Hacker News Reader

Comment by klodolph | Hacker News Reader

Word is one of the most common tools for writing papers. There was a study a while back that looked at the quality of Word or Latex papers. The study found that researchers using Word were more effective, and the working theory that explained it was that people who focused on the research did better research, and people focusing on the typesetting do better typesetting.

View on HN

Ironically, the paper itself was republished to fix figure placement [1].

To your main point, no, theoretical physics paper are almost all written in latex. I can't recall word-written theory paper. Experimental papers are sometimes word, but pretty rarely. You can try randomly sample papers from cond-mat arxiv to verify it.

[1] https://journals.plos.org/plosone/article?id=10.1371/journal...

> There was a study a while back

A couple of posts about how poor that paper was:

• https://lemire.me/blog/2015/01/14/knauff-and-nejasmic-recomm...

• https://blog.cr.yp.to/20201206-msword.html

Latex is a waste of human potential.

I've done professional typesetting and cataloging with QuarkXPress and InDesign both and it was extremely fast, that's for professional high quality publishing. .doc and .md is fine for information first publishing.

Latex is not simple and is in my opinion just a big nerd snipe, its the intrusive thought of layout software, for humanities sake we'd be better off with simpler tools like markdown + math notation and if latex had been never invented.

Reminds of "command line is faster than GUI" folks who get timed doing every task slower. "But it feels faster!"

Now repeat what you did

Every task? If I want to rename all of the .pdf files in the folder to .avi I would just do it in the command line

Or would you rather have me google a GUI that lets me "quickly rename" and download a few programs that have this capability? Or do a few hundred files by hand?

There are these things called programming/scripting languages, look it up. python for example. And you end up with very short and simple - and most importantly - human readable programms instead of cryptic Unix two letter commands with bunch of arcane single letter flags and other legacy pdp11 nonsense from 1970ties

The GUI for that is called file manager, there are quite a few available and they can be far more sophisticated than an ad-hoc command pipeline. If you manage your files on a regular basis (manual sorting, metadata, mass renaming etc), you already have a serious file manager that allows you to do this.

Command line is... fine for doing this occasionally, but it's hardly "quick" or usable for this, I'm saying that as an advanced commandline user. You're just using the tool that you already have and know, and it happens to be the system shell. It doesn't have to be.

(another question is that classic file management at the scale where a sophisticated tool is needed is mostly automated nowadays, and yes there's scripting as well)

I think it would be nice if everyone can decide for themselves which software they prefer. Some people prefer Latex, others prefer Word or InDesign etc.

I would prefer not to have gatekeeping either way, both "It was written using Word, it must be fake" and "Latex should never have been invented."

I'm not saying latex should be banned or disallowed anywhere, no actual gatekeeping in that regard.

There are big geek communities that lead newbies astray by recommending it, it's a nerd snipe that wastes a lot of brain cycles better used elsewhere. It's only little b bad, not big B bad.

I am saying it's a poor tool, a waste of time and an evolutionary dead end, people are allowed to fetishize poor tools, efficiency, simplicity and legibility are very poor in latex world with a high learning curve for a task that is at best tertiary to the task of doing real research, it's a tool that promotes rabbit trails, bike shedding and procrastination.

And the small amount of research that has looked into this has apparently born this out in at least some small degree. It's not a hot take if it's got backing.

I can believe that this is true today. As recently as 15 or 20 years ago, Word was awful enough that seeing through significant technical documents was a terrible experience. Horror stories of large corrupted documents and mysterious formatting behaviors were very common.

Today, Word is much more capable as a scientist's tool.

Reply on news.ycombinator.com