[jltsiren]

You normally submit a LaTeX or Word document, and the publisher does the final typesetting. Even in computer science, where people often spend a lot of time tweaking the typesetting, the pdf generated by the authors is essentially a preview. There are often visible differences between it and the publisher's version.

[IshKebab]

Yeah this is one of the craziest things about the scientific publishing industry.

Journals justify their fees by claiming its for typesetting, but all they are really doing is adding extra work to nit pick bibliography formats and so on (see the comments in this article about sentence case). Nobody cares about that. I don't think anyone even reads "journals" any more (except maybe Nature/Science etc.). They mostly just read individual papers and then there's no consistency to maintain.

In a sane world journals would accept PDFs. They would check that the format roughly matches what they expect but not insist on doing the type setting themselves.

Oh well, maybe one day.

[aragilar]

I would note arXiv requires the source as well, and having the source is what is enabling the HTML experiments they're doing.

On consistency, what the journals provide is some level of QA (how much is a function of field and journal, rather than the what is charges), and the template is the journal's brand, so both the authors and journals benefit from the style (I can tell the difference between the different (all similar quality) journals in my field at a glance by the style).

It's also worth noting that there's a whole much of metadata that needs to be collected (whether you agree with it or not, funders require it), so a PDF isn't going to cut it here either.

[jonathanstrange]

Citation and bibliography guidelines are by far the things that authors neglect the most, and they are absolutely essential to ensure quality.

Using PDF as an input format would make editing and typesetting practically impossible. Not that I haven't seen volumes where publishers did that but the results are abysmal and in my experience that only occurred with local "grey literature" like really crappy conference proceedings edited in an institute.

[IshKebab]

> Using PDF as an input format would make editing and typesetting practically impossible.

But they don't do any editing or typesetting. They say "use our template" and "the author's second initial needs to be italic in citations". That's my whole point.

[colejohnson66]

> Journals justify their fees by claiming its for typesetting [..]

Because they used to actually be doing that. Historically, science journals were pay-to-play because the journal had to typeset your document and print it. But with the advent of computers, they had to pivot while still retaining their revenue streams.