oooooooo boy i didn't see that #RELX's 2022 annual report has been released. Time for some #HateReading

report: https://www.relx.com/investors/annual-reports/2022
investor presentation: https://www.relx.com/investors/investor-presentations/2023

#SurveillancePublishing

@neuralreckoning @Samuelmoore
This is why it is absolutely essential to understand the way that #SurveillancePublishing changes the calculus of these questions.

The services Samuel is describing also include mechanisms for ensuring that prestige and other incentives survive any sort of formal efforts to decouple metrics/evaluation/etc. from eg. citation counts. That takes a few forms, including building paper recommendation/'discovery' systems that will systematically favor flagship/higher APC journals, as well as selling surveillance-backed researcher/research evaluation products like SciVal to both researchers and granting agencies alike. The example of South Korea is extremely telling in this regard: https://jon-e.net/infrastructure/#one-more-turn-of-the-screw-the-ability-of-the-former-publishers

The shift from subscriptions to APCs seems subtle, but changes nearly all of the strategic calculations from the prior era. Prestige is operationalized in entirely different ways, as is profit: as Samuel says, we focus a bit too much on the high end of APCs, when the lower end are just as pernicious - reading between the lines of investor calls, and other patterns, a clear business model of generating a smoother prestige gradient to generate maximal price discrimination for work at any 'prestige level,' where the higher volume 'lower tier' journals are an integral part of that model ( https://jon-e.net/infrastructure/#information-capitalism-in-its-terrifying-splendor-here-too-pits ).

So even funders explicitly not relying on traditional prestige metrics wouldn't be enough (to the degree that's even possible, you can't exactly tell people they aren't allowed to care about Nature papers against a strong acculturated backdrop of caring about them) - if they control the recommendation, authoring, and other systems that govern being able to find a paper, as well as selling employee ranking tools based on compliance with their system, the situation looks like much more of a need for a whole-of-infrastructure approach rather than treating publication or even prestige in isolation.

Which is yet another time I find myself doing my steve ballmer chant - "it's the infrastructure, the infrastructure, the infrastructure, the infrastructure"

I think the calculus of #RegisteredReports might have flipped in a #SurveillancePublishing APC-driven #OpenAccess world.

Subscription models meant that a journal could command a high price by being in high demand in a self-reinforcing cycle where since most libraries subscribed to them then they would have high readership, etc. Multiply that by the power of bundling or whatever. Libraries being a conduit could tell who read what and tailor subscriptions accordingly. actual loss of readership could impact subscription cost during negotiations, so null results are less attractive because they command fewer readers and citations and whatnot. publication bias ensues. the classic story.

in an APC world, where the profit is derived from authors willing to directly pay more for the attendant view count and citation, the registered report is instead more like a commitment to pay at some future time to publish. if the prices keep going up, the journal effectively invests in your need to publish as a security.

this is doubly perverse in a surveillance publishing system, where the publishers operate paper recommendation and rating systems linked to funding and employment decisions. in that case, they can just manufacture the view count and citation - and even literal "scientific value score" - as a function of APC price, so null results aren't even a problem since the exclusivity-prestige link is partially dissolved.

I wonder if causes for publication bias could have changed substantially enough that registered reports could backfire as a means of combating publication bias. Since the primary filter is the perceived importance of a piece of work - assuming the authors could pass some competency and design check normal to the field - which is most likely to be at least partially evaluated by the same system of self-fulfilling metrics used in the recommendation/scoring systems for funders and employers, they might directly reinforce hype cycles. couple that again with the prestige gradient model of APC pricing where one publisher owns many Journals at different prestige levels and can bounce you down the ladder to one with a lower but still high APC.

Journals then would then be effectively sorting papers by APC according to the propensity for views/citations, regardless of outcome. It's sort of a combination of payola and security. plz lmk where I'm missing something here bc not just trying to shit on the parade.

@risottobias many if not most of the problems of contemporary academia are rooted in our allegiance to the traditional journal mode of communication, particularly with their evolution into #SurveillancePublishing, and most of our attempts to resolve those problems involve either publishing more journal papers or founding more journals about it.

"Published in journal" does not necessarily imply "uncreative" for all subjects but when thinking about root causes of problems in academia it almost always does (unless it specifically problematizes the notion of journals as such and notes their role in the communicative regime the paper exists in)