4
$\begingroup$

I searched through all 296 "materials" questions on the Chem SE: https://chemistry.stackexchange.com/questions/tagged/materials, and looked for questions that also had any of the "computational" or "theory" or "theoretical" tags. In total there was only one!

I then did the same on the Physics SE and found that there was again, only one question that had both a materials tag and a computational tag.

It therefore appears that Materials Modeling questions do not get asked on either SE. A huge number of "computational chemistry" and a huge number of "computational physics" questions get asked on each SE, but it is surprising that not so many get asked for "chemical physics" of materials (questions about chemical physics of gas-phase systems do get asked a lot though, and we can find them by searching the "ab-initio" or "quantum chemistry" tags).

Furthermore, there's 1007 materials modeling questions on Research Gate just on using the software VASP: https://www.researchgate.net/topic/VASP, and 15000+ questions on the VASP forum itself: https://cms.mpi.univie.ac.at/vasp-forum/forum.php, and according to Facebook, there's an average of 2 posts/week on the VASP facebook group: https://www.facebook.com/groups/Vasp.Code/

Then a similar story can be told for other materials modeling software like QuantumEspresso, CPMD, CP2K, ABINT, CASTEP, DFTB+, and LAMPPS.

Combining the user forums for all of these programs into one stack exchange, there could be a lot of people asking about the modeling of the chemical physics of materials. Perhaps other chemical physics software such as CFOUR, MOLPRO, DIRAC, ORCA, MRCC, ADF, GAUSSIAN, PSI4, DALTON, and MOLCAS may also have their very active forum users start using Stack Exchange more and more.

So I wonder what people think is the reason why the Chemistry and Physics SEs have only 1 question each, with a materials tag plus a computational or theory tag, when there's tens of thousands of people asking questions on the forum for just one of the materials modeling software packages?

Is there a different stack exchange where such questions get asked more often (such as some sort of Engineering SE)? If there is not, I wonder if people here might consider giving input on what the questions should look like on a Materials Modeling SE: https://area51.stackexchange.com/proposals/122958/materials-modeling

| |
$\endgroup$
  • 3
    $\begingroup$ There is Computational Science. Most questions I've seen on Physics or Chemistry about modeling are either 'what is wrong with my code' or 'how do I get package X to do Y', neither one of which has either Physics or Chemistry as the core problem. Those things should be asked elsewhere (I mean, if there is a forum for VASP, why would you not ask something about VASP there?). $\endgroup$ – Jon Custer Aug 20 '19 at 12:41
  • $\begingroup$ @JonCuster: I have read and considered your comment. I wonder if you would mind taking a look at the questions on the Materials Modeling proposal? I don't think they resemble the "computational science" SE very much at all. I looked at SciComp, and all 50 questions on the first page are about numerical algorithms or applied math or solving differential equations or FFT. I don't think the top questions proposed on MMSE so far would survive at all on SciComp! $\endgroup$ – user1271772 Aug 22 '19 at 21:04
  • 2
    $\begingroup$ @JonCuster: As for the VASP forum. (1) What if I have a question about quasiparticle simulation or DFPT, and can't get the answer I'm looking for from the VASP forum because it turns out that (without me knowing), VASP doesn't actually do those things but Quantum Espresso (QE) does? This is why we need a general place where users of all software can see the question. (2) The VASP forum isn't highly regarded and people say that QE's mailing list is one of the huge advantages of QE. (3) People don't get points on the VASP forum for doing a good job helping people. (4) The VASP website could go $\endgroup$ – user1271772 Aug 22 '19 at 21:10
  • 1
    $\begingroup$ ...down anytime, whereas the SE network is a lot more stable! So the MMSE can be a place where anyone with questions about materials modeling in general, could go and get answers from users of QuantumEspresso, CPMD, CP2K, ABINT, CASTEP, DFTB+, LAMPPS, and VASP (whoever sees it first or is best suited to give the best answer!). It's better than going to each of those forums, registering an account in each place, and asking in 10 different places! $\endgroup$ – user1271772 Aug 22 '19 at 21:12
  • $\begingroup$ I think a lot of the questions on the material modeling proposal would fit well here or on Physics. As to why they aren't asked here or there frequently, I'm not sure, but the fact that they aren't raises concerns about how much interest a new site in the SE network would actually get. I think it would be easier to just promote chem as a place where these questions could be asked. $\endgroup$ – Tyberius Sep 15 '19 at 4:27
  • $\begingroup$ @Tyberius: (1) many of the questions currently proposed would not fit on the Chem.SE. Many would not fit on Phys.SE. Some of the questions would probably not be well received on either place (or on CompSci or SO or Mathematics). (2) You seem concerned about a potential lack of interest since there's been almost no "materials computation" questions on Chem or Physics. But as a follower of several mailing lists for materials modeling, I see questions being asked all the time, they are just asked in a less sophisticated platform (no scoring system, no moderation, no unified site for... $\endgroup$ – user1271772 Sep 16 '19 at 18:34
  • $\begingroup$ ... questions from all of those mailing lists. (3) About promoting Chem to ask these questions: Unfortunately most of these questions are more for Physics than for Chem, but an even bigger problem is that Chem and Phys are VERY Broad SE's which contain so much that is not at all of interest to me or my colleagues. I never visit the Chem or Phys SEs because the first 100 questions usualy see are either about some homework problem, or about some physics I have no idea about (gravity, string theory, medical imaging, etc.) or chemistry I don't know about (reactions, organic, analytical, kinetics). $\endgroup$ – user1271772 Sep 16 '19 at 18:38
  • 1
    $\begingroup$ @user1271772 From my perspective, the current top 5 proposed questions on the area51 proposal would all be on topic here, along with a number of others. But I see your point about materials modeling potentially being worth separating so as not to sort through the unrelated parts of chem, phys, comp-sci, etc. As someone who does computational research, I can see why having a collected SE for this would be better than having to pore over a bunch of different mailing lists. I'll be following how it progresses. $\endgroup$ – Tyberius Sep 17 '19 at 17:52
  • 1
    $\begingroup$ @Tyberius: I noticed you followed the proposal earlier today. Thank you so much for your support !!!! I do agree that the top 5 questions (4 of them are mine) might not be "closed" here on Chem.SE, but they are all about solid-state material structure rather than individual molecules or reactions. Actually I have to change my mind about that now! I just searched "plane-wave" in Chem.SE and the first answer was closed, and the comments tell the user to go to Physics.SE or Mathematics.SE: chemistry.stackexchange.com/questions/14951/… $\endgroup$ – user1271772 Sep 17 '19 at 20:39
5
$\begingroup$

Now there is: Materials Modeling Stackexchange

| |
$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .