Hacker News Reader

Comment by darawk | Hacker News Reader

Yes, definitely. Is this even a discussion? Scientific journal companies extract enormous rents and add literally zero value. I'm not using 'literally' in the not-actually-literally sense that people sometimes use it. I mean literally zero value. They either don't pay, or barely pay, the peer researchers who review the material (the only valuable part). Aside from that, they do what, exactly? Publish it? The thing that costs effectively zero dollars to do on the internet now? It's disgusting that this business still exists.

View on HN

> Aside from that, they do what, exactly?

In CS the ACM organizes conferences... beyond that they are an effectively useless organization.

Ya, but the ACM is also a professional organization. That's a fine service to provide, but it doesn't need to be comingled with publishing.

A number of journals provide copy-editing and layout services. My own papers have been improved by both.

Why is this even relevant? Few people actually read published papers compared to, say, blogs, so the standard needn't be so high - inconveniencing 100 people is better than inconveniencing 1000. Yet nobody pays for copy editing and layout of their blog. It's not only trivial but also not important if it isn't quite right.

If readability is important for journal articles, then layout is the least important way to achieve that. The whole style of writing needs to be changed. They're notoriously hard to read even for other researchers in similar fields. I'm not saying dumb it down but, for example, perhaps they could be structured a bit like a newspaper story with a quick overview using common vocabulary first, then more details and more precise language later. That's just one idea but my general point is that readability is not what they're aiming for anyway so why bother making it look pretty?

Because it isn't about volume, and blogging is an inherently informal style in many cases (says someone who both publishes academic papers and has a blog).

As for readability, clarity of writing and visual clarity are not perfectly correlated. I have read very well-written papers that were still painful to interact with as physical objects.

This is already the case. Every scientific journal article I've seen starts with an abstract, which is exactly the summary you desire.

That's a start but doesn't go nearly far enough to relieve the burden of reading papers that researchers have to suffer through. An abstract usually has a high density jargon and undefined terms that are specific to the niche they're working in. What's easy to read is news articles that are based on scientific papers. The fact that journalists write articles instead of just copying and pasting the abstract shows that abstracts aren't very readable.

Also, once you're past the abstract, it's still heavy going. Compare that to text books which tend to be a bit easier to follow.

I'm not saying these are the answers. Just highlighting the problem of readability. That seems more serious than not having a space between a unit and a number, or a formula having a disproportionately big fraction. Even more important than spelling and grammar which humans can mostly correct as they go.

"The fact that journalists write articles instead of just copying and pasting the abstract shows that abstracts aren't very readable."

No, it shows that journalists write for a different audience.

"What's easy to read is news articles that are based on scientific papers."

That shows journalists wrote for a wider audience.

The articles that journalists write are considered better by laymen, but typically, they provide insufficient information for experts to see what exactly is new in the article and/or simplify so much that experts find the information presented downright wrong (search HN on battery or fusion break-throughs for examples)

But yes, it is true that quite a few scientists could do with a good editor that helps them write text that is easier to understand for experts. The current citation-happy culture doesn't help there, with abstracts that have more "(Newton, 1754)" phrases than actual text describing the work done.

What's easy to read is news articles that are based on scientific papers.

The reason they are so easy to read is that the journalists tend to simplify and generalize the results to such an abstract level that the 'research' they present in their article bares only passing resemblance to the actual research presented in the actual paper. And that's assuming a best case scenario where they don't simply misunderstand the whole thing.

All that being said, I agree that most papers could do with a good copy editor. Fortunately there are hundreds of freelancers willing and able to do that and many universities even have someone on staff or retainer that can help out.

Research articles are dense and full of jargon because that is the most efficient way to communicate with their intended audience, which is other researchers.

Making a standard research article readable to the lay audience, even if it is also educated to some degree in the field, would double or triple its length.

Fair. Likely the worlds most expensive copy-editing and layout services, though :) (considering the access price of the journals as the cost of those services).

On the other hand, in my experience the copy-editing process commonly makes things significantly worse, as I described here: https://news.ycombinator.com/item?id=8776429

There's an army of people on Fiverr and the like who would make your papers look fine, for very little money.

A few thoughts:

1. They don't have a standard journal style. I want my Lancet paper to look like a Lancet paper, not just "be fine".

2. I'd have to find them. The journals do it for me automatically on submission.

3. For a lab just starting out, there is no such thing as "very little money".

1. They can make it look however you like. They don't because they're not allowed into that particular market.

2. They're all over those sites. And the journal takes a lot, apparently.

3. A few tens of dollars? How many papers does an academic do in a year? They really can't afford a few hundred bucks?

I think if the publishers weren't in their near-monopoly position, money could be found for this. Heck, someone will probably write an AI that does the formatting in the right style soon.

There could be some hidden costs that the publishers incur, no? I'm just playing tenth man here.

I'm not the most knowledgeable about scientific journal companies, but hosting thousands of papers online/offline and having to vet all these individually may be quite an expensive process, right?

OTOH, the infamous Alexandre managed okay with Sci-Hub, but maybe it was because she didn't have to vet these?

In the end, seeing the profits these companies make I suppose nullifies any devil's advocate position here. Some of the upper-echelon positions in these journal companies are multi-billionaires @_@

So, the vetting process is actually extremely valuable. Refereeing papers is a super important service. It's just not one that these journals meaningfully provide. They either don't pay the peer reviewers, or they pay them something completely nominal. They do it because they want to, and for prestige. That is the only hard part of operating a scientific journal. The rest is just hosting and distributing static content. If you have a team of peer reviewers ready to go, you could literally setup your own scientific journal on WordPress for free, right now.

Reply on news.ycombinator.com