Friday assorted links


1. I think so, but maybe they never were that good.

Leave it to an economics blog to make a case for the veracity of diminishing returns.

Does the quality of moderation deteriorate over time?

And is that based on enthusiasm/level of caring about the comments?

Most sites I've seen which are successful in terms of comments understand that a modicum of moderation automation prevents a ton of trolls.

On a high visibility blog (like this one), or a high visibility open commenting platform, you need some kind of moderation that goes beyond simple spam removal, also removing low quality non-spam (President Bolsanero/Brazil/Cuck accusations/etc). Otherwise, the noise and low quality posts drive out the good.

Also, the 15 minute thing is a bit silly. I'm an earlier commenter here because I stumbled into a fresh post, on a topic that interested me. But I don't have MR on auto-refresh, staring at it on 15 minute intervals throughout the day. I assume the same goes for others.

A problem with MR comments - because there are so many posts here, a given post is only the first or second post on the blog for a short time, pushing down older posts, and their comments, quickly. May I suggest a prominently linked "open thread" post once or twice a week, as SlateStarCodex does?

Agreed. Very simple automated moderation would help a lot. One person posting with multiple names is one of my pet peeves, it basically destroys credibility very fast. You could also add a quick word filter ("cuck" for instance) which could tame some trolls.

I find comments on slatestar codex high quality throughout, and given the length of some of Scott's posts, I sometimes opt to read the comments instead of the article. I rarely feel like I've made a bad decision by doing this. No offense guys, but I don't feel like MR has equally good comments.

Agreed. Slate Star Codex has an account system (which means posters can be banned) and moderation that is forgiving but vigilant. Scott Alexander understands that if you don't take active measures to preserve the walled garden, it will become overgrown with weeds or destroyed.

Marginal Revolution should do the same. Pick a few level-headed posters to be moderators.

"I accept The Power!"

I think most of Tyler's points remain true today.

That being said, MarginalRevolution has the most bare bones posting software that I see anymore. Despite this, there are still plenty of interesting and valuable posts. The points made in Tyler's/Alex's posts and the point and counter-point below are often far better than what I see written in newspapers or online articles.

An open thread combined with a ban on off topic comments.

Or just a ban on off topic comments period.

Yes we should both be immediately banned.

I wholeheartedly endorse this idea.

Maybe that has always been your purpose, to apply the heckler's veto.

But it might be useful to actually explore the difference between our positions.

I'm fine reading any conversation tangentially related to the top post. Naturally I prefer that such comments be honest, well presented, and founded in fact.

You're different, right? You are fundamentally about not seeing things that you don't want to see.

I’ll be much more inclined to believe your sincerity if you can go three days without an off topic series of comments about the POTUS.

Is that a logical statement, for a citizen in a democracy, arguably in a time of crisis?

I very much doubt the logical response is to spam an academic economics blog comments section with off-topic partisan hyperventilating.

"The McConnell-Trump impeachment game"

Published here, at Marginal Revolution, on October 27, 2019.

Not a bad piece, in retrospect.

I don't think he has anyone else to talk to.

Remember, it's not partisan, it's "moderate." Everyone else is the extreme one. Of course.

Comments about blog comment quality are themselves low quality comments.

I’ll lower you one and reduce the quality even further.

Hold my beer.

1. I've also suspected that the quality of comments goes down over time not just at the individual post level, but at the site level. For example, comparing comments on this blog from 2010 and 2020, I find the comments from 2010 on average tend to be more insightful and less trollish. My hypothesis is that, as time goes on, successful blogs tend to attract more readers, and also for a variety of reasons the marginal reader is less sophisticated than the average. Pair this with the likelihood that the least informed are most likely to comment, and the effect is diluted comment quality.

I come here because of ease of commenting and I have much too much time on my hands. Likely, that and five years "put out to pasture" are sources of my declining quality.

Re: "Assorted Links." Opinion Is Not Truth Department: Many links are assorted but (to me) some are not high-value.

I should stop reading the crazy talk comments. It's like a moth drawn to a flame. Years ago, one of my more lettered colleagues was asked why he had come to the executive floors. He responded, in ear shot of the regional exec, he was doing field research in proctology. The highly-capable can utter such things.

Completely agree - this is why comment sections tend to be of higher quality in the early going, but diminish over time. I'm referring to comment quality across posts, over time, not comment quality within a particular post.

"America's Favorite Architecture"

About what you would expect, the top 15 are well-known and highly visible monuments or buildings famous enough to essentially function as monuments in the public imagination. Not until #16 do you get a real, informed choice. I'm sort of heartened that the only Vegas architecture on the list is the Bellagio at #22. My personal favorite didn't make the list at all:

Can't help but wonder when Mar-a-Lago will be quietly added to that list.

And just think today's cruise ships, at about 300 m in length, are twice the length of the Smith Tower.

The Chrysler building at #9 must be worth a mention surely. Glad to see the Gamble house there if only at #66.

RE: #3, no doubt someone, somewhere will produce that study as compelling evidence of the virtue of autarky.

#3: The list is too Cathedral-heavy. I would ban Catholics from participating in future polls on the subject. I also wonder how many people who voted FLW's Falling Water that high have ever actually been there. It's a rather ugly house built over a pretty stretch of a stream, and it's had structural issues, basically since the construction was complete. A more appropriate name might have been Falling Ceiling. Wright's design was largely an exercise in value subtraction.

America? The survey asked fewer than 5,000 for responses. A little over half, 2,248, were AIA members.

Good one: I had to look up 'autarky.'

I worked five years in the ESB. It doesn't seem that great looking out a lower floor office window.

Fallingwater is an early entry into what is too often called "good architecture" around the world these days. I call it the "box in the forest." Basically, build a box, and let the natural beauty of its surroundings do the heavy work of actually making the building seem attractive.

Let's just ignore the fact that it only works where you can ensure a square mile of land free of other development.

For any site, the initial visitors are highly motivated to learn, and are seeking out as many sources as they can find. These are the most thoughtful readers, and likely to leave thoughtful comments, or no comments at all.

As time progresses and the site's readership grows, each marginal reader is less invested in the platform, and probably heard about it from a friend, or a link from somewhere else. They're less connected to the site and the articles, and are less likely to leave good comments.

I have a reasonably popular YouTube channel and have witnessed the same decline in comment quality over time. I've also seen an increase in the number of trolls and people just generally behaving badly.

I'll also add that Tyler and I seem to be talking about 2 different issues. Tyler is talking about the quality of comments over a relatively short period of time, per blog post.

My point is related to the decline of comment quality across blog posts, or in my case, YT videos.

"I'll also add that Tyler and I seem to be talking about 2 different issues. Tyler is talking about the quality of comments over a relatively short period of time, per blog post."

Right most of the commenters here are talking about issue 1 whereas Tyler's OP was clearly about issue 2.

I didn't read all the comments there but someone apparently mentioned diminishing marginal utility or marginal returns, which I think is the main explanation for issue 2: after any of us, including Tyler, have read 15 comments, the 16th is almost always going to be less informative, innovative, or interesting. But if we started with #16 and read backwards to #1, we'd probably find #1 to be less good on average than the other fifteen.

"For any site, the initial visitors are highly motivated to learn, and are seeking out as many sources as they can find."

I would love to see data supporting this. I watch a lot of edutanement YouTube videos (I'd rather watch a documentary about the hazards of Victorian homes than a sitcom), and often the rabbit holes I go down exploring interesting threads take me to some pretty old (in internet terms) videos. In my professional life it's worse. One of my most-frequently referenced books is from the 1930s (Fenneman's "Physiography of Western United States"), and I've cited papers so old they were written in Latin before.

I would say, in the absence of hard data, that two things are happening with fast commenters:

1) They are looking for something to comment on. This includes folks with something worth saying--but it also includes trolls, folks with knee-jerk reactions, and the like.
2) People who follow this blog will check it more often, leading to faster response times. Folks just finding the blog may not find it until long after the post has been made, and while they're unlikely to comment on a blog post a year or two old, they may post on something an hour or two old.

Those honestly engaged in pursuit of knowledge, or even entertainment, likely don't care how old a video is. And in some cases older may be better--the YouTube channel King of Random has experienced pretty significant scope creep over the past few years, for example, with the older videos being more DIY stuff and the newer being more "cool things you can do with expensive equipment" stuff.

I'm afraid I don't have data - I'm speculating.

To me blogs are like restaurants - you have the adventurous leaders who go and find the interesting content (or food), and they tell other people about it. The people who blaze the trail are more interested & invested in what the find, and are more likely to have interesting thoughts to share.

The folks who hear about it from the adventurous folks are likely less interested - otherwise, they would have gone and found it on their own.

In restaurants you can tell a lot about the quality based on the clientele. For example:

* Lots of elderly people - restaurant WAS good back in the day, but is no longer. It survives by invoking good memories in the patrons who remember what it used to be

* Lots of hipsters - this is what happens to good restaurants after the word gets out. Typically this is a sign that prices are going to go up, and quality is going to decline. This is what the adventurous diners leave and find something else.

Hope that helps.

"The folks who hear about it from the adventurous folks are likely less interested - otherwise, they would have gone and found it on their own."

I don't think this is quite right. There are multiple things going on here that get conflated into one number. For example, there's time--some people have more time (due to jobs, family, access to internet, etc) than others. Someone working two jobs to make ends meet may be very interested in economics, but not have time to search out blogs.

Second, early adopters always end up dealing with a lot of sub-par stuff. Doesn't matter if it's a blog, an electronic device, or a new scientific theory: early stages include a lot of garbage. Early adopters believe that the advantages gained from adopting new things early make up for this. Later adopters--many of whom have the same level of interest in the topic--believe that the best option is to wait for the early adopters to cull the chafe first. Both are viable strategies, and orthogonal to interest in the topic.

I'm also not convinced that the number of interesting thoughts is correlated to the concepts of early adopters or late adopters. Look at how many scientific theories are revised long after the theory is first adopted.

Not always; see for the best quality comments I've seen anywhere

To be fair, posting literally hundreds of sequential blog items (not really literally) about obscure Russian literary references in between very occasional things which are of actual interest, uh, well, it thins out the herd.

3. No love for the missions.

Maybe because they are vaguely "foreign." I love them.

Notice the almost complete absence of brutalism, International Style, etc. in the list. Frank Lloyd Wright now feels like a traditionalist, blending as he did into Craftsman and Art Deco styles.

I think an underrated quality of an architect's work is how well it may be successfully mimicked or adapted by others: essentially judging the starchitect not simply for his own work, but - in my own town, say - the more pedestrian bank building or library that owe something of their existence to him. However arresting, or even on rare occasion successful, to me Brutalism fails on this count.

I don't read that many comments; it's not as though my comments are directed to readers of the blog. A recurring criticism of comments by readers is that they are off-topic. My question: how can a Straussian blog ever have off-topic comments? How would one know they are off-topic?

LOL! You don't read comments? Perhaps you don't read other people's comments but I see you doing a CNTRL+F+"rayward"!

Ray, I do read your comments - I am interested in your much younger girlfriend. I don't know how to do a search for comments, but I do check to see if you have made comments to my comments. LOL!

Ray... and Rayward ... it is an effort for me to think guys like you, who are not pro-life like I am, who are not anti-racist like I am (think about it , the most anti-racist people on the planet are people like me who are ABSOLUTELY against abortion - liberal abortion regimes are ALWAYS racist against the less preferred race ....)... it is an effort for me to remember you have a lot of good things to say despite your defective priors ..... trust me I know what I am talking about.

Ray, your girlfriend was half you age ten years ago, if I remember correctly.
So she is now in her thirties.

Look, you may think nobody notices what you say.
I notice.
I would not want you as a son in law but if your wife loves you, or even if she does not really love you but loves you more than any other guy she has ever met, man up dude,and stop wasting your time on chess and clever comments on internet sites, and ...

and tell us all, years from now, about how right you were to want to be a dad.


"I don't like bullies"

trust me

i pity the fool who does not know i Know what i am talking about.

11:12 was a quote from Dwayne Johnson
11:55 was a quote from Mister Rogers or Mister Tea, not sure which any more

5. Interesting, I think studies of disgust should do a better job of distinguishing between disgust and fear (even the abstract cited in the Twitter page uses the term “disgust/threat”). I see disgust and fear/threat as two extremely different things. I’m disgusted by cockroaches, but I don’t feel afraid of or threatened by them. On the other hand, I feel afraid and threatened of neo-Nazis, but they don’t disgust me. Disgust and fear are very different emotions and also trigger very different responses; disgust often causes one to want to destroy the source of the disgust, whereas fear usually causes one to want to flee. For instance, if I see a disgusting cockroach, my instinct would be to stomp it but if I encountered a terrifying gangster in a dark alley, my instinct would be to run rather than fight. I don’t like terms like homophobia, xenophobia, etc. when they are used to refer to disgust rather than fear.

I think conservatives are more likely to experience disgust but conservatives and liberals are equally likely to experience fear. And note that even the paper cited states: “Even in the failed replication studies discussed above, the results tend to skew towards the hypothesized direction of the relationship, even if they are not statistically significant.” When you have several studies finding results in the same direction, that’s persuasive evidence even if each study taken individually would not be statistically significant at the conventional 5% level.

You might end up with some marginal effect, left over. But the interesting and provocative claim by the "disgust" literature* was that it was a significant factor which "explained" conservatism.

A minor difference of degree in a pretty standard, necessarily and healthy emotional response that humans have evolved doesn't really give that, and it just becomes a minor, probably not causal or very interesting, correlate of conservatism.

(Just like minor differences on rage, depression, neuroticism and so that can show up transiently in various analyses don't really cut much mustard as explanations...)

*and by this I include some affiliated models which incorporated it, like Haidt's Moral Foundations, where it forms the basis for a theoretical "Purity" moral pillar.

I've always viewed politics as a wheel, not a spectrum. Go far enough left and you will meet the far right, and vice versa, and feeling disgust is the hallmark.

2. Yes, it is. And the efforts to resuscitate it aren't going very well. In fact, it was failing before the internet clobbered hard-copy news, too.

Yankees are dedicated to condemning the censored media efforts of absolutist governments like those of China, Russia, etc. But journals in the US, rather than examine and question the activities of the various levels of coercion, parrot their line. Cops that simply do their well-paid job are painted as heroes, those that join the ranks of criminals are mentioned with little sense of outrage.

If there is one role that the news media should whole-heartedly embrace it is holding government at all levels, and the personalities involved, responsible for both success and failure. Local news organizations have become simple public relations departments for the government, Pravda with advertising.

Did I miss something? I don't think I saw local TV news mentioned anywhere in that article.

A friend worked for a local newspaper for years. It closed about a decade ago, after close to a century of reporting local news. He started a news website hoping to get something going, with little success. Few subscribers, few advertisers, enough to pay the costs but not enough to hire anyone. I talked to him recently and he is seeing a turn around. Enough to hire someone, and get some local coverage of events, which is attracting readers and advertisers. Nothing of the scale of the paper but trending in the right direction. There is a hunger for information and local advertisers want a local outlet. I thought that was good news.

5. I listened to Haidt and much made sense until he tried predicting. It is quite obvious right now; a substantial part of the US citizenry is disgusted but definitely not conservative.

Small c conservative maybe. But that doesn't translate into ideological conservatism. Lots of examples; Labour unions are conservative to the point of reactionary but not Conservative. In Canada the reaction to the Alberta oil patch is opposite; there is a visceral disgust that is open and clear among the left towards the industry that provides large numbers of well paid working class jobs.

His point that some people have a more vigorous disgust reaction is true, and of you look a bit further you likely would find that a vibrant and working community that has stood for many generations would have a balance of both. The tension is what makes it work. A working sewage and water distribution system with a vigorous artistic community.

It is a waste of time to try to pin labels on people.

4 An interesting anecdote. Someone I know has a cruise holiday planned. They got an email this week saying that if you have these symptoms, fever, cough, etc. don't come and we will refund your money. And you will be examined before you board the ship.

Of course this isn't about winnowing out those who are sick, it is about giving healthy people confidence to follow through with their holiday plans. IIRC when SARS made its appearance the cruise ship industry was seriously hurt.

"Of course this isn't about winnowing out those who are sick, it is about giving healthy people confidence to follow through with their holiday plans. "

I suspect its both. But of the two, not having a few sick people cause a disasterous cruise for thousands of other passengers, causing a potential quarantine and devaluing the brand, is probably the biggest factor.

To follow up: every cruise line in the world has to be looking at the ongoing disaster happening on the Diamond Princess in Japan and they are absolutely terrified.

"The cruise line operator Royal Caribbean has said it will refuse any passengers with Chinese, Hong Kong or Macao passports, regardless of when they were last in China."

1. MR is better than most at avoiding the "deteriorating comment quality" issue, but not close to SlateStarCodex, which shows almost no deterioration at all AFAICT.

Those SSC commenters do indeed have stamina. Could it have something generational to do with experience playing very long* video games?

*If long is the right word. With Pong, it was Think Long, Think Wrong.

SSC has the big dip in quality after Scott Alexander was harassed for allegedly being antisemitic or something. He cracked down on the comments and stopped writing about politics. Maybe he got too successful and became a target.

Or maybe I just read the comments too much and it’s all become familiar to me. I’d be interested in someone here’s opinion.

The MR comments section is usually about the right length for me, SSC for instance is too long and you can’t really follow the conversation, but of course that is better than no conversation. The other thing I like about MR comment section is the mix of regulars who are highly predictable and drop ins, often experts, who bring new insights. Of course the best thing is the hosts, who have managed to find thoughtful posts for more than a decade. Internet has no consumer surplus my arse!

+1, that's nearly exactly my opinion.

Yes, same with me. I only glance at SSC, but I don't think I've seen even a single trolling comment there (maybe it's moderated? I'm not a regular reader there).

But I read MR, not SSC where both the posts and the comments have too many words for too few insights. Which isn't to say that it's not good, it is good. But not as good as MR in signal-to-noise ratio, even with MR's comments that provide negative utility. It starts of course with Alex and especially with Tyler, but if there weren't comments here I doubt that I'd read MR.

Gosh, Tyler. Does your claim that comments after the 15th are mostly not so good mean that in the Shelf Life of Public Intellectuals thread when way far into it I noted that due to having run my department's seminar series for a long time I got to meet "a very smart and interesting polymath" and that I have not regretted having done so that this was not a worthy comment? :-)

#1....There's no doubt in my mind that this blog is about as good as it gets, and Alex and Tyler deserve some sort of award. I look forward to the blog everyday. I read every comment, and wouldn't do so if I didn't enjoy it. I've been reading and commenting since Sept of 08, a great time to have begun. I encourage people to comment and participate. The only thing that bothers me are the infantile morons whose posts disrespect this blog and its readers. One last thing...I'm glad the comments aren't voted on.

Why are you glad the comments aren't voted on? The system seems to work reasonably well for places like Slashdot. Can you share your concerns with comment voting?

Sure. I'll get a vast majority of down votes. I'll probably still comment, since I enjoy this blog so much, but I don't need a daily reminder that I'm not as clever as I think I am. As Jackson Browne wrote, don't confront me with my failings, I have not forgotten them. It's a lot easier to click on a down button than to argue with me.

Surprised to not see the National Museum of the American Indian on that list.

#6...To be honest, I'm only a Democrat now because of my dislike of Trump and McConnell. In 2016, I was for James Webb, and I still think he would be a better President than anybody running. I was disgusted by his treatment during the primaries. I've a bad feeling that the party is heading for a Corbyn. As in Britain, I think this polarization meme is obscuring how many people are in the middle/moderate camp.

If the Dems do nominate Sanders, basically the American version of Corbyn, they will not beat Trump. Sanders might even get more votes (like Clinton did), but I don't see how he can win the electoral college with the natural Republican advantage it offers.

James Carville agrees:

"“We’re losing our damn minds ... “Eighteen percent of the population controls 52 Senate seats,” Carville said. “We’ve got to be a majoritarian party. The urban core is not gonna get it done. What we need is power! Do you understand? That’s what this is about.”

Trump won something like 85% of US counties. If Democrats want to win they have to be able to expand out of their urban bubble.

"Trump won something like 85% of US counties" -- that's even less meaningful than saying Clinton won the popular vote. Trump's actual margin of victory was 100,000 votes -- his swing voters could have all fit in the Penn State football stadium -- spread across three states. Slightly broader appeal from Clinton in Pennsylvania and Wisconsin and higher black turnout in Michigan would have handed the victory to her -- and led to breathless pundit pieces about the permanent Democratic majority!

Presidential electoral maps also distort a lot of political discourse by making everything about red/blue states. If people actually paid attention to the House of Representatives and the non-famous people who make it up and where they are from, discourse might shift a bit. About 100 out of the 232 Democrats who make up the current House are not from the Northeast or the West Coast. Texas alone sent 13 Democrats to the House in 2018, Ohio sent 4, Kentucky, Utah and Kansas 1 each, etc.

Thanks JWatts. It was very interesting and timely.

Sanders could win a sizable EC victory on economic issues (healthcare, minimum wage, higher taxes on the super rich), but the momentum of the Democratic Party over the past 5+ years since Obama entered lame duck territory is continuous and uncompromising leftward push on social issues and immigration. Even if he has to just go along with it for show, it alienates a lot of moderate voters who would vote for him on economic issues to hear things about prisoners voting from their jail cells, an end to immigration enforcement, and no compromise on transgender bathroom/locker issues.

Came an article today about the "mobilization theory" (cult?) which seems to be trending among Democrats*.

Idea is apparently no or few undecided swing voters exist, in the sense of people who reliably turn out to vote, but who truly do not decide who to vote for until the last minute and are really reachable and persuadable. Instead a pool of people who are strongly partisan or at least predictable in their voting, but who have to be "mobilized" to vote exists, largely by negative partisanship. These people drop in and out of elections giving an "illusion" of a pool of relatively static size which really changes its mind. The efforts that are needed to get these people "mobilized" are then apparently thought to be exactly counter what is conventionally thought to be needed to win over a persuadable moderate centre.

Now, all sounds like a nice thing to want to believe for the left wing of the Democratic Party - they'd probably love a secret army of partisans that it turns out all along could just be motivated to pass them their vote by really vigorous attack politics on the Republicans. Dovetails so nicely with, uh, well, what they want to do anyway. However, skeptical that there really these kind of "strongly partisan yet for some reason do not reliably vote" subsets about to create an illusion of swing voting.

(One interesting aspect is apparently this is all the brainchild of a particular Niskanen employed outlier in her polling methodology, Bitecofer, who, uh, is very *narrative fulfilling* for the 2015-2020 era in a number of ways in her personal style and conduct - random swearing, "punk feminist" emblems, a self aggrandizing personal style. Wonder if that had any influence...)

4. There is clearly potential for coronavirus disasters in Africa. But on the up side:

1. Health professionals throughout the continent now have access to barrier protection (gloves, masks) even if the populace does not.

2. The incubation period is short enough for the virus to potentially burn out in many regions.

3. Poor transport infrastructure will protect many locations.

3. Africa's median age is around 20. Widespread coronavirus infection could be a massive human disaster, but its economic effect will be comparatively low, with the vast bulk of the working age population recovering even in areas where no treatment is received. Coronavirus of course has the potential to take a horrible toll on the immune compromised.

7. DC is world-leading at whipping up opposition to governments, but he failed in DoE when trying to whip up support for the government trying to make exactly this kind of big change - his boss got seriously demoted, exactly like DC, to a non-policy role as Chief Whip.

Comments for this post are closed