How Twitter moderated the Covid debate

If we want governmental mandates to regulate social media, let's do it. Make it law that crackpot ideas MUST be shared, and limit exactly how platforms can moderate their platform. The problem with this approach is that in the quest for freedom of speech, we'll have stripped freedom from owners of the platform to moderate their platform how they see fit.

The reality is that truly unfettered free speech & unmoderated discussions/debates devolves into all sorts of distractions from the primary goals of the platform (e.g. spam, hate speech, illegal activity, off-topic discussion). There must always be some level of moderation and censorship, even if that is at a basic level watching for spam bots & illegal activity. Typically that's not enough to ensure decorum, so you end up with additional community guidelines to set the tone for the platform.

My point here is that freedom of speech on online platforms is a mythical unicorn, that never existed, nor will it ever exist. It seems more like a wedge issue certain people like to bring up because they're unhappy with how a platform has acted towards them, and want to claim superiority by championing "free speech", even if they don't actually believe in it. For all the belly aching Musk did about Twitter, he is doing exactly the same things he complained about. That'd be fine, if he wasn't such a hypocrite about it & understood the ramifications and expectations of free speech, and the complexities of competent content moderation.

I'd also disagree that crackpot ideas are best debated in the open on social media. I think in a well moderated forum where people are held accountable, and expected to cite their sources yes. In the firehose that is social media where attention spans are limited, consumption is often on mobile devices which may hamper ability to do in-depth research, it leaves much to be desired as a place to find deeper truth easily. Also with puppet accounts, the moment you debunk one thread, ten more have popped up regurgitating the same talking points.

> Botching the methodology on the paper you’re using to argue for removing safeguards is bad but it’s worse not to mention that you were funded by one of the industries affected.

You're being too coy. Which industry and which paper? I'm not sure what's preventing you from being more specific, especially since you were lamenting vague examples earlier.

> unless you’re defining that term as meaning that you don’t get free promotion.

So if you can seemingly access a service but not be provided that service then you haven't been deplatformed? That sounds a lot like being given a telephone and a dial tone but you can only call out to prearranged numbers and meanwhile no one can call in nor find out your number and yet the phone company claims they're not denying you service. In other words, preposterous.

> Testing rate is a meaningless statistic, because the criteria for testing vary between countries and over time.

Testing is useful for two reasons - helping people who may have the disease (obviously), and for seroprevalence. Since Japan's testing was so low those seroprevalence statistics are lacking. As we know now, severity of the disease is linked to the overall health of the individual. It's quite possible those waves you call nontrivial were just as large but that the health of the population meant the severity of symptoms (which is what actually makes a wave trivial or not, along with availability of health services) was lower and continued to be lower.

I've been in Japan for the whole of the pandemic and to claim that people were isolated - either from each other or the outside world in any way that is meaningful for disease or contact tracing was of a high level is to contradict reality.

Finally, the article I shared by Bhattacharya and Packalen goes over contact tracing and points out why it is futile in the case of COVID-19, but Martin Kulldorff, Bhattacharya's infamous co-signatory on the Great Barrington Declaration sums it up in two tweets[1][2]:

> Contact tracing, testing and isolation is important against many infectious disease outbreaks, such as Ebola and post-vaccine measles. It is ineffective, naïve and counter-productive against COVID19, influenza, pre-vaccine measles, etc, and by definition, against any pandemic.

> Problems when one or more of: (1) disease is widely spread, (2) there are many mild or asymptomatic cases, (3) there is pre-symptomatic transmission, (4) we cannot identify the first index case in a given population, or how that person got infected.

[1] https://twitter.com/MartinKulldorff/status/13073778096192880...

[2] https://twitter.com/MartinKulldorff/status/13073812979171614...

> No I trust experts in the field who by and large say CSAM is WAY WAY WAY down on the platform since Elon took over

Which experts and how is this being measured? Because I can't find sources that confirm that claim: https://www.forbes.com/sites/alexandralevine/2022/11/18/elon...

Do you have ANY actual numbers to back that up, at all? Let alone numbers measured by independent experts and not just filtered through Twitter PR?

> I care about the amount of content, not the number of moderators.

I don't know how to get across to you that human moderation requires humans.

The act of moderation requires a human being to sit down, look at a piece of content, and decide if it's in violation of a policy. Then that process is repeated over and over again for the duration of the workday. Having bodies in the chairs to look at the content is not optional.

Particularly not where CSAM is concerned, which is an area where moderator turnover is astronomical because moderating CSAM is effectively the equivalent of feeding your brain poison every day until you have a breakdown and have to go to therapy. It is a nasty, horrible job and there really isn't a way around the fact that it requires a lot of moderators working in rotations. There's a limit to how often someone can be shown horrifying images in a workweek before they decide that they would rather be doing anything at all other than continuing to show up at work.

That part of the moderation job isn't glamorous, and it usually happens behind the scenes with contractors. It's an awful part of the tech industry that doesn't get much attention because it's unpleasant to talk about and unpleasant to acknowledge how much we rely on underpaid contractors ruining their own lives to keep our platforms safe. But the gist is that an engineer doing "hardcore programming" is not a substitute for having thousands and thousands of moderators actually reviewing images.

----

> Just look at the tweets from Ella Irwin

That Elon's current head of safety is praising him isn't surprising -- one thing we've all learned from the way he's run Twitter is that you don't get to work for Elon at all unless you're pretty proactive about praising him.

But there have been zero independent measurements that confirm these claims, and actual experts in the field (read, not just a current employee and a couple fan/activist accounts) are concerned about the direction the platform is headed and about the lack of investment in moderation teams.

But since those experts are talking to journalists and not just posting to Twitter, maybe it doesn't matter what they think /s.

"Conspiracy theories are looking for ways to become something that fits people's predispositions." Maybe that is exactly yours predisposition?