Let’s stop reviewing for publishers that profit from research! Join the strike! https://josephpcohen.com/w/reviewerstrike/

I wish this would work but it won’t. They’ll just get shittier reviewers who’re willing to do it. ImmigrNt researchers use reviews as evidence of expertise so there’s at least one group of folk who are incentivized to review for anyone.

Yes, there is a career incentive to review, but it's much weaker than the one to publish, and comes later in career, so hopefully more people can afford it.

And yes, there will probably always be people willing to do reviews for free. But if the quality/delay/expertise of reviews from commercial publishers starts to degrade, it encourages authors to look for alternatives.

The issue is as a semi disingenuous researcher (which honestly most are) you actually want reviewers who are not experts because they won’t ask the hard questions and do a thorough review. So if you can get your nat neur submission reviewed by a bunch of desperate postdocs aiming to please then you’re hitting jackpot.

True, but academic journals usually try to maintain some sort of acceptance ratio, and avoid accepting every submitted article. (If they do, they will have the reputation of not being selective, which will harm their rankings and will make people try submitting their work to more selective venues first.)

Of course, for commercial publishers, there is an incentive to publish more -- so there is tension between taking more articles (more revenue in APCs, more volume, easier to attract subscriptions, more perceived impact for extensive properties like h-index) and being more selective (more prestige).

So worse reviews probably doesn't mean more papers accepted but a more random selection of accepted articles.

Assuming that articles have some intrinsic "quality" that most researchers would agree on, and that good reviews are better correlated to this quality than bad reviews, one would expect that getting worse reviews would make the journal selection process (even) more noisy. This can harm the prestige of commercial journals (if they start rejecting articles perceived as "good" and publish articles perceived as "bad").

I also have an initiative to boycott reviewing for closed-access journals with currently >350 signers. If you agree, consider signing it :) https://nofreeviewnoreview.org/

Indeed, for-profit publishing only works because researchers are willing to do free reviewing work for such publishers.

To be honest, it is a bit naive to think people will boycott one of the most prestigious journal in their field for this reason

There needs to be nonprofit co-op publisher(s) for the scalable benefit of many institutions. Pack-up and form equivalent new journals.

Let’s stop reviewing for publishers that profit from research! https://josephpcohen.com/w/reviewerstrike/

It is hard to make this requirement because then there is a whole set of ideas that cannot be referenced. It would close off the community that imposed that rule. In the past people have cited Geoffrey Hinton's slides for a concept origin. I don't think it is good to limit citations. Credit is important.

But it would be nice to see some rules on forums that say you cannot post paywalled items there. It would hopefully give more incentive to authors to not publish paywalled articles.

But I think a reviewer strike is a better solution that can result in a complete shift away from paywalls. The market is exploiting free labor, so without the free labor the business models must change.

We have collections which I guess should be featured on the front page. https://academictorrents.com/collections.php

There is metadata. It is stored in bibtex along with every torrent. This format allows it to be a freeform database where the user can add fields as they want. We (Academic Torrents) can then build new ways to display this metadata. Also the "abstract" part of the metadata is rendered as markdown on the details page of a torrent. Here is a good example: https://academictorrents.com/details/d52ccc21455c7a82fd6e589...

Ok, I see that there is code provided there. Better than nothing but geez, it is not really what metadata should be like

  def get_labels(rightside):
    met = {}
    met['brain'] = (
        1. * (rightside != 0).sum() / (rightside == 0).sum())
    met['tumor'] = (
        1. * (rightside > 2).sum() / ((rightside != 0).sum() + 1e-10))
    met['has_enough_brain'] = met['brain'] > 0.30
    met['has_tumor'] = met['tumor'] > 0.01
    return met

I will say that it is very handy to know exactly how the labels were computed.

What I really meant is a way to search and select data based on metadata. For example has_tumor.

Also note how everything is still one single blob, to get one line of any of the files, one would need to download everything.

Bittorrent does support partial downloads that request only some files or byte ranges out of a torrent. Some of the torrents are just compressed zip's but for the others you could look at the code / documentation to see which files were relevant before downloading 10GB of data.

I think the abstract is sufficient for searching data; expecting some kind of smart database that can handle all the weird formats science uses is a bit much.

There are even torrent clients that export a FUSE VFS so you can use your standard tools.

| one would need to download everything

Just download it then. We got mp3 albums off Napster on modems back in the day, surely getting that torrent is easier and faster today.

Sometimes if a torrent doesn't have seeds it is a signal that people do not value that data. Not all data should be stored forever. The community should vote with their servers what we should store forever and at what level of redundancy.

You can browse collections and then order by seeders. Here are some collections I think should be mirrored:

http://academictorrents.com/collection/deep-learning?sort_fi...

http://academictorrents.com/collection/medical?sort_field=se...

http://academictorrents.com/collection/joes-recommended-mirr...

I'm assuming those globe icons means there's a web seed (HTTP server) involved? If so, I'd prioritise the ones that don't have that backing.

Yes. But webseeds also go away over time when a department shuts down a server or a student account expires.

The data is hosted by the community and we also coordinate hosting from our sponsors (Listed on the home page)

We work with academic institutions to ensure they allow this service. Please report universities which block the service using the feedback button shown in the lower right of the webpage.

We also encourage HTTP seeds to be specified (aka url-lists) by the uploader to offer a backup URL which can be contacted automatically if BT is blocked. We also offer a python API designed for clusters and university computers written in pure python which supports HTTP seeds: https://github.com/AcademicTorrents/python-r-api

The project is run by the U.S. 501(c)3 Non-profit called Institute for Reproducible Research (http://reproducibilityinstitute.org) and this site has an overhead cost of ~$500/year. We plan to fund this project for at least the next 30 years. The community hosts the data and we also coordinate donations of hosting from our sponsors (listed on the home page).

We also run the project ShortScience.org! Check it out!

> The community hosts the data

This is the key thing for me, so there are no guarantees of the data being available? Or is all the data backed by an owner hosted box (and backups)?

The main goal is to facilitate distribution. We work with research groups and provide "best effort" hosting with the resources from our sponsors. We expect the uploader has a backup of the data in all cases.

Sometimes people upload 1TB files which are not intended to be mirrored or not of interest to many people. We don't want people who donate hosting to mirror this content unless they really want to. But we also want to make it easy and automatic to mirror content. Using collections, which each have an RSS feed, content can be curated by someone you trust to decide what should be mirrored. I curate many collections including videos lectures, deep learning, and medical datasets.

Got it. Not trying to knock the hard work put into this, I'm actually thrilled to see this initiative and only intend to be constructive. Personally, I would rather be asked to trust that site admins were auditing each torrent to ensure it at least looks legitimate, before passing final legal responsibility on to me as a seeder. Leaving users to identify contributors they can trust to never include "Pirated Movie 2018" into their donated seedbox sounds like quite a hurdle to attracting new seeders willing to participate in a "legitimate bittorrent use case" project.

We perform our own audits of all data and request a justification why it is academic data if it appears to be "Movie 2018". We think the collections model is the best balance between a walled garden and zero censorship. You can be assured that no collections curated by me will have a bad torrent in it! Here are a few:

http://academictorrents.com/collection/medical

http://academictorrents.com/collection/video-lectures

http://academictorrents.com/collection/joes-recommended-mirr...

http://academictorrents.com/collection/computer-vision

http://academictorrents.com/collection/deep-learning

Isn't this the purpose of IPFS? You get to host what you want.