@NikeDattani We just had a massive influx of questions of varying quality. Is this an acceptable thing or should we close a few?

@TristanMaxson The influx was because we just hit 180 days in Beta: area51.stackexchange.com/proposals/122958/matter-modeling, which means 6 months, which means last chance to fulfill commitment. User 2ndquantized was successful, but Bernhard's influx of questions was a bit too late (you can see nothing under "current commitments" and "fulfilled commitments").

@TristanMaxson As for closing: are there any specific ones you had in mind?

See this user's questions from the last hour. mattermodeling.stackexchange.com/users/1195/ashique-lal

Most questions are 1-2 sentences with no focus.

@NikeDattani Also to comment on your statement from the other day, I think it's still best to be careful not to reformulate a user's question just because it is a HNQ. I think the question strayed quite a bit from the original question, but the asker did not seem to care too much. In other cases, it might be more problematic.

I would prefer in these cases, asking the user to edit things instead.

@TristanMaxson I do this all the time to make questions become more friendly to the HNQ algorithm. You can look here and see my edits on a lot of these posts: chat.stackexchange.com/rooms/109935/hot-network-questions-. I've changed questions far more severely than I did it in this case.

Not only did the user not care, but they "accepted" my answer, which answers not just their specific concern about what "level of theory" means for B3LYP/6-31G, but also what "level of theory" means in general, which included what it means for their specific B3LYP/6-31G, in my table of examples.

@TristanMaxson Ideally user's would edit things themselves, I agree. But the HNQ algorithm is extremely time-sensitive, dropping the HNQ score exponentially as 1/(1+hours)^1.4. We can't rely on people to respond to comments. I appreciate what you're saying in spirit, but I've made 1000+ edits: mattermodeling.stackexchange.com/… and no one once complained about it.

Until you :)

If others complain about it we can re-visit this.

It is likely more of a problem with SE then making HNQ so time sensitive, in which case I am not too upset by it. Honestly, most problems stem from SE being in maintenance mode and nothing ever really changing.

mattermodeling.stackexchange.com/questions/2226/… - This is something I've wondered about too. What insight could we get from a quick analysis of the charge densities that DFT codes output? This kind of analysis could be of help in deciding what to calculate next or even what direction a project could go forward. It would be great if @Jack would post this on DFT groups at places like Facebook.

Posted the wrong link before. I meant this one - mattermodeling.stackexchange.com/questions/2482/…

@HitanshuSachania I haven't seen Jack in this chat room, so he's unlikely to see this message.

Yea, and he has no contact info on his page.

@HitanshuSachania I've invited him now to the VASP chat room. VASP is his 2nd top tag. Your question to Jack is about the specific science right? Not about "Matter Modeling Meta" which is about community discussion about the site itself?

Yes, that is perfect. Btw, how did you invite him? Can anyone invite other users?

In my experience charge density analysis falls under one of two categories, either visual analysis which tends to be subtraction / addition of densities (such as isolated adsorbate vs adsorbate on surface) or post processing methods such as AIM.

I find many people are skeptical of the visual analysis methods though so be aware of that.

@NikeDattani SE, in general, has a tremendous scope to improve. Chat rooms and comments could use a complete overhaul. They are quite user-unfriendly😅.

Infact, in the example on that question specifically, how do you know that at slightly different isosurface values / DFT+U that both do not act similarly. I have avoided answering that question because I think I am too skeptical of the analysis suggested in that question.

@TristanMaxson Absolutely, though I don't generally come across any visualization in terms of charge density in papers, then again I've barely started in this field 2 years ago.

@TristanMaxson You could answer that question and include your doubt about the example in it. I've seen popular posts on stack overflow that have answers with uncertainty.

@HitanshuSachania You will occasionally see it, but normally it is a pain to understand what is actually occurring. Electrostatic potential maps are actually one form of this type of analysis, but with extra steps. Those tend to be seen a bit more often.

@TristanMaxson interesting.

The other really common one is the bader charge analysis, but I think that doesn't fit the spirit of the question. I may write an answer about electrostatic potential maps if nobody else does, I am not really an expert by any means in them.

@TristanMaxson now that I think of it, a post doc in my lab was looking to use charge densities to study how well does the adsorbate atom attaches to an oxide surface in a single-atom catalyst.

Quantifying it tends to be very subjective.

Often you look for a buildup of charge density between the adsorbate and surface and see if there are any non-local effects as well.

@TristanMaxson I guess Jack expects each answer to focus on just one thing, so Bader charge analysis could be an answer exclusively.

But everytime I have seen it personally, someone has raised concern that their functional is not likely to be fit to reproduce the correct behavior (even tho functionals technically should be fit to model the right electron density at least)

@TristanMaxson if they aren't fit to reproducing the correct behaviour, it sort of also means those functionals are no good for relaxation or and DFT calculation as such, isn't it? Afterall, electron charge density is all that matters principally.

I agree, but the argument is hard to defend since electron density is normally not what is fit directly

This room quickly became the second most used room after "Modeling Matters", and it's newer than all the other ones!