Combating Misinformation
about.fb.com
about.fb.com
Does this pass the mask test? As we all remember, the WHO said to not wear masks early during the outbreak. Twitter users, in particular, was way ahead on urging people TO wear a mask. Would Facebook now block that or reduce its distribution and potentially risk lives?
They are only using fact checkers because it gives them someone to point to and say "Look we didn't make this call, they did". This is like a money laundering scheme but with brand reputation risk instead of money.
Facebook will retain absolute control and the people running this initiative will do whatever the brass tells them to, this program is just for deflecting complaints.
The problem is with individuals and organizations that knowingly continue to promote misinformation like vaccines are bad for you.
What would you have Facebook do, fact check using Twitter because the CDC made this one bad call? Do you think the modal CDC consensus is worse than that of Twitter?
Really, Facebook and others need to just give up on this. There is no principled commitment to anything here. "Misinformation" isn't even a meaningful concept. It's arrogant and abusive for randos at FB or other social media firms to believe they are so smart they can sift the entire world for what's true and what's not. Nobody can do that. Nobody.
Please read the whole section because I cherry picked just one paragraph without full context. I just wanted to demonstrate that nothing is actually all that obvious when it comes to the pandemic.
> Evidence on the protective effect of mask use in community settings
> At present there is only limited and inconsistent scientific evidence to support the effectiveness of masking of healthy people in the community to prevent infection with respiratory viruses, including SARS-CoV-2 (75). A large randomized community-based trial in which 4862 healthy participants were divided into a group wearing medical/surgical masks and a control group found no difference in infection with SARS-CoV-2 (76). A recent systematic review found nine trials (of which eight were cluster-randomized controlled trials in which clusters of people, versus individuals, were randomized) comparing medical/surgical masks versus no masks to prevent the spread of viral respiratory illness. Two trials were with healthcare workers and seven in the community. The review concluded that wearing a mask may make little or no difference to the prevention of influenza-like illness (ILI) (RR 0.99, 95%CI 0.82 to 1.18) or laboratory confirmed illness (LCI) (RR 0.91, 95%CI 0.66-1.26) (44); the certainty of the evidence was low for ILI, moderate for LCI.
The truth is never easy to ascertain. In fact, one of the highest cited papers from an epidemiologist is titled « Why Most Published Research Findings Are False ». I found it to be a good meta-analysis that could easily challenge my previous field of computer vision where often fantastical results were obtained on a data set but could not be reproduced elsewhere.
https://journals.plos.org/plosmedicine/article?id=10.1371/jo...
The recent decade has been obsessed with truth as an road to ideal life.
I believe in local first reality now, relativistically so.
@BadMedicalTakes (https://twitter.com/badmedicaltakes) is just one quick way to see some of the stuff thats out there.
Can we at least all agree that germs cause disease, there are no nanobots in any of the COVID vaccines, and that masks are not dangerous?
It seems next to impossible to both allow individuals to share things and make sure that you only see "true" things in that community. And it seems that given the numbers involved, there will always be some amount of "accidental" misinformation that has a real impact, and deliberate misinformation from bad actors looking to exploit the system...
Your mistake is thinking the problem is in the communities, as opposed to the technology platforms themselves.
Personally, as a starting point I'd want them to:
1. Kill algorithmic curation that is based on "engagement". These digital skinner boxes are a cancer on society.
2. Introduce much stronger privacy laws that discourage or prevent data collection used to drive targeted advertising. Reduce or eliminate the financial incentives driving #1.
3. Provide complete and open transparency around how these platforms engage in their moderation decisions. Reporting leads one to suspect Zuck stepped into to help out right-wing agitators. Twitter is making their own arbitrary decisions. We all need to know if these decisions are happening and how they're being made.
2. This has some merit and would be a great start (we can dream, right?), but I don't think it would solve misinformation.
3. That sounds nice, but how would it work in practice. What if moderators get death threats from bad actors?
2. Nothing I listed would "solve" it. There is no one "solution".
3. We already have laws to deal with harassment.
Frankly, I feel like you're falling for the nirvana fallacy, here:
Yes, FB has been pretty egregious in certain areas, but the question of how to prevent misinformation in online communities is a tricky one, that goes beyond the details of a particular platform and into human nature. Government can (should?) help in certain ways, but even in the best scenarios, that will have limitations, as well.
Anyway, I appreciate your thoughts.
To cut the long story short, there are many people who believe that we need to have a fundamentally different approach to how we form communities. Decentralization is one of those trends in tech that could help us get there.
There's also a big difference between different perspectives and things that are blatantly wrong.
Decentralization is interesting, but 20 years ago, there were many small "social networks", even if they didn't go by that name. FB and it's portfolio are the first ones to create 1B+ member "communities". There are very strong economic incentives for smaller communities to grow into larger communities (FB illustrates the possibilities and perils of focusing on growth).
This is perhaps too big a topic to discuss on HN comments, but the idea that there are absolute & unchanging truths goes all the way back to Plato. The Western Civilization is founded on top of a very specific worldview that is "centered" on these absolute truths. This is the reason why a lot of traditionalists see postmodernism as the root of all evil.
>There's also a big difference between different perspectives and things that are blatantly wrong.
I don't disagree. What I'd like to focus on is not basic logic like 1 + 1 = 2, but rather the implications of these. That some people are born with XY or XX chromosomes is a fact. What we do with those facts is entirely context dependent.
>Decentralization is interesting... There are very strong economic incentives for smaller communities to grow into larger communities
Recent decentralization trends are designed to combat this exactly at the protocol level.
Ha ha-- so true. But thanks for making a valiant effort. ;-)
> but the idea that there are absolute & unchanging truths goes all the way back to Plato. The Western Civilization is founded on top of a very specific worldview that is "centered" on these absolute truths. This is the reason why a lot of traditionalists see postmodernism as the root of all evil.
You could argue that it predates Plato... But even before postmodernism, people seemed to have trouble agreeing on one truth.
>There's also a big difference between different perspectives and things that are blatantly wrong. >> I don't disagree. What I'd like to focus on is not basic logic like 1 + 1 = 2, but rather the implications of these. That some people are born with XY or XX chromosomes is a fact. What we do with those facts is entirely context dependent.
This is a great example of an issue that I've seen on FB. Most people do have XY or XX, but a small percentage of people have something different. Beyond that, the way those genes get expressed can lead to different phenotype outcomes, which is a scientific "fact", but I've seen a lot of people, including from folks who claim to have MDs, who say that it's just XX or XY. Is this "misinformation"? Probably depends on who you ask...
>Decentralization is interesting... There are very strong economic incentives for smaller communities to grow into larger communities >> Recent decentralization trends are designed to combat this exactly at the protocol level.
Please explain more. I don't dispute that their are notions of smaller communities, I'm just thinking that there are structural reasons why those smaller communities will get eclipsed by communities focused on growth.
Oh wait, nevermind, this is "Combating Misinformation WHILE STAYING a multibillion rev business who makes money from users' attention"
(context https://twitter.com/doctorow/status/1374785753629163521)
Oh wait. Now that is considered information and the opposite is considered misinformation.
This is so great. /s
If they meant the first article, it should be flagged [2020]
[0]: https://about.fb.com/news/2021/03/how-were-tackling-misinfor...
There is real enduring value in being a source for information that is true and trusted. It is possible that Facebook might choose that path.
The harder question is understanding what is true. "Galileo's head was on the block; the crime was looking up the truth."
Facebook (or YouTube, or Twitter, or whoever) can try to come up with even more infoboxes or disclaimers, but it will continue to exist as long as it is given a platform.
The original sin of social media comes down to the idea that anyone can post, and that virality begets virality. Shocking content is presented on the same level as traditional media -- and in many cases, can exceed those traditional news sources' reach.
FWIW, and disclosure -- we're building a platform that aims to combat misinfo by making it only for news. (https://blog.nillium.com/were-not-an-aggregator/).
Our reporters follow an Editorial Policy, that comes with consequences if they break the guidelines. Virtually every respectable news org has something like this -- we're just making ours public. https://www.forthapp.com/docs/policy.html
Until we hold the reporting produced by professional reporters -- reviewed by editors, fact checked, and held in check by an editorial process -- at a higher esteem than what Firstname Bunchanumbers says, misinformation will continue.
Having worked in a news organization, I wish I had more faith in this statement. Most reporters are woefully out of touch or depth when covering complex topics. As a finance major, I'm often surprised how often the explanations for basic topics are wrong or presented without any nuance.
That implies all ideas are given equal visibility, but we already know that social networks prioritize content that's outrageous or controversial over content that isn't.
You want a free exchange of ideas? Start by fixing the algorithms.
[1] https://www.nytimes.com/2018/10/15/technology/myanmar-facebo...