Do you think Springer pays for reviewing? No publisher does. It's all voluntary work by the same people who write the papers. It's the modern day slavery.
The parent comment implied that reviewers would be anonymized with this Github-based approach unless the PR was approved, which seems like it would lead to more acceptances so that the reviewers could get credit.
By providing opportunity to find errors in research of people you don't like!
More seriously though, peer review is usually free or almost free, reviewers do not get anything near to the ridiculous amount of the money that "open access" journals demand.
Besides, having articles on github, with all the data, so that any reader can check the results, and comment, would be much more powerful than the type of peer review we have today.
First, I'd question that you really need to incentivize it. People - myself included - are doing research in a topic because they are passionate about it and most are more than happy to look at new research and ideas about things they are passionate about. Will you really need incentivization to get people to interact with the topic they have devoted their life towards studying?
Let's say however that you need incentivization, I suppose one approach would be at a legislative level where you require all researchers publish on a decentralized open access platform where every citizen - they are the one's funding the research after all - can freely and uninhibitedly access the information.
> Will you really need incentivization to get people to interact with the topic they have devoted their life towards studying?
Perhaps not. But people keep telling me in this thread that reviewers work because the journals expect them to or they will get future papers rejected/lose conference access, not for the love of the topic or because they get paid or otherwise any kind of special visibility for it. I wonder whether an approach that promised only publicity for having done reviews or relying on altruism would really produce the quality of reviews necessary.
> I suppose one approach would be at a legislative level where you require all researchers publish on a decentralized open access platform where every citizen - they are the one's funding the research after all - can freely and uninhibitedly access the information.
That's not really incentivizing the _reviewers_, though, right?
The same way we do now? There's a strong cultural expectation that one will agree to review, or have one's graduate students review, a reasonable number of papers. If you're a member of the Foo committee of the Society for Bar, you're not going to refuse to pull your weight as a reviewer for the Society for Bar's International Journal of Foo unless you want to get the cold shoulder at your next conference. It's not like anybody gets paid for this today.
But no seriously, this seems like an actual technology / problem fit, although you'd need a decent amount of tooling, product design, etc for it to work - but some of that's already being explored by the "open source + blockchain" endeavors.
In basic form, I imagine you'd basically turn the transaction fees into review fees.
(The product design and tooling comes from how you prevent abuses to that basic format, I think)
I work for an Open Access publisher/research discovery technology company, and we literally have this on the drawing board. We're small, but well connected in the world of publishing and research.
I for one would be also be interested in links.
One thing we don't want to do is reinvent the actual blockchain technology, so we're also interested in finding partnerships in terms of platform technology.
A proof of stake smart contract would be perfect for this. The currency is reputation earned by writing well-done reviews and 'staked' behind reviews they believe are well-done. Papers need a certain number of well-done reviews with the stakes on both sides accumulating behind approve/deny. You have to pay with the currency to have your paper reviewed and the currencies are distributed to the reviewers to allow monetization via exchanges with a small amount withdrawn as a siphon to support the continued development of the network and marketing/public relations.
Agreed! There's a lot in the blockchain model that seems to lend itself to this problem in particular.
Ideally, we'd also like to set up a foundation to maintain and manage the ecosystem and invite both old and new players to participate. A lot of blockchain efforts seem to be a little too much like a hope for a monopoly wrapped in a thick layer of talk about free markets and democracy.
We're actively looking for collaboration and partnerships around this. If you, or anyone, would like to get involved we'd be happy to discuss. My email is in my profile.
Neat. I'm at a crypto conference in a couple weeks, I'll see if there's any sparks on this idea.
I agree with an above post that the starting point is that you get paid to review, and you pay to get reviewed; although you'd need another way to monetize the token or it's limited to this use case.
The overlap with open-source seem clear: How do you validate the value of the crowd-sourced contributions? In code, you could do it via TDD...
OH! OR - (well, you could make both), you pay to review by _betting for it's validity_. There may be a way to combine or borrow from each approach to fill in the gaps in the other.
I'm one of the main developers of OpenReview.net, and I would love to discuss this more with you, but like the other commenter, I can't figure out how to get your email from your profile (I'm also generally new to HN and may just be missing something - for example, is there a direct message-like option here?)
I think this was meant as a suggestion, I've also never seen this happen. However, I am wondering why and so may I ask why you do not want to be mentioned as reviewer on a paper, maybe even with your review?
not parent, but research association conference presentations might help to centralize announcements/review efforts but also keep things under researcher purview..
