There is arxiv, though not everything in there is peer reviewed.

Arxiv is a pre-print repository, It by definition doesn't have peer reviewed articles (at the time of submission)

This has happened. F1000 is an example.

A big issue is funding bodies/grant reviewers are mostly old guard who are happy with the status quo. Recently, a big journal was changing in a way that upset a large swath of such people. Unfortunately i can't remember what journal. But it was obvious they benefited substantially as they're more established.

eLife is most likely what you’re thinking of:

https://www.the-scientist.com/news-opinion/q-a-why-elife-is-...

Thank you for the correction. Yep it's eLife

Several governments have lists of "good" journals that "count" (tenure, grant considerations, etc), and those lists take into account things like impact factor over the past several years. Needless to say, there is a lot of inertia in the system beyond the authors themselves.

>> I'm honestly confused why researchers in one or a few fields haven't already gotten together

I am not knowledgeable, but it looks like just game theory.

Ideally, I would wish you to publish in open access while I reap the immediate career benefits of publishing in a prestige journal.

It's like recruiting the "best" faculty together and creating a new school. People would still not view the degree from it the same way they do one from Harvard. History and brand matter more than you would think.

> I'm honestly confused why researchers in one or a few fields haven't already gotten together and said e.g. "okay, instead of considering publishing in 'Cell' to be the most prestigious for biology, let start an open-source co-op journal called 'Flagella' that by design has exactly the same level of publication pickiness, and works with exactly the same peer-reviewers, as 'Cell'; and then let's all walk around getting signatures on these threshold trigger agreements at our field's conventions, where those agreements would say that that once we have enough people behind this, we'll all switch over to publishing in 'Flagella' at once, considering it to be just as prestigious as 'Cell'; and after that point, we will all be contractually bound to never publish in 'Cell' again."

There are several reasons why this has not happened.

1. The most important, researchers are not publishers and are already completely overloaded with other work (most academics typically work 60h+ a week). They neither have the time not expertise to get this started and make no mistake there would be a significant effort required to get this started.

2. There is a certain percentage of researchers who are quite happy with the status quo. These are typically the very successful researchers which don't lack funding so don't see a problem with this (obviously there are exceptions) . Unfortunately they are often also the ones who have a disproportionate influence so if they are not on board the effort is much harder.

3. The vast majority of researchers highly depends on publications in high impact journals for their grants, careers, jobs... (for younger researchers a publication in e.g. Nature can really set off your career). For a new journal to get a high impact takes at least several years, so this is a bit of a chicken and egg problem. Before the journal has the high impact most can't afford to send their best work there, but without that work you don't get the impact.

#2 and #3 are arguments against the naive strawman version of this that I didn't say, where people just try to start a new journal and get people to publish in it. A threshold trigger agreement entirely short-circuits these problems: the journal doesn't exist at all, until enough people sign onto it; and then suddenly, after the threshold passes, the journal is immediately not just existent but pre-eminent in its field, and the old journal immediately kneecapped by a majority of the field being legally obligated to no longer publish in it, or subscribe to it. The "researchers happy with the status quo" would be left behind in a world where their comfy old journal suddenly stops publishing before the next issue comes out, due to an effective bank-run of churn; at which point they'll have to either raise their heads and look around to find out what's replaced it, or become irrelevant.

(This is a not an obscure technique — it's the bit of game-theory that Groupon is based on.)

this has happened, math has several such journals

You're right, and many/most researchers want to publish in Nature, or Cell, or a handful of journals. They don't want the cheaper and well-regarded journal that's perfectly tailored to their research. Part of it is ego, part of it is funding requirements, part of it is academia's requirements. Publishers know that pay-to-read's days are counted, but it's actually a mission to convince researchers to use open access.