hzrn wrote:
> wait, how would this rafsi work in a lujvo? for example, "selatmi"
> (according to sutysisku, to be clear) apparently decomposes to "se" and
> "latmi". i dont think adding a hyphen would help here either (decomposes
> to "se" and "[hyphen]atmi".


This is a common problem with zevla. As mentioned, appending .ybu and .y'y
to the cmavo rafsi, gismu rafsi, or full zevla which are acting as rafsi
is the usual strategy. I have thought that it might be okay to use 《ZEVLy》
instead of 《ZEVLA'y》, iff we implement a rule that no two distinct zevla
can have distinct meanings while differing by only their final vowel
cluster (howsoever long that may be – medial glides are another issue
which would need to be resolved for this purpose), but that does not apply
here. I just provide the thought for your edification.