Hacker News Reader

Comment by convolvatron | Hacker News Reader

you're right about the phrase, its basically an assertion that "we're all cheating scum, so I have no choice but to be a cheating scum myself", which is hugely corrosive. and in this case its the funding system more broadly that's imposing these non-goals from above that are incentivizing bad science.

but why are they imposing these structures? to try to weed out the cheating scum. once you start walking down that path, you're signing up for a distortion of value.

View on HN

As I said to the parent poster, that's not what it means at all. It means that you should look at the system's incentives, not the behavior of individuals as the root cause of any issues.

You don't need to be a "cheating scum" to succeed, but there are not enough checks in place to prevent that from being a successful strategy for someone who wouldn't succeed otherwise.

The people who need to change the most are the nameless "they" who issue funding because they have the most control over these systems, along with the publishing cartel which has almost no redeeming value in today's environment.

Nobody says the phrase when they are calling people to look at a system's incentives. They use the phrase as a response to personal criticism excusing and rationalizing their own bad behavior.

It is a deflection of personal responsibility, full stop.

That's objectively false, with the article in question being example #1.

And yet changing the game generally has better results than trying to change the players.

Agreed. Accountability matters, but changing the game usually scales better than hoping for better individual behaviour under the same pressures. Academia needs systems that reward transparency, verification, and contribution more directly. That is part of what we are building with Liberata if of interest: https://liberata.info/beta-signup

Some unsolicited feedback from someone who was at one point part of your target audience for this product:

* You need to put some text on the pages on your site describing what this actually is and who is working on it. No one is taking the time to watch videos (especially one that is 9 minutes long!??!?!?!)

* The problems listed don't give me the impression that the team has much experience with publishing in academia and are generally unfocused

* Related to the item above, you're both making this way more complicated than it needs to be, and completely ignoring or glossing over the primary issues at hand (the network effect of the existing publisher cartel and the tension between the requirements for obtaining funding and producing reproducible research findings)

I don't think you're in a position to have much impact on the major issues, making any impact on the more minor ones kind of irrelevant, but if I'm wrong: I would focus on making it easy for academic communities to start their own open access journals (e.g. the Journal of Machine Learning Research) and provide a tool to automate citation checking to start, with steps towards content management for the content that would allow an external party to reproduce the results from the paper).

Recruiting people who can break away from the current model to build a new one is the only real chance of success, and your role would presumably be to make it easy for them to make that change. This is not a novel concept, and many platforms already exist to enable open access journal publication, yet their adoption is not widespread.

The question I think you should address on your website (in text form) is: What are you doing differently that will cause a different outcome?

Reply on news.ycombinator.com