User talk:73.250.212.243

Dear 73.250.212.243, The fact that the NIST (at the time of publication of the Stephan 2019 paper) has two websites (please check the refs. 215 and 216 in the mentioned citation) with in parts contradictory statements is allready strongly misleading. One of the websites has been finally shut down today it seems. Moreover, it has been shown in Ref. 12 that several of the LJ data given and recommended by the NIST website is thermodynamically inconsistent. In particular, much more consistent and reliable reference LJ data is not listed by the NIST website. Therefore, stating or implying that the NIST website would be a reliable database for LJ simulation data is highly questionable. Also, your comment in your revision "(also see lack of data in supporting information of Ref. [12])" is unclear. What do you mean with lacking data in Ref. 12.

A problem of the data provided by the NIST website lies in its intransparancy. Several readers understand the given data in a way that it was determined by Errington (e.g. Thjis van Westen and Gross 2017) and it is only reprinted on the NIST website. This is in fact how the NIST website reads. However, comparing the data provided by the NIST website and that of Errington yields some differences. This can be seen from comparing the Errington data comprised in Ref. 12 (obtained from private communication with Errington himself with the authors from Ref.12) with the data provided by the NIST [at least at the time of the publication of Ref. 12]. Hence, the data was probably computed by the NIST people themselves. Yet, even this is unclear. Also several simulation details are not stated. This is probably also due to the fact that the simulation data in question was not published in a peer-reviewed process.

Overall, the statement that "Presently, no data repository covers and maintains this database (or any other model potential) – even data and results stated by the NIST website should be treated with caution (not reproducible and misleading referencing [12])." seems adequate to me: First, there is clearly no data repository that maintains thermophysical property data of the LJ substance. The data provided by the NIST website is not only extremely outdated and proven in parts thermodynamically inconsistent, but most importantly it is totally incomplete. It covers less than 5% of the presently available thermophysical property data of the LJ substance. This was clearly shwon in the Stephan et al. 2019 paper. There is no doubt, it is unfortunate that no institution takes care of that data. But this makes stating the absence of caretaking only more relevant. Secondly, as discussed above, there are severe problems with the NIST website data as it is. It was not published in a peer-reviewed process (on the contrary to the entire database provided in Ref. 12). The simulation protocol used by the NIST website people remains obscure. And the referencing used by the NIST website caused some severe conflicts in follow up work such that also other peoples work becomes not clearly reproducible (e.g. the work of Thjis van Westen and Gross 2017). This can be seen from comparing the Errington data provided by Errington to the authors of Ref. 12 (see Ref. 127 in Ref. 12) with the NIST website data. Third, very clearly, the LJ data provided and 'recommended' by NIST should be treated with caution as outlined in detail here.

Please be careful on making rather blunt changes without carefully checking and cross-checking the facts.

[12] Stephan et al. J. Chem. Inf. Mod. (2019)