Why not positional variation?

[RadioFM]

Although the best default approach should probably be to consider each distinct variation as its own glyph and not make too many assumptions about a system we cannot read, I too am toying with the idea of some glyphs being positional variants.
I'm not so sure about the other flourishes, but I personally feel that n (EVA-n) may just be a word-final variant of i (EVA-i). I myself am not savvy on paleography, but I'm sure I have seen some other few manuscripts having n as a mere "final" i minim with a flourish. Please, feel free to correct me on this.

Nevertheless, for the sake of argument, let's say we cannot know for sure n is just another way of writing a final i.
As far as I can tell, unlike r or s, the relatively few times n is followed by another character, it tends to either occur on "LAAFU" line positions (i.e. beginning/ending of lines, or cramped beside the edge of a drawing), other times within anomalously long words, sequences with ambiguous whitespaces, or in Labelese. We do see some words starting with 'flourished' glyphs such as y, r and s, but not once one that starts with n.

Also, there are almost no words ending in i in the entirety of the manuscript, You are not allowed to view links. Register or Login to view..

It is also worth noting that in f27v:
    n is completely missing in the entirety of f27v
    i seems to be also missing, except for a couple of glyphs that I read as awkwardly written e, dcar and daidy

I get the feeling that other "flourished" variations behave differently, and that there may be something different going on with them.
For instance, take the glyphs generated from e and i according to the You are not allowed to view links. Register or Login to view.. r and s share the same flourish, and both occur fairly often in the MS. Same goes for y and l, but the respective "counterpart" to n (b, EVA-b) - which occurs not nearly as often as n - always occurs in similar word positions as n, and often preceded by a sequence of e "minims" (just like n is often preceded by i). See folios f29v, You are not allowed to view links. Register or Login to view. and You are not allowed to view links. Register or Login to view. for a couple of these occurrences.

I know for certain long time users of this forum already know Emma May Smith has previously delved in a similar topic on her hypothesis of a and y equivalence, but I'll leave it here just in case: You are not allowed to view links. Register or Login to view.

---

Going off on a tangent here, but...
IF there is meaningful, underlying plaintext encoded in f27v, the absence of n in such a rather long-running string of text suggests to me that there is redundancy in the coding, and n, in, iin, iiin sequences can somehow be replaced.

Also, even though nearly no words end in i, with scribe(s) apparently preferring n as a finishing i variant, we do observe many words ending in e and the absence of b in many of the VMS quires, which marks a noticeable contrast between two seemingly similar variants of flourished minims.