Hacker News Reader

Comment by dannykwells | Hacker News Reader

aRxiv / bioaRxiv / medaRxiv exist, are free and are backed by extremely well known names and institutions in the scientific community. They do not require some random coin, you can just, well, upload a preprint of your work.

They also did the hard work on making sure journals will accept papers uploaded to there.

Maybe they don't have a flashy website but I would much much rather use those than upload my work - which I have invested years of my time + thousands or millions of dollars in - to some random site with founders who are just "from San Francisco".

For those trying to "accelerate science" - please try to take even a modicum of time to understand how science is even being done currently and start from there, instead of making up solutions to problems that don't exist.

View on HN

> For those trying to "accelerate science" - please try to take even a modicum of time to understand how science is even being done currently and start from there, instead of making up solutions to problems that don't exist.

:%s/science/industry/g

So many of these cryptocurrency-based “solutions” could have this same comment aimed at them. So much time/effort/money spent rolling out web services that don’t even understand the basics of the industry they’re trying to “disrupt”.

I’m not anti-cryptocurrency at all (and have my spread that I support) but this “take x and add a token to motivate people to post x to our site” is getting pretty old.

> aRxiv / bioaRxiv / medaRxiv exist, are free and are backed by extremely well known names and institutions in the scientific community. They do not require some random coin, you can just, well, upload a preprint of your work

yeah, but they don't use crypto :)

The websites support TLS just fine, and they work. Same with Sci-Hub. We don't need cryptocurrency for this purpose (as usual).

Non crypto no cool.

(I'm being sarcastic here if it wasn't clear)

I think your skepticism is warranted, and to be honest, is probably shared by the majority of academic scientists.

To give a little background - I'm the CSO of ResearchHub and I come from scientific background. After college I started a PhD program in molecular biology and I've worked in epigenetics and synthetic biology labs. I ended up dropping out after seeing how difficult it was to achieve success as an early-career scientists. A few random stories:

-A friend of mine was told to leave her PhD program with a masters degree because she was unable to replicate a famous result in her field. Three years after she left, the paper she was told to replicate was retracted.

It turns out that >50+% of research is unable to be replicated, so I'm sure this is not unique to her experience.

-A PhD student in biochemistry lab I rotated through was in his 9th year of the program (normal is 5-7), completely out of funding, sleeping on a friends couch, and paying for his experiments out of pocket just to get the publication he needed to graduate with a degree.

Academia is riddled with perverse bibliometric incentives that encourage researchers to partake in behaviors which hurt the ecosystem as a whole. For example, an article shared on HN a few days ago showed how irreproducible papers are cited 153x more than reproducible papers. This is because more interesting results are more likely to receive citations, but also less likely to replicate.

After I left my graduate program I was admitted to medical school. My institution was fantastic because they got med students into the hospital to interview patients during the first week of classes. Without typing a novel, many of our front-line therapies are surprisingly old and ineffective. The waste caused by a broken incentive structure for academic research is felt within clinical medicine.

You are absolutely right that arXiv, bioRxiv, and MedRxiv are fabulous projects that have pushed the culture of open science forward tremendously. Also, should out Open Science Framework - they are amazing.

While these tools help scholars share their work in the open, we hope that ResearchHub and ResearchCoin will help to fill a grab by creating new incentives that can help to loosen the grip of bibliometrics on a scientist's career prospects and funding decisions.

ResearchCoin was included because it allows our community to create a dynamic system for rewarding open scientific publishing/discussion that can change over time. In addition, the ResearchHub DAO to allows scientists to help govern how rewards are distributed. The admittedly quite lofty goal is to create a democratic and evidence-based incentive structure for science.

While it is possible that the cryptocurrency component will make it easy for some to dismiss ResearchHub, we believe crypto is an elegant solution to address some of the root problems in academia.

I have some questions about the cryptocurrency component that I could not see addressed on the website.

1. Where does RSC originate? I see that there are 1 billion RSC in existence. Can more be mined? Do you just generate new ones at will? Is this basically a 100% pre-mine of a token?

2. Who owns the RSC that exist? You plan to distribute some proportion of them. Presumably these can be converted to fiat dollars at some point. Supposing they become valuable, can you just sell off the 99% or so that you haven't freely distributed? What keeps the DAO acting in the community's interest?

3. Popularity is a poor indicator to choose to optimize for if you're looking to incentivize high quality. Even if you were able to build a rock-solid system free of bot accounts and manipulation, which is not your current system, the research and summaries receiving the most RSC would likely be clickbaity papers or those which can appeal to the widest audience. Do you have a way to address these issues?

4. What gives RSC a real-world value?

5. How do you ensure the right person is rewarded RSC? If I upload someone else's paper and get a reward, that doesn't seem right. Especially if a few months down the road they wanted to upload the paper themselves to profit in RSC from it. But presumably you would discourage duplicates, so they couldn't re-upload even if they tried. Similarly, how would incentives be structured for collaborative research? 3+ authors to a paper is not uncommon. It seems like the current system is just to reward the uploader, whoever they may be, like earning a "finder's fee". But then you just end up rewarding whoever can scrape and upload papers the fastest rather than those who put in the work to produce quality research.

I welcome the idea of a community built around open and accessible scientific research. But when you add financial incentives, especially when they're this easy to game and with a cryptocurrency entirely controlled by some opaque entity it raises some major red flags. Hope you have the time to answer some of the above.

Thanks for all of these questions. I hope my answers are able to help. If you'd like - it would be great to connect to have a more in-depth conversation at some point. Feel free to reach out to hello@researchhub.com if want to chat!

1. Yes, RSC is 100% pre-mined and earmarked for distribution to the community, team, founders, and ResearchHub corporate entity. We plan to distribute about 5-6% of the total RSC supply per year to the community as rewards for posting science on the platform.

2. In the future, if we run out of RSC and need to add to the total supply the ResearchHub DAO will be able to vote to mint new tokens. There are two mechanisms at play that keep the DAO honest:

* An open source codebase - If the ResearchHub team ever goes rogue and fails to act as stewards of the community, the community will be able to fork our codebase and manage a new version of ResearchHub how they see fit. See the Steem vs. Hive example to see how open-sourcing can give community an option to exit: https://hackernoon.com/inside-trons-steem-takeover-attempt-a...

* RSC holding represent ownership in the DAO. For every coin held, an individual wallet gets one vote. This means that the biggest hodlers of RSC will have the most influence.

On the surface the concept of more coins = more influence might seem like a bad idea, but we believe that carefully distributing RSC into the hands of the scientists and community members who care the most about our mission will result in the best governance.

Regarding guarantees against the team dumping tokens - we currently have vesting smart contracts for employee token grants and would be open to enacting other measures to ensure a healthy market for RSC if requested by the community.

3. I wrote a little bit about this in another response. Let me copy and paste here:

The way coins are awarded to posts can change over time via a democratic process facilitated by a DAO. In theory, this could create the infrastructure for the scientific community to self-govern how their work is assessed and rewarded.

Perhaps a 1-10 peer review rating would provide a better signal than upvotes? Or maybe basic research deserves 1.5x rewards because it helps to fertilize the landscape of translational research. If the community wants to make changes to Researchhub's incentives, we will be able to iterate over time until we find a formula that properly encourages the behaviors that benefit science the most.

4. As of today RSC gives its holdlers the ability to vote in the ResearchHub DAO and participate in the governance of the network. Within ResearchHub you can tip users for their posts by "supporting" their content with RSC.

5. We’re currently building out a feature to help address this. Any scientist who has authored a paper shared within Researchhub will have a pot of RSC waiting for them upon sign up. It is in our best interest to properly credit reward the authors of manuscripts because in an ideal world they would have a larger influence in the ResearchHub DAO compared to the person who simply posted a link to their paper in our forum.

> Thanks for all of these questions. I hope my answers are able to help.

"While it is possible that the cryptocurrency component will make it easy for some to dismiss ResearchHub, we believe crypto is an elegant solution to address some of the root problems in academia."

"let me answer or non-answer these questions in multiple separate responses instead on the website I'm running that is definitely not a scam"

> ResearchCoin was included because it allows our community to create a dynamic system for rewarding open scientific publishing/discussion that can change over time. In addition, the ResearchHub DAO to allows scientists to help govern how rewards are distributed. The admittedly quite lofty goal is to create a democratic and evidence-based incentive structure for science.

I really don't understand how it helps research, can you please elaborate.

Do you mean researchers get paid in a coin, which they can sell, which will fund further research? If the coins became super valuable wouldn't this simply do the opposite: incentivize cheap/easy papers about popular/meme subjects in order to get upvotes?

Perhaps a 1-10 peer review rating would provide a better signal than upvotes? Or maybe basic research deserves 1.5x rewards because it helps to fertilize the landscape of translational research.

If the community wants to make changes to Researchhub's incentives, we will be able to iterate over time until we find a formula that properly encourages the behaviors that benefit science the most.

To speak to your example - if the v1 incentive structure encourages cheap/easy/meme science, we would be able to recognize this and make improvements to hopefully achieve better results.

This is an improvement over the current static system which values bibliometrics because it has the ability to adapt and improve over time.

Thank you for your answers. Best of luck.

> for the scientific community to self-govern how their work is assessed and rewarded.

Doesn't this already happen? Science chose bibliometrics as its God. Making a new God out of crypto and crowdsourcing doesn't feel like it addresses the root concern, which is human bias in the scientific method. It sounds like the system you're divising is just a new way for humans to bungle things with bias (no offense intended). I think you're on to something with trying to develop a new method, but perhaps we should figure out how to remove the bias before involving money and human heuristics?

I appreciate you sharing your stories. And it seems like you really have the hands-on experience with science and know the major struggles.

Now, when you talk about solutions it gets really fuzzy. There're very specific problems that exist and you need to create a well-defined incentive. As a graduate student in bioinformatics, my main struggle is the source code. People either don't share any at all, or share something you can't possibly run.

> we believe crypto is an elegant solution to address some of the root problems in academia.

So what are the root problems? First off, the code isn't written by the software engineers, but rather by the biologists who learned how to write some R or Python scripts. Even the CS graduates who switched to bioinformatics have no idea how to develop quality software.

Next, the universities often have their own hardware for computation(HPC). When running on HPC, you often use the pre-installed modules. For example, my university's HPC doesn't even allow running Docker. So in order to make my code runnable somewhere else, I need to put in an incredible amount of effort. Would I do it for some crypto? No.

Let’s assume ResearchCoin takes off as a new way to assess academic output, and it will become a factor for tenure committees.

The stakes are extremely high in the academic job market, competition is fierce and all incumbents are highly skilled and tremendously resourceful.

Are you prepared to maintain the integrity of ResearchCoin against all sorts of attempts to gamble the system, or even adversarial attacks by those whose careers depend on it?

> While it is possible that the cryptocurrency component will make it easy for some to dismiss ResearchHub, we believe crypto is an elegant solution to address some of the root problems in academia.

And this is where you list the root problems in academia and show how yet another virtual token solves them, and this helps people to see its value, and not dismiss it easily.

Oh, wait. You don't. Not here, not on your website. I mean, you do mention that "Further details about ResearchCoin can also be found on the ResearchHub Notion page.", but don't even link to that page.

Hi joyecesticks, if I could make one suggestion: if you list the DOI in the metadata of an article page (e.g. by sticking something like `<meta name="citation_doi" content="10.1101/2021.05.24.445440">` in the `<head>`) you'll make it easier for search engines like Google Scholar to find them and associate them with other versions of the article.

Additionally, it allows tools like https://plaudit.pub to add functionality like allowing people with a browser extension to endorse the articles. (Disclosure: I volunteer for Plaudit - hit me up if you want to integrate it into ResearchHub natively!)

Although the DOI is most important, there is other metadata you can expose. For more info, see: https://scholar.google.com/intl/en/scholar/inclusion.html#in...

Very interesting. What about adding a discussion forum feature to this, people with an orcid discussing an article with a doi?

Certainly an option, mainly limited by a lack of resources to implement that :)

Well I think it’s fine to try new models, what’s the harm? I also applaud attempts to use cryptocurrency for something useful.

Academia is quite thoroughly broken, everyone in it hates it. The reality is with the existing publishing model you do upload your precious work to some random site. There it is judged by faceless editors, and sold for profit if they deign to publish it, after you have done professional level editing and graphic design for free. And you do this only because of a consensual hallucination that this is not only essential but the true measure of one’s worth as a scientist.

Reply on news.ycombinator.com