Good to know! Thanks for the update too!

I think part of it is the relative lack of people. But I also think that things go trending more on Twitter, which is both good and bad. The bad is obvious. But the good is that it created unifying experiences for academic Twitter to nucleate around to make it feel part of the same community.

I guess where I'm confused is that the M06-2X geometries will not be minima or TS with wB97M-V, which was used for model training. So, some small amount of strain could have a large impact on things. Maybe I'm thinking about this the wrong way?

(I'm not implying the models will do well after this, to be clear)

Is this a fair comparison though if you are doing a static calculation on a structure from a different level of theory than what was used for the MLIP? I think it'd be more useful to either have the same level of theory or to do the MLIP optimizations here.

I have just been sending this to people. You know, from the last time he pulled this shit www.funraniumlabs.com/2017/03/want...

😅 I feel this...

I agree it's a terrible idea, but just a note that this plugin is by a third party -- not Google Scholar.

😂 That will do it! I like your approach.

(I'm not sure how I've gotten this far, I must admit!)

It continues to be my opinion that if the journals take our work for free, then I should be able to give it to them in whatever (reasonable) format I'd like and let them deal with the rest. Of course, no matter how nicely formatted it is, they often still butcher it come time for the proofs. 😭

Thanks. It has been escalated for a month, so I'm not sure there is room for further elevation but I appreciate you looking into this.

With their most famous application being separations... it is hard to argue with that! Doesn't get more engineering than separations.