opencompl / paper-template Goto Github PK

This is a bit hard to debug atm, a warning might be nice if possible

In this template, we basically write all the text in the paper.tex file. This has the downside that one has to skim through a lot of text when wanting to find a particular section to work on.

In my thesis template, I had a folder Chapters that contained single files for each section (i.e. 01-intro.tex, 06-conclusion.tex), which all were imported into the equivalent of paper.tex in this template. Not sure how you feel about it, but to me, this structure is easier to work on.

Tagging some people to get their opinions: @tobiasgrosser @bollu @math-fehr @Superty @Dinistro

This was suggested by @math-fehr on today's meeting.

See for example:

On each commit, we spend 1:30 to pull the latex-action. I wonder if a self-hosted builder on a cheap google cloud machine would allow us to cache the docker image such that this time would drop to instant?

Google provides e.g., for 5 USD/month a micro instance, which would give us unlimed builds for a month.

https://docs.github.com/en/actions/hosting-your-own-runners/adding-self-hosted-runners

I really dislike multi-file tex projects as this makes search and replace and jumping through files really painful. It would be excellent if we could specify the abstract again at a single location in paper.tex

I saw in the Makefile that there is a target for CameraIEEE but what about CameraACM?

I am wondering if we can find ways to reduce this noise without adding even more abstractions. The more complex this repo becomes the more likely things break and the slower we can work with things. We use the ACM templates a lot more often than the IEEE templates so the common case we should optimize for is ACM. Even having some manual porting efforts for the IEEE template seems to be reasonable for me.