Some remarks:

If you really prefer using variables in subscripts, please note that something like (A

_{x}, B

_{1-x}) is wrong, because (1-x) B cannot replace x A - except x = b/(a+b) (where a and b are the charges of A and B, respectively), in which case the use of a variable does obviously not make any sense, or the charge is balanced by coupled substitution, in which case you'll have to obscure the formula further by introducing even more variables.

In other words: if you are using variable subscripts, you have to omit the comma: (A

_{x}B

_{1-x}) is correct. Note however, that this has exactly the same meaning as (A, B).

Now, if you still want to keep the variables, please also note:

(1) There is no need to keep the brackets in an expression such as (A

_{x}B

_{1-x}). In order to avoid confusion, the brackets should be omitted - unless you want to indicate that A and B are occupying a specific site, in which case they are better replaced by curly brackets though.

(2) You have to define x unambiguously, as this is not self-explaining. This condition (here, 0 <= x < 0.5) is an integral part of the formula.

Now, let's have a closer look at Rui's last proposal. Note that I have added the missing condition for x and also left out the comma, for the reasons explained above:

Li

_{1-x}(Fe

^{3+}_{x}Mn

^{2+}_{1-x}) (PO

_{4}), 0 <= x < 0.5

This can be modified by applying simple mathematical logics (in fact, the following transformations don't have anything to do with chemistry):

First, the brackets can be omitted:

Li

_{1-x}Fe

^{3+}_{x}Mn

^{2+}_{1-x} (PO

_{4}), 0 <= x < 0.5

Note that I keep the bracket for the anion. This is to make clear that "PO4" is a single anion. Now, some of you might want to add that this is obvious anyway. Note however, that it is much less obvious when the mineral contains amphoteric elements, i.e. elements that can (and often do) occur both as cations and as constituents of oxoanions: in those cases, the bracket must be used in order to make clear what is meant. Instead of using the brackets when we (the specialists) think they are needed and leaving them out when we (the specialists) think they are not, we have decided to use them consistently.

Second, lithium and manganese have the same subscript and can be combined. Note that I need brackets now in order to indicate that the subscript applies for two cations (those within the brackets) and not only one (again, this is a mathematical convention, not a chemical one):

Fe

^{3+}_{x}(LiMn

^{2+})

_{1-x} (PO

_{4}), 0 <= x < 0.5

All those notations have exactly the same meaning. However, I still prefer to keep formulas as simple as possible and this implies avoiding variables in subscripts when they are actually not needed. If you could follow the discussion down to this point, you will easily notice that the formula written above has the same meaning as the one I add below:

(Fe

^{3+},LiMn

^{2+})(PO

_{4})

We really should stop questioning our notations of chemical formulas every time we are asked a question about the meaning of a particular notation or about seeming discrepancies between our formulas and those found on different websites. Understanding chemical formulas of complicated minerals requires a chemical knowledge that exceeds what is taught on high schools. There is no way to simplify chemical formulas so as to be understood by anybody - pretty much in the same way as there is no way to guarantee that a phrase in a foreign language will be understood by anybody, no matter how simple it is written.

Consequently, there will always be questions from people who just don't understand the formula, no matter which notation we preferrably adopt, and ask us because they want to learn. Trying to adopt a simplfied notation in a fruitless attempt to avoid

such questions in the future will only raise more questions in the long run. More importantly however, it will also not resolve the respective person's problem; this still has to be taken care of by an appropriate explanation (which, in some cases, will be rather lengthy). There also will always be people who do not understand the formula and do not want to learn, but just spot a difference between our formula and the formula on another website and think that this alone qualifies as an error. Although I generally dislike smarties, this behaviour has to be encouraged as well, since there is always a finite chance that they did really find an error.

This said, we still need to answer one of Vik's questions:

It is the composition of a mineral (or, more generally, of a chemical compound) that is defined, but not the formula itself - a slight but important difference. Formulas are an important tool in a chemist's daily work and as such, they have to be flexible (in fact, if we have to be strict, we're mostly using systematic compound names, not formulas). The only requirement for a formula is that it must properly reflect the composition, but there are no definitions whatsoever concerning the level of detail or the notation. Even the IUPAC has surprisingly few recommendations on the formulas of inorganic compounds. They generally recommend to use sum formulas (i.e. formulas with a zero level of structural detail), because they are easily standardized (on the other hand, they also are horribly impractical for more complicated compounds). They also recommend (very amusing, actually) to write the individual elements in alphabetical order. There are many IUPAC rules for organic compounds (where they are sorely needed, not only to ged rid of the flood of superfluous trivial names), but they are entirely irrelevant for minerals.