Can you confirm that you were using Houdini 11 here: “plain old particle fluid surface node Flip workflow”? If so, the reason they ditched the popnet is simple: it's slow, and the “ease of use” was not worth it. You can always add a popnet back in if that's what you want to use. For emission, like pyro, they switched to a more sops based workflow, where you build your volumes and points in sops, then source them in the dop network. The new sourcing workflow is great, and a huge improvement over what they had in h11. Peter Quint has some videos on vimeo going over the pyro stuff in Houdini 12 that should get you up to speed (the same nodes are used for flip).
So, for flip, instead of popnets, it's better to use a sop solver with vops or to use fields to direct your simulations. It's faster, and just as versatile, albeit a bit more manual work the first time you go through it. You can use shelf tools for the majority of the work you'll do, and then you can tweak as needed. I never have to use vex (by typing) and usually stick with vops when I need, and if it's something useful, I'll save a preset or otl out of it. I'll also use nodes like gas linear combination if I need to mix fields, but you can also use it on geometry attributes too.
The difference between 11 and 12 was big, but between 12 and 12.5 it was pretty minimal in terms of the workflow. The meshing is different (openvdb) and not mentioned in the docs anywhere, but that's about it, and the particle fluid surface node is still there (and much faster), so you can always use that too. As for docs, I think these all do a pretty good job…
http://www.sidefx.com/docs/houdini12.0/dyno/liquids [
sidefx.com] see “Particle fluid networks” – this is basically what I described above.
http://www.sidefx.com/docs/houdini12.5/dyno/liquids [
sidefx.com] << I think the page is identical for 12.5
http://www.sidefx.com/docs/houdini12.5/shelf/wavetank [
sidefx.com] << specifically has Understanding the network of nodes, which mentions the fluidtank_interior node.