Making the software tag work for you

Question

While looking through our current slate of tags, the software tag stuck out to me. Its our second most used tag, but that seems to because its serving a couple different purposes:

1. To label questions that use any Matter Modeling software (e.g. How to convert a molecular structure made in Avogadro to a LAMMPS coordinate file?, MM-GBSA from Induced Fit Docking on Maestro)
2. To look for software that serves some function (e.g. Molecular Dynamics simulation software for Windows, Plotting of natural bond orbitals (NBOs), Software to draw 3d micelle configuration, Is there a repository for DFT pre/postprocessing scripts?)
3. Advice/insight on the development of Matter Modeling software (e.g. Since MKL is not optimized for AMD hardware, should I use a math library specific to AMD, or would an open-source one be just as good?, Did the 2019 discovery of O(N log(N)) multiplication have a practical outcome?, What is a good programming language for matter modeling?)

My view on these is that [1] is probably an overly general use of the tag (can just include tags for the specific program(s), while [2] and [3] are different enough to warrant separate tag (perhaps software-recommendations or just software for [2] and software-development for [3]).

What are your thoughts on this? Does the software tag match your expectations the way it is currently or would you likely to see it broken into more specific tags (not necessarily the ones I recommended)?

Answer 1

My view on these is that [1] is probably an overly general use of the tag (can just include tags for the specific program(s), while [2] and [3] are different enough to warrant separate tag (perhaps software-recommendations or just software for [2] and software-development for [3]).

I agree, [1] is overly general, [2] should be renamed software-recommendation and [3] should be renamed software-development (not software). In particular this provides a nice and easy dichotomy between software that's already developed and software which you want to write.

Good job on identifying that distinction!

if you agree and want to make these changes upvote my answer so we know to make that change

Answer 2

I think the people that this affects the most, are the watchers of the tag, and the people subscribed to the tag to get emails whenever a new question is asked with that tag.

Let's see what they say before doing a rather large re-tagging procedure.

Later it will also be the people/companies sponsoring a particular tag assuming we want to help SE gain some revenue and therefore make it more likely for our site to stay (and that we may want to benefit from whatever the sponsors provide the users, which in the case of QuantumComputing.SE was just t-shirts, but could presumably be more from a different company).

My personal feeling right now is that most of our tags are rather small, and while the DFT and software ones are the biggest, they are no where near as big as the tags you see on StackOverflow, Unix/Linux, SuperUser, etc.

Maybe we can start with someone who dislikes the present tagging convention, to list out all the ways in which it's harmful. In both Tyberius's question and Cody's answer, it's suggested that it's "overly general" but what specifically is the harm in that?

In my opinion, a good reason to change a current tagging convention is that the subscribers to the tag are receiving way too many emails/day and 66% of them are not interesting to them, so they would like the tag to be split into 3 different sub-tags. However there might be subscribers/sponsors that prefer to have all three of them falling under one tag, for easier categorization. For these reasons I think we should hear from the subscribers and watchers before being too quick to change things. Maybe we don't yet even have enough. I have been avoiding putting too much thought into these things because it's still very very early days and we don't know what the site is going to look like 1 year from now. We might make a decision now and then regret it later and it will be too hard (or much harder) to switch back. For these reasons I've been focusing more on our other big list of things to do which includes addressing the unanswered queue to reach the A51 recommendation of 90% answered, and a lot of advertising to bring new users into the site while it's still new and fresh (rather than old and established and not appealing to join late).