Which side is all about censoring curse words?
Only
harmless curse words. If it's an ethnic, racial, sexual, or religious slur it's fine. Not only is that kind of word fine, but it's even against the 1st Amendment for someone to react negatively to you saying it.
Ehe... just ask Chinese dissidents how censoring words work. People switch to eufemisms. And the the eufemism becomes the new bad word. The Chinese government then censor those words. This is why 18th century newspapers are so hard to understand today. In order to avoid banned words people had to dance around the topic and implying, without saying. But did that stop those newspapers? Nope.
This idea of helping groups by avoiding bad words is well suited for our modern Internet age. It's a super easy way for individuals to quickly identify the people who are evil online. But it's an illusion. You're not helping anyone more than thoughts and prayers ever did. Instead it makes communicating convoluted and only hampers communication. Since every group now has specialized ingroup language (you need to master or be seen as evil) it drives groups apart.
We need to keep repeating the mantra "it's ok to be offended". It is ok to be offended and it is ok to offend. Not only does the Internet allow quick global communication. But it also allows us not to see things we don't want to see. Nobody needs to see anything they don't want to.
Researchers love counting keywords because it's quantifiable. But what does it prove? A computer scientist friend of mine was on a research project financed by the Swedish government to track homophobia through keywords (he was himself gay). They tried correlating the use of homophobic slurs online and violence against gays. They did this Internationally. I only know of this because I was talking to him throughout this process. I don't know if it got published. But what he told me is that whatever metric used for homophobic slurs used in a culture does not lead to to anti gay violence in any way that is comparable between cultures, or even within cultures. It was highly specific and regional. Which suggests that it's more complicated than that a straight correlation can be used.