Ask HN: What added value do academic publishers provide? | Hacker News Reader

Ask HN: What added value do academic publishers provide?

8 pointsp0llard11 comments

As per the title; this is something I've wondered about for quite a while, but the recent news about MIT and Elsevier (and UC a while ago, although it looks like negotiations have started anew there) reminded me. It seems to be to be the case that they don't, especially now that most journals/proceedings are published solely in digital format; there would seem to be literally zero cost or skill involved in watermarking a PDF and hosting it online, heck, the arXiv does this for free, and it seems that more and more this is all academic publishers are doing. If reviewers were being paid then that might be one explanation, but they aren't.

For textbooks, etc., I can see that there is some cost involved in providing proof reading services, but the cost of academic textbooks is clearly far too high if this is the only service being provided; the existence of freelance proof readers and well established print-on-demand services leaves me at a loss as to why academic publishers still exist at all. But since they do, and universities/institutions haven't banded into a cartel to run their own publishing, it would seem that there must be some reason that the publishers are still in business; it is just down to the corruption that seems to be rife at the very top ranks of academia (publishers giving kickbacks to chancellors), or is there a valid reason for this?

The good ones are indeed pretty good filters. I'm not smart enough nor free enough to verify that a study is good, is important, is free of errors -- and so I benefit from the service provided by, say, Nature Photonics in knowing that whatever I'm reading is important, significant in some manner, and interesting.

Good journals have a high bar for writing and diagrams. In order to understand things, it helps that they're communicated clearly. You're likely not going to find incomprehensible poorly written text or hard-to-understand figures in Science or Cell journal.

As an example, I encourage you right now to go to https://science.sciencemag.org/ to see their coverage of covid-19. You'll get better cutting-edge information and perspective on the issue than anywhere else.

Ah, but that distinction was made by the reviewers, who are (in the overwhelmingly vast majority of cases) unpaid; the publisher itself isn't providing that as a service, at least as far as I understand.

Frankly the only problem I see in this grand situation is the part about reviewing.

But what I see as being the problem is something different: my problem is that it's rigged. I.e., do you know that when you submit to Nature, you can specify to whom your paper should and should not be sent to for review? This encourages back-hand deals. Invariably, one tends to suggest their old pals, and naturally those pals show the journal two glowing thumbs being up and tell them to send it straight to the presses, with a cute little remark that this or that figure should maybe be made more clear.

I remember talking to a colleague about this about 5 years ago, the reviewers should not be anonymous, for various reasons: a) they should be recognized for their diligent work of verifying things, b) there should be some motivation to play a straight game, some element of accountability: perhaps the knowledge that you stand as the person who okay'd something that might prove later to be suspect is good enough motivation to play it clean.

And as it happens, some big journals are beginning to try this policy out! I hope and think this takes over in most places.

> I.e., do you know that when you submit to Nature, you can specify to whom your paper should and should not be sent to for review?

It's the same in e.g. ACM conferences; indeed there is a developing scandal surrounding organised fraud in ACM/IEEE conferences (conferences are more prestigious than journals in CS, unlike other fields) facilitated by this very mechanism.

The reviewers themselves are not really anonymous are they? In general I think it's usually pretty well known who the reviewers for a given journal/conference are, even if this isn't explicitly published. The papers themselves are usually anonymised, but that's a different matter, no?

I recently saw a comment here about this. I don't remember the specifics, but basically the argument was based on trustworthy long-term hosting of obscure material.

Can universities not manage that themselves? Archival isn't a new problem and I don't believe it's a problem that should be tackled by a profit making company; if there were a non-profit responsible for safe-guarding the collective knowledge of humanity that would be one thing, but I'm pretty sure publishers aren't treating this as their aim.

This discussion came up in thread about the internet archive.

It is probably just inertia from before the internet.

When viewed through the lens of maintaining accessibility for hundreds of years or more, I am comfortable with slow change.

The question is how much do you trust the publishers to do this?

They find the volunteers, coordinate them, figure out and handle the paperwork, and politely hassle them until they do their work. Basically, a similar function to university administrators, mid-level government employees, corporate project managers, and non-profit directors.

Yes sure, but they also command an enormous premium for doing so; perhaps I'm too naive, but I would hope that in a "rational market" this wouldn't be possible. I would expect universities to be running their own publishing entirely to avoid paying for Elsevier's CEO's lifestyle.

Some fields already do have their own publishing orgs, like ACM and IEEE. I don't think I've published anything of note in one of the standard publishers in a while. The field-specific organizations seem to be notably better, but are more limited in scope: https://libraries.acm.org/subscriptions-access/academic/dl-p...

Part of it is that I imagine a lot of this work is typical administrative office work, and not many people are excited to do it. You're probably not disrupting anything, things will be more complex than you expect (as is anything that operates and charges people internationally), and you won't be seen as a hero (probably more like the opposite).

Yep, I'm a computer scientist and the I'd say probably the majority of the papers I read are from ACM conference proceedings, weirdly the journal/conference prestige split seems to be reversed in CS compared to other disciplines.

Outside CS however it does seem that traditional journals (published by Elsevier, Springer, etc.) are still very much de rigueur.

Reply on news.ycombinator.com