Should all equations which appear in a thesis be numbered? I was just wondering if there is any sort of consensus on the topic of when to number math expressions. 
For example different lines in a proof, these should be tagged or not tagged?
 A: I suggest only numbering expressions which will be referred to later in the paper.  If "for future discussion" means a later paper, I think referencing equations in earlier papers solely via a number is a pretty good way to seriously discourage readership.
A: Unless your department or university requires it, there is no need to number all equations. In fact I have not seen any dissertation where all equations are numbered.
A: One can also be of the opinion that all equations should be numbered, for easier future discussion and references.
Probably best to make this Community Wiki and let the Yes and No answers accumulate votes by popularity.
A: In these days of electronic publications there should an option: of the equations which author is willing to number, some should have an explicit number, and other numbers should be hidden, and seen only when a reader makes a request (e.g. by pointing your cursor to the equation).
This way you can have it both ways. You may number even every equation in your paper, while only some of them will be visible directly, and the rest after taking an action.
