[michaelmior]

> which happens to add value to nobody but their own pockets.

While publishers as they exist now are not necessary for this, the publishing process does typically incorporate peer review which has significant value. I agree that things need to change, but I don't think it's true that zero value is added by publishers.

[sonzohan]

> does typically incorporate peer review which has significant value.

Significant value which is given gratis by said peers, which journals use to boost their reputability and, by association, their profits. Publishers are profiting off of free labor from subject matter experts. Even more disappointing is this free labor is viewed as a right of passage. Don't forget that the author spent hundreds to thousands of dollars to access these unpaid peers. Publishers are increasingly well-known as scammy. It's why MIT ended their Elsevier contract, and why many other R1s are following suit.

Also don't get me started on the dubious quality of peer review in todays "Publish or Perish" climate.

[jampekka]

> It's why MIT ended their Elsevier contract, and why many other R1s are following suit.

I think focusing on Elsevier is/was a mistake. All the publishers are running the exact same racket. Even non-profits like IEEE. It's a disgrace.

[relaxing]

Elsevier are the biggest and most aggressive in driving up prices.

But you’re right the others aren’t far behind.

[michaelmior]

I agree that the value is primarily given by the peers. However, someone needs to coordinate the process. That said, while I can understand viewing it as free labor, it is often (indirectly) part of the job description of faculty members.

[thomasahle]

In many fields, particularly those that rely on conferences over journals, peer review is organized by others, not the publisher. So here they are truly doing nothing but hosting the papers behind their pay wall.

[jampekka]

Peer review and editing is done free by academics. Publishers "facilitate" this by having horrible unusable sofware platforms from the 90's for this. These are actually worse than nothing, as review would be a lot easier to organize and conduct via e-mail.

Publishers don't add value, they subtract it.

People probably have hard time undersanding this because the system is so absurd and so obvious racket they don't think such can exist.

[michaelmior]

I am an academic who spends a nontrivial amount of time reviewing papers. My experience has been that the platforms used for submitting reviews are perfectly fine. Often outdated, sure, but I've never really found this to be a significant barrier. I would personally find it more challenging to manage reviews via email.

[jampekka]

For reviewers they are usually least bad, but still bad. For example Manuscript Central still requires pop-ups and can be more or less impossible to use with mobile devices. It also has separate login for all journals and password rules that makes it very common for people (including me) to have to reset the password every time they need to log in.

For editors they tend to be a lot worse. E.g. with Editorial Manager adding reviewers is really painful and the search is totally useless. The process to add and be added as a reviewer takes multiple messages in a quite confusing process. Decision letters are sent with bizarre template hacks that are really easy to mess up.

Some submission systems still require formatting of images etc to some arcane formats like tiff and eps. Or to have figures and captions as separate pages. These cause significant work for submitters and are a disaster to review.

Compared to typical enterprise systems they could be even worse, but I'd say still worse than average in that class too.

Don't know if these are a significant barrier. For me they have been and I've resigned as an editor due to the unusability. But I find that they easily cause enough loss in time and nerves of authors, reviewers and editors to be of negative value.

[michaelmior]

I will admit that MC is pretty annoying. Particularly the issue of having separate login/passwords for all journals. I haven't been in the position of Editor at this point, so I can't speak from that perspective.

I haven't yet run into submissions systems as arcane as those you mention in terms of format requirements, but that does indeed sound very messy and I hope I never see it.

[shaky-carrousel]

Peer reviewing could be made easier by public key cryptography. You could establish a web of trust between reviewers.

[jampekka]

I see no need for that. There are no problems with verifiying identities of reviewers.

[shaky-carrousel]

It could enable to have a decentralised web of trust of reviewers. In the same vein it is used to verify identities, you can use it to assess the reliability of a reviewer. A reviewer could sign another reviewer's key with the confidence it has about their reviewing.

[jampekka]

This is done via academic credentials, mostly by publishing in the same field.

I don't think there are any major technical challenges in publishing or peer review. It's not complicated at all. The problems are sociological.

[shaky-carrousel]

Well, you depend on a trustable publisher. With a web of trust you won't need that. People could publish anywhere.

[fanf2]

Peer review is done for free by academics. The publisher does not contribute to peer review.

[michaelmior]

The publisher does often assist with coordinating the peer review process which is a nontrivial task. I'm not saying this in itself necessitates everything else surrounding the publishing ecosystem, but I don't think it's accurate to say the publisher does not contribute.

[jampekka]

In my editor experience the assistance is minimal at best. Editors (almost all upaid) assess suitability of manuscripts, find the reviewers and handle all communication.