As I am involved in this, I think I should answer, too.
The real culprit of all this is that the question Use a quality glossary and acronym list in LyxUse a quality glossary and acronym list in Lyx is a form of an XY problem. The OP probably wanted to make the answer of How to combine Acronym and GlossaryHow to combine Acronym and Glossary work within LyX. He actually asked something different: instead of making the glossaries
approach work in LyX, he tried to use nomencl
, and get the first-use/subsequent-use scheme that glossaries
provides into that. Now, there are two approaches: answer what the OP should have asked, or answer what the OP actually asked. I usually chose the second, and so I did here, suggesting in the first paragraph of my answer that this might be the wrong way to address the problem at hand. You seem to chose the first approach, making a pure glossaries
solution work within LyX. And I think here started our misunderstanding:
You commented several times suggesting that my answer does not use the full glossaries
machinery. This is true, but fails to see the point of using the nomencl
machinery (\makenomenclature
... \nomenclature{...}
... \printnomenclature
). So to answer your first question: no, in my approach \makeglossaries
is not needed, as we don't write a glo
file, and won't read a gls
file. Similarly, the answer to your second question is that my answer did not mention how to create the nls
file, but a gls
file is not needed.
Your third question is the hardest one. I think we worked out in chat that my solution can work. I fully agree that it might be hard to use, in particular for new users that might prefer a more LyXy way. Does it make my answer "not useful"? That is up to the individual user, I guess.