3 ms·

I do research in physics. Typically we use git and latex. That's because it is easy to branch out and change the structure of the document, you can include the

by nagrom 15y ago

I do research in physics. Typically we use git and latex. That's because it is easy to branch out and change the structure of the document, you can include the scripts and data used to generate data in the same repository and there's a nice, easy-to-use front end in Github. You get most of the features of word, and the math looks better. And it's platform-agnostic! It doesn't matter whether you make that document on Windows, Mac or Linux, and it doesn't matter which software you use either. And cross-linking in the document is easy and a file-based database can be used to store the bibliography. Word is very hard to you use once you get used to that system...it's too inflexible and a complicated document gets mangled when crossing the OS boundary.

spindritf 15y ago
> I do research in physics. Typically we use git and latex. That's only convenient for an already existing research group. The usual workflow of incidental cooperation, like a course instructor sending additional materials to students, is to generate a pdf from the LaTeX documents they wrote or otherwise obtained and e-mail it.
zerostar07 15y ago
The fact is that it's much harder to train people who haven't had programming experience (e.g. biologists or medical doctors) to use this stuff. Word does the same thing pretty well (better imho, the reviews interface is intuitive), and has plenty of citation managers too
- _delirium 15y ago
  It's also close to impossible if the journal or conference doesn't give you a LaTeX stylesheet, which is uncommon outside physics/math/CS. Sure, you can in principle produce your own stylesheet conforming to the formatting requirements, but even most seasoned LaTeX users find the stylesheet language to be black magic (I dread having to make even minor edits to a stylesheet). Matching citation styles used to be a problem also, but tools like Jurabib and Biber have mostly solved that problem.

by nagrom 15y ago

spindritf 15y ago

> I do research in physics. Typically we use git and latex. That's only convenient for an already existing research group. The usual workflow of incidental cooperation, like a course instructor sending additional materials to students, is to generate a pdf from the LaTeX documents they wrote or otherwise obtained and e-mail it.

zerostar07 15y ago

The fact is that it's much harder to train people who haven't had programming experience (e.g. biologists or medical doctors) to use this stuff. Word does the same thing pretty well (better imho, the reviews interface is intuitive), and has plenty of citation managers too

_delirium 15y ago
It's also close to impossible if the journal or conference doesn't give you a LaTeX stylesheet, which is uncommon outside physics/math/CS. Sure, you can in principle produce your own stylesheet conforming to the formatting requirements, but even most seasoned LaTeX users find the stylesheet language to be black magic (I dread having to make even minor edits to a stylesheet). Matching citation styles used to be a problem also, but tools like Jurabib and Biber have mostly solved that problem.