Thread by @mbeisen

Did a quick read of @BrunaLab's study of APCs and the geographic diversity of authors. It's really good, and is another important reason why we need to move beyond APCs to a publishing system that is free for everyone to read AND publish direct.mit.edu/qss/article/2/…

Publishing biases are incredibly difficult to address, and there are always a bazillion caveats, which they acknowledge, but they did as good a job as one can do with this kind of question.

Their data are pretty convincing that authors from low income countries are significantly less likely to choose to pay APCs when non-APC options are available. This is not the first observation of this effect, and I think we should stipulate it.

The question of why this is is more complicated however. They attribute the effect to being primarily about APCs being a barrier, while I think there are good reasons to think it is also about asymmetric incentives (e.g. funder and institutional mandates) to publish OA.

But exactly how these and other effects partition out, my experience with @PLOS and @eLife is that there is a large geographic asymmetry in how scientists see APCs. Some of this is directly financial obviously. But it goes beyond that as well.

In the early days @PLOS we had a blanket no questions asked waiver policy (this was ultimately eroded, over my objections, by other members of our board), and we have one now @eLife. And even in these fairly frictionless situations, people were reluctant to ask for waivers.

And I don't think it's all that complicated to understand why - people don't like being treated differently than their peers and colleagues, even when it's nominally to their advantage.

People accept travel grants and fee waivers for conferences when it is the only way for them to attend. But when given the choice between journals that are free for everyone to publish in vs ones that are are free for them to publish in they consistently choose the former.

We can argue about whether this makes sense - and point out that the publishing costs of non-open access journals are being subsidized by subscriptions - but this has not and likely never will be a winning argument.

Instead we should just accept the reality that publishing is subsidized - whether by subscriptions or APCs - primarily by funders and institutions with the resources to do so, and should continue to be because publishing is the lifeblood of science.

It is detrimental for a million reasons for this subsidy to come in the form of subscriptions, as they necessitate denying access to people without subscriptions, but it is also detrimental to replace barriers to reading with barriers to publishing.

My biggest quibble with @BrunaLab is this

https://twitter.com/BrunaLab/status/1494454667794915332?s=20&t=PZckEWTiEACaBx5zwYPX0Q

.

While there are notable exceptions, society publishers by and large have been just as bad - and in many worse - than commercial publishers. So this isn't a solution, without system change it's a capitulation.

The solution - something we should all now be fighting for - is a system in which global science funders are taxed (in a highly progressive manner) to support key publishing services with no transaction costs of any kind. Nobody pays to publish. Nobody pays to read.

Publishing is only one barrier to participation in science, and not the biggest one (almost certainly research funding is). But this is something we could easily do, and it would be an important signal that science actually cares about being a truly global endeavor.