similar filtering mechanisms: the balance of costs was all wrong—a large number of people had to take affirmative action to ignore the small number of bad apples. It turned out to be better to actively curate the default than to thrust the burden of filtering signal from noise onto each and every user.
To give an Extropian-list-specific example: determined harassment was why Nick Szabo stopped posting there. The filters didn’t help there.
I’m curious: Can you tell me/link me more please?
No; a lot of the materials are now private, I don’t think Nick wants to drag old stuff up, and if the harasser was the same Detweiler dude who did some later harassing, he may well have been mentally ill and not really responsible for his actions.
Thanks! I guess the main thing I wanted to check was that you meant Nick was the one being harassed rather than the other way round, which you have indeed answered.
Evidence? Aren’t such filters still available in Usenet readers? My theory is that such code was just never implemented in the shiny new web.
And with collaborative filtering, everyone doesn’t need to make every adjustment themselves. That’s the point. You delegate ratings to others, or combinations of others.
But is plopping someone in an ignore file supposed to be so difficult? Should be easier than ever. Have a plonk button on every post to add the guy to your kill file. “Hmmm, this guy is a dick. Plonk.” Couldn’t be easier. Just as easy as clicking a point of karma.
What was the nature of the harassment, and how would it be prevented in the current list software?
I didn’t specify ‘failure of Usenet readers’. I specified failure of Usenet.
Still a serious UI burden which doesn’t scale. Torture vs dust specks.
It’s difficult in the way that constant strain and vigilance is so difficult. Trivial inconveniences on every post.
By flat-out banning the harasser.
Usenet fails, therefore killfiles suck? I still don’t see evidence.
Collaborative filtering is about the only way to scale.
No more strain or vigilance necessary than a click. I don’t find that so taxing.
Ok, so the current list software is no better. How is that an indictment of collaborative filtering or killfiles? Yeah, they can’t solve all problems.
Usenet’s failure is often attributed to the defaulting to allowing everyone and expecting users to killfile their way to a good experience, which doesn’t work for keeping communities vibrant or dealing with spam. Hence, the decline of Usenet as alternatives opened up and Usenet failed to scale to Internet access getting wider.
Or tons of moderation and voting. Seems to work for Reddit.
Trivial inconvenience.
The question is whether they solve any problems. If they’re so great, why are they so rare?
Got a source? Having previously pretty much lived on Usenet and now not having fired up a newsreader in years—while frequenting reunions of two Usenet groups I used to be on, one on Facebook and one on G+ - I’m interested in anything written on the subject; I think it’s one there’s not enough well-written post-mortems of.
I don’t think killfiles were a significant factor myself, but I admit I’m basing that opinion just on “it sounds wrong”, not any actual data.
I’d have attributed the decline of Usenet and mailing lists to (1) not being on the Web (that’s the biggie) (2) barrier to entry to create a new discussion forum (even alt.* had process). Mostly (1) - the wine-users list (for Wine, the Windows compatibility layer for Linux) has a two-way gateway to a web forum, and immediately the forum was available the volume was 10x.
I also posted some hypothesising as to why there are no good Web-based Usenet readers—and why forums aren’t backed by NNTP—here, with a bunch of people I met on Usenet commenting. tl;dr that the unit of NNTP is the message, but the unit of forums is the thread. Same applies to mailing lists, which is why GMane seems weird considered as a “forum”.
Not really. This is my own lived experience comparing Usenet to Google Groups, Reddit, web forums, and Wikipedia, and noting the explosion of user-contribution in the shift from Overcoming Bias to LessWrong. You could easily prove Usenet is declined, but I’m not sure what research you could do to prove that the incentives were structured wrong or that features like killfiles fostered complacency & reluctance to change, other than to note how all of Usenet’s replacements were strikingly different from it in similar ways.
My read is that killfiles were a major aspect of systematically bad design of Usenet which made it uncompetitive and unscalable: it increased user costs it should not have, adding friction and trivial inconveniences. Killfiles express a fundamental contempt for user time: if there are 100 readers and 1 spammer, it should not take 100 reader actions to deal with the 1 spammers, as killfiles inherently tilt matters. What would be much better is if 10 readers take an action like downvoting and spare the remaining 90. Rinse and repeat. What is better, dealing with spam/trolls while using O(1) or O(n) in reader time?
The non-Web thing is another example of this. Yes, an uber-nerd (and buybuydandavis is exemplifying this attitude in this thread) may contemptuously look at it as an irrelevant problem: ‘what sort of person can’t maintain a good killfile? or figure out how to deal with NNTP servers and ports and local clients?’ But it’s a big deal when repeatedly incurred by millions of people who do not wish to become uber-nerds and to whom costs matter.
Of course, all of this could have been fixed. But they weren’t fixed in time, and so Usenet stagnated and died.
Another experience here from a long-time former user of Usenet, overlapping yours to some extent.
comp.sources.* was made obsolete by the web and cheap disc space. The binaries newsgroups also, except for legally questionable content that no-one wanted the exposure of personally hosting. (I understand the binaries groups still play this role to some extent.)
I dropped sci.logic and sci.math years before I dropped Usenet altogether, and for the same reason that if I was looking today for discussion on such topics, I wouldn’t look there. There’s only so long you can go on skipping past the same old arguments over whether 0.999… equals 1.
rec.arts.sf.* took a big hit when LiveJournal was invented. Many of its prominent posters left to start their own blogs. Rasf carried on for years after that, but it never really recovered to its earlier level, and slowly dwindled year by year. Some rasf stalwarts mocked those who left, accusing them of wanting their own little fiefdom where they could censor opposing viewpoints. They spoke as if this was a Bad Thing. It’s certainly a different thing from Usenet, but if you want a place on the net for pleasant conversation among friends, a blog under your own control is the way to have that. Rasf was that, for many of its members, for many years, but blogs do it better.
Usenet was never designed, it just grew. There were various bodies and people involved with managing it, but they generally played King Log, leaving it up to the users to manage the creation of newsgroups and stamping the resulting consensus. Kill files didn’t come from a design team, they were invented one day by Larry Wall), and taken up by everyone because they saw what a brilliant idea it was. That everyone had to manage their own kill file was, from the point of view of what Usenet was, a virtue, not a flaw. Everyone could speak, no matter what they had to say, but no-one had to listen. The libertarian ideal of free speech. I say this not particularly to defend it, but just to say that that is how people saw these things, that was the animating spirit of Usenet.
Then spam was invented, eternal September began, blogging developed, and mass public access arrived. Usenet managed to respond to all of those things, but it couldn’t change what it fundamentally was, because what it was was what those who loved Usenet wanted it to be.
Here I disagree. Usenet could not and cannot be fixed, any more than we could have brontosauruses roaming around the modern world. Usenet was a creature of the technology of its time and the spirit of its participants. There may be some lessons to learn from the history of Usenet, or some ideas worth taking up, but in the present world there is no place for Usenet.
As I recall, at least the parts of usenet where I hung out (rec.art.sf.written, fandom, and composition, and soc.support.fat-acceptance) weren’t that badly plagued by spam (there were volunteers dealing with spam for usenet), but trolls were a problem.
I think it has more to do with the fact that Overcoming Bias didn’t allow users to post.
OB allowed users to send in emails and they would be posted, which is not a high bar (lower than, say, learning a Usenet reader) and a fair number of people contributed. It’s just that LW made it much easier and unsurprisingly got way more contributions. This apparently came as a big surprise to Eliezer (but not me, because of my long experience with Wikipedia; it was a bit of a Nupedia vs Wikipedia scenario to my eyes).
vBulletin which is very popular has “ignore” mechanism: put a user on ignore and you don’t see his posts. Yep, it’s just as easy as pressing a button.
I like ignore buttons. Cleans out the crap very quickly. And provides useful feedback to people joining lists who want to talk to people. As grown up after grown up plonks you, those who might get the message do.
Most ignore functions send no information to the ignored. No one ever gets the message because no message is sent.
If I’m engaged with someone, I tend to plonk publicly, so the fellow knows I won’t be responding any longer, and others get the idea as well.
But I’ll silently too.
No, I don’t think that’s true. You’re arguing that internet user interfaces become better at hosting debates over time. If I believed that, I’d also believe that the user interfaces for holding rational discussion have gradually improved, from Usenet, to bulletin boards, to Facebook and Wordpress, to Twitter and Tumblr.
No, I’m not. I’m saying the interfaces got better at certain features of UX, like dealing with spam and trolls. Usenet could be intrinsically better at debate (in the hypothetical universe where it had a restricted userbase and wasn’t dying of spam and other issues).
eg. imagine a forum where all comments had to be accompanied by an argument map but the forum didn’t have any way of banning/deleting accounts. I have little doubt that the debates would be of higher quality, since argument maps have been shown repeatedly to help, but would anyone use that forum for very long? I have much doubt.