Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

[flagged]



I find fact checkers quite interesting in the way they've been deployed over the last few years. As if one source is the "truth" and that settles things. Usually when you get to the end of "false" claims, you find some tidbits that make you wonder... such as:

"Several posts circulating online wrongly suggest that the pills are the same because ivermectin also acts like a protease inhibitor and keeps the virus from replicating. Ivermectin continues to be studied in relation to COVID-19, however, it has yet to be proven that it can treat COVID-19."


> Usually when you get to the end of "false" claims, you find some tidbits that make you wonder... such as:

Could you please explain what exactly did you interpreted from a statement that boils down to "no one showed that this actually works"? The fact check looks pretty cut and dry to me.


The problem with "fact checkers" is you can design them any way you want to get the answer you want. Here's how this is done with this one:

Yes, the claim that Pfizer's new pill is "just repackaged Ivermectin" is certainly verifiably false, so the second line of the article can claim false followed by a supporting expert quote to drive it home. Statistically something like 95% of readers will stop there. Case closed. They're convinced.

But it's not hard to find lots of front-line doctors who are seeing success with Ivermectin. Even the 4th-to-last paragraph in the factcheck article (that I cited) admits that Ivermectin "acts like a protease inhibitor" which is how the new patented "covid pills" will also act.

If you follow many of these voices like I do, you see patterns that they aren't really even claiming that the new drug is exactly like Ivermectin. So even framing the question like this is a failure and doesn't capture the actually-useful voices out there. Probably just the Twitter mobs which always seem like the majority these days, but are far from it.

This doesn't prove anything, nor am I claiming that Ivermectin works, but advancing the conversation is so much more nuanced than framing something and then calling it false. We, dear Hacker News readers, should know this better than most.


> But it's not hard to find lots of front-line doctors who are seeing success with Ivermectin.

If that was true then how do you explain the fact that so far no study was able observe or reproduce that effect under a controlled environment?


Technically not correct. Most blinded, randomized controlled trials showed improvements in symptoms and timeline [1,2,3, among others], but they are generally underpowered and so many consider the results to be not compelling. All the headlines you may have read that ivermectin studies found no results are way overblow; science journalism has never been great to begin with, and now politics has thoroughly infected this so signal to noise on this question is terrible. Some better trials are underway now, so we'll see.

Not surprising, the patentable drug got a decent trial since pharma actually has a much bigger financial incentive there.

[1] https://ejmo.org/pdf/A%20Comparative%20Study%20on%20Ivermect...

[2] https://journals.sagepub.com/doi/pdf/10.1177/030006052110135...

[3] https://www.sciencedirect.com/science/article/pii/S258953702...


It is not hard to find lots of front-line doctors who have success with sugar pills (see: homeopathy). This means absolutely nothing.

People always seem to have trouble with statistics when stuff is not digital-like clear cut. In many areas (e.g. anything involving social studies), your noise levels are huge. The fact that you hear "some success stories" usually means that there is absolutely nothing, rather than indicative of something. When there is interest (and thus publication bias), the noise level actually increases.

For something in medicine to be actually plausibly effective, what you need to hear is more like "few failure stories". "Some success stories" doesn't even start to cut it. Our current research/publish system just cannot be used to identify treatments that have low effectiveness (which may actually be a good thing). We went over this topic not long ago in HN.


Not op, but "We don't know" != "False"

Also, op does not claim "no one showed this actually works." Your comment is not in good faith.


The top-line claim in the linked fact-check is "Pfizer’s new pill to treat COVID-19 is just ivermectin repackaged".

The claim is explicitly countered by claims in the fact check: "They are dramatically different molecules. The drugs are different in their structure and their molecular size".

I have not fact-checked the fact check, but to suggest that the headline claim of "false" is misleading or inconsistent with the body of the article is simply wrong.

The comment about Ivermectin still being studied and not yet known to be effective is not relevant to the claim under review, but merely added to provide the reader with context.


Thanks for the response. You're right, but, I'm not trying to say that the headline claim of "false" is misleading with the body of the article.

I'm trying to say something more nuanced... that the headline claim of "false" is misleading from the actual conversations happening out there right now. Which in a way is not a problem with just fact checkers, but the Internet as a whole.

BUT... In my opinion, "fact checkers" by nature of their name have a greater responsibility to Internet discourse than the random comment thread or opinion article because they claim to be an authoritative source.

I can't help but think of the Ministry of Truth.


So basically someone proved their position thoroughly with regards to this singular claim, but you’re still skeptical towards the fact checkers because …? Kinda seems like the fact checkers were right here, and yet you still persist for whatever reason.


Why keep reading an article that's placed wrong information front and center? The false info should lower your prior on anything else the article says...

Unless of course you're just looking to confirm your biases, in which case you'll find whatever you want to find.


> Not op, but "We don't know" != "False"

That's not what the fat checker states. At all.

The fact checker states, and I quote:

> however, it has yet to be proven that it can treat COVID-19."

"It is yet to be proven" means, quite literally, anyone who looked into it never saw it work, ever. At all.


> "It is yet to be proven" means, quite literally, anyone who looked into it never saw it work, ever. At all.

This claim is quite literally false. If no one saw an effect, then we wouldn't still be talking about it. Most of the studies listed here [1] all claim to have seen it work. You can of course claim these are poor or unconvincing studies for various reasons, but they still falsify your claim that no one who looked into it ever saw it work.

[1] https://ivmmeta.com/


Your link is an oft-debunked misinformation campaign[1].

In my opinion, the best current review is the Cochrane review[2]. And indeed, I think "anyone who looked into it never saw it work, ever. At all" is overstating the case. There is weak evidence that it works. That could easily go one of two ways: a highly powered RCT (like ACTIV-6) could show that it doesn't work at all, or we could find that it shows some improvement but is definitely not a silver bullet.

[1]: https://ebm.bmj.com/content/early/2021/05/26/bmjebm-2021-111...

[2]: https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD...


Sure, don't take ivmmeta.com analysis at face value, but I was pointing out that they link to numerous legitimate studies which did find positive results, thus refuting the OP's claim that nobody who looked into this ever found anything.


That's fair. The Cochrane review links to many of the same studies, for what it's worth.


> "It is yet to be proven" means, quite literally, anyone who looked into it never saw it work, ever. At all.

[yet].

There are studies ongoing. Not in the anti-vaxxer "I'm still doing research" sense, but in the sense that there are actual non-Facebook studies ongoing. Even that AP fact checker article mentions that.

When the context says there are studies ongoing and yet to be proven is mentioned, it means that the conclusion has not been drawn.

In isolation, when yet to be proven is written with no other context, it means what you interpreted: that nobody has ever seen it work, and we need a nice way to say that.

The fact checker and this conversation both accurately conveyed how different ivermectin is from what Pfizer is developing: the compounds are different, the molecules are different, everything is different. They don't say the result is different. Ivermectin - a completely different compound and molecule - has a hypothesis and limited result showing it functions as a protease inhibitor. It does omit where there are similarities being studied and what those hypothesis are.

This is actually new information to me and I found that looking for a response to you. This isn't about Ivermectin or laughing at Republicans turn their whole states blue from a lack of oxygen as well as the remaining population, its just about what is omitted from the fact checking article.


Just like every other 「fact-check」 I’ve seen recently, this just attacks the most weak and absurd version of the fact. The reason people are saying it’s repackaged ivermectin is not because it’s literally the same chemical in new packaging, but because the mechanism of action for Pfizer’s new antiviral (3CL protease inhibitor) is something that Ivermectin also does.


This is from the fact check and it sounds to me like they're treating the claims fairly and not using a ridiculous version.

'Several posts circulating online wrongly suggest that the pills are the same because ivermectin also acts like a protease inhibitor and keeps the virus from replicating. Ivermectin continues to be studied in relation to COVID-19, however, it has yet to be proven that it can treat COVID-19.

Images of the structure of Pfizer’s new pill and ivermectin can be found online and do not show similarities like the posts describe.

Ivermectin binds to glutamate-gated chloride channels and is used to treat parasite infections, said Joseph Glajch, a consultant in pharmaceutical and analytical chemistry.

“These two are so far apart,” he said. “If you look at how they interact with the body..., they don’t even go to the same pathways or receptors.”'


They are dancing around the fact that ivermectin is indeed a 3CL protease inhibitor, despite that the molecule is very different from the new anti-virals being developed, and despite that 3CL protease inhibition was not primarily what made ivermectin an anti parasitic.

That’s how you know these fact checks are bullshit propaganda. They obviously did their research, but they will discuss everything short of the thing that will actually give credence to the claim being fact checked.

It doesn’t matter that the primary mechanism of action of ivermectin is different from the thing that makes it anti-viral. It doesn’t matter that the molecule looks very different from the new antivirals. These are not important and are just distractions from one simple fact:

the new antivirals work by inhibiting 3CL protease, ivermectin inhibits 3CL protease too.


So who subversive-checks the fact-checker publications? I hope no-one since the idea of an outsourced fact-checker is farcical and we shouldn't fall for any factception play.

Just because they label themselves as meta-news fact-checkers doesn't mean we forget news includes meta news or that fact-checking is an act of bias-prone publication.

The eternal problem of news is one of defining the level of controversy accurately, if we could do that there'd be no need for more than one news publication. And yet somehow no-one that benefits from constant fact-checker approval mentions that these fact-checkers act entirely like syndicated news publications with a lower word count (which doesn't make any claims of being a single-source-of-truth more valid). They are only blunting the minds of their audience by their omissions.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: