<html><head><meta http-equiv="Content-Type" content="text/html; charset=us-ascii"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class=""><br class=""><div><br class=""><blockquote type="cite" class=""><div class="">On Mar 17, 2018, at 1:43 PM, Corinne Cath <<a href="mailto:corinnecath@gmail.com" class="">corinnecath@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div class="gmail_default" style="font-size: 11px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; font-family: verdana, sans-serif;">In terms of overarching questions, I would be interested in learning more about how issues like sexism, racism, ableism play out on the mailinglists, as I heard a lot about that from my interviews. Its hard, but not impossible to do sentiment analysis for those kinds of conversation. I have reached out to a friend who does that for reddit forums to see if her code is on git.</div></div></blockquote></div><br class=""><div class="">I would certainly be interested in this if it's something that we can implement with some reasonable measure.</div><div class=""><br class=""></div><div class="">One cute idea I had in mind was calculating the differences in the use of "thanks" and "actually" in the email messages, as there was a paper that defined that as a distinguishing metric between certain prominent Linux developers, and it seemed like a nice heuristic for welcoming vs. antagonistic messaging. </div><div class=""><br class=""></div><div class="">If there are more significant or useful sentiment analysis tools that we can evaluate, I think that would be useful. It would be a nice input to the gender estimation work, to see if different behaviors on the mailing list are correlated. It might also help us identify which groups are involved in controversy, flaming, dysfunction, etc.</div><div class=""><br class=""></div><div class="">Corinne, if you hear back from your contact, or if others have any suggestions on sentiment analysis we can easily integrate, let us know.</div><div class=""><br class=""></div><div class="">Cheers,</div><div class="">Nick</div></body></html>