Although no one likes to talk about them, unmarked questions of the 'ona li seli?' form occur from time to time. The problem with them is that, since we don't know anything about tp spoken sentence contours, we don't know what the open question sounds like. Anglophones, of course, have that rising-toward-resolution pitch glide (or did, until it became the teenies normal sentence ending). Other languages have other patterns, including none, for languages which are always explicit about questions. But the question mark solves this problem in written tp. And one form of this is not only common but fundamental, the 'anu' choice question: 'sina wile e telo pimeja anu telo lipu?', which is overtly indistinguishable from a declaration about your possible wants. (This is fundamental because it underlies 'x ala x' -- originally 'x ala anu x' -- and, more obviously, S anu seme?', whose analysis is rather complex.)
The problem with 'seme anu S?' is that the 'seme anu' is likely to get caught up with the subject of the S, suggesting that the choice was not "yes" and "no" but about what the proper subject of the S was. To be sure, the claim that 'x ala x' and 'S anu seme' are equivalent is fairly clearly wrong anyhow. The first allows conventionally only the answers '(x) ala' and 'x', whereas the second allows almost anything relevant (broadly interpreted), typically alternate suggestions (which might come up after an 'ala' by way of explanation or suggestions how to proceed). The 'seme' seems, in most cases. to match S as its category and so allow -- in two steps -- for a totally new sentence directly.
Its not clear whether the various question types carry any special pragmatic or rhetorical or conventional effects beyond those mentioned here, like the old "expects the answer "yes"", for example.