search

IACR / latex

Latex classes for IACR publications. We will start with the new journal proposal.
9 stars 2 forks source link

We should check for natbib and forbid it #213

Closed kmccurley closed 11 months ago

kmccurley commented 12 months ago

The natbib package interferes with how bibliography styles work, and we should forbid it. At least one editor has complained that natbib messes up Springer style.

I found one paper (crypto 2023 #206) that uses natbib. In this case the author just wanted to use \citet to expand the author name prior to the reference. Thus, when the author enters However, as observed by \citet{EC:HirZik10} the result looks like

However, as observed by Hirt and Zikas [58],

I believe this is caused in part because Springer uses the widely-hated numeric style. The alphaurl style that we use would give HK10, which is at least a hint as to who the authors are. if the authors want to expand the names on a reference beyond this, then they can write it out in full or define their own \citet macro. Note that the authors don't even use natbib properly because the documentation for natbib says to avoid using the standard \cite but authors mix it with \citet to occasionally expand the author names.

I looked at other papers uploaded for LNCS volumes and it appears to be fairly rare to use natbib (one paper each from Crypto 2023, Eurocrypt 2022, and Eurocrypt 2023). Nobody uses \citep but a few used \citet. All of these authors appear to mix \cite with \citet in spite of the instructions from natbib.

kmccurley commented 11 months ago

Perhaps we should reconsider this based on the fact that acmart supports it. We should discuss this together to evaluate the tradeoffs.

jwbos commented 11 months ago

Indeed, I first would like to understand the drawbacks and if this indeed changing the bibtex style.

kmccurley commented 11 months ago

After reading up about it, I think ACM supports so many different journals and conferences that they need to support multiple bibliographic styles. I don't think we need to support the numeric style at all, and that means natbib is pretty irrelevant.

jwbos commented 11 months ago

I started reading up on natbib as well and don't understand / see the benefits. I agree that we don't need to support this in our style.

kmccurley commented 11 months ago

I have a pull request to forbid loading natbib (along with a test).