Online archives where scientists post research spark information revolution | Hacker News Reader

Online archives where scientists post research spark information revolution

washingtonpost.com

39 pointsbookofjoe22 comments

Open articleView on HN

https://archive.ph/ms9eC

This is about preprint servers. Cool.

But for some reason I had inferred from the headline this was about sharing the actual data. As in the source datasets.

Ages ago, IIRC, Univ of WA had a project called something like ShareSQL. Made by astronomy or geology dept?

Idea was to put your datasets online in read-only one-off databases.

Sorry, I can't quickly refind this project. "ShareSQL" is too generic for searching. And probably not even the right name.

Happily, hosting datasets online is getting a lot easier. IIRC, startup fly.io hit HN front page last week. They'll host your Postgres instance. (Like a blogger for databases. Ha. That's a pretty good tagline.) I'm sure there's dozens more startups.

Any way. We're long overdue for sharing of data. I don't really follow this git ecosystem, but I've gleaned that hosting binary blobs is getting easier. And ever more accessible, like the hosting of SQLlite instances accessible via HTTP, also recently front paged here on HN.

The opening of this piece reminded me of papers showing that retracted peer reviewed papers still get cited (for evidentiary support) at high rates in the scientific literature.

The ivermectin case is interesting too, as it's not exactly like ivermectin hasn't been scrutinized at this point.

I'm not sure peer review as part of the publication process per se matters as much as it's made out to be.

Having seen this before [0] I fully expect in 10 to 15 years to start hearing about how these servers must combat misinformation and ban anyone who is posting non-mainstream science there.

[0] https://en.wikipedia.org/wiki/Wisdom_of_the_crowd

Yeah. Combating misinformation and figuring out "what is actually true" is REALLY hard - even for objective things like "how do I install a package"

Arxiv.org has been online for 30 years. Is there a specific reason you think they (or the other preprint servers) might grow less open in ten more years? They have a Covid research banner on the front page now, but it’s a reasonable blanket “don’t get your medical advice from a preprint server” kind of deal

Arxiv doesn’t formally review submissions, but to post there you have to pass a topic filter (to make sure you fit in a covered sub field) and meet some other criteria

This part is probably the closest to a filter on non-mainstream science: https://arxiv.org/help/moderation#scholarly-interest

It’s more geared towards filtering pseudoscience and non-research stuff than blocking articles because mods don’t agree with the conclusion.

The photo caption is a good example of misinformation. There is no evidence that anyone was harmed by taking prescribed (human) ivermectin:

"A debunked ivermectin study was later retracted from a server, but the damage was difficult to contain: It had been viewed more than 150,000 times and covered widely by the news media not to mention contributed to a host of serious side effects among those who ingested ivermectin with no proven benefit against covid-19. (Mike Stewart/AP)"

Reply on news.ycombinator.com