The MP team has made some changes on our backend so as to enable a lot of additional data (the site now has 20% more materials than an hour ago!) and greater ease extending our data sets going forward. Because we’ve changed a lot, there are bound to be bugs that we simply missed in our extensive testing. We would like you to find these and post them publicly so that we can fix them and the community can hold us to the high standard we all want out of this project.
Besides additional data, have there been changes to existing data? Last I checked (April 2018) the band gaps for mp-867108 and mp-861660 were 172 meV and 14 meV, but they both are now 0.
For both of those cases, we previously only had band gaps from static calculations. The new data set has some additional calculations, including more calculations along the key symmetry lines in reciprocal space. This is a better type of calculation for determining band gap and for both of these materials those calculations have determined that the gap closes somewhere along the high symmetry paths. We’re currently troubleshooting an issue where our build system is plotting some bandstructures, but we will fix that soon and you should be able to see the corresponding bandstructure.