anybody doing an SATB setting of something expected to be sung with the congregation should make sure that - even if the melody is the soprano (or alto) - it should also sound acceptable in the men's octave.

Exactly. That is one of the essences of good hymn writing for congregational singing. One should expect a significant number of men's voices singing the melody down an octave from the soprano ... unless it's one of those hymns with the melody (ie., the "tenor") in the tenor part.

Anecdote: When accompanying hymns on a Rodgers that had a "solo" setting (that coupled the highest note played on the Great to the Swell), to bring out the melody line, I generally made sure that a 16ft was among the stops drawn on the Swell, so as to give that octave-below-the-soprano feel to the texture. Of course, the 16ft (and on final stanzas 32ft) stops on the Pedal reinforced the bass line well, even when there was a danger of the sort of close harmony that SkirpR mentions.

In my own hymn writing, I almost invariably envisage men's voices (if only in the congregation) doubling the soprano melody. And of course, I do write lots of descants for which it would be expected that the men's voices of the choir would sing the melody. In those instances where I've written soprano "double descants", it is the case that, say, for a four stanza hymn, descant 1 is assigned to the second stanza, descant 2 to the third stanza, and both (with divided sopranos) to the last stanza. For all the descant stanzas, the altos and men sing the melody. There are also instances where I've had tenors sing one of the descants and the sopranos the other, both joining on the final stanza, in which case, all voices not singing descant sing the melody in the appropriate octave.

I don't see why these principles could not be applied to setting the Gloria, if one is predisposed to having some sort of descant to ornament the main melody. I do think, however, that in any through composed Gloria ... SATB, possibly with descant(s) ... the initial use of that Gloria with a congregation would entail using only unison voices at first, until the congregation is comfortable with the setting. Then one could have the choir sing parts, with say one or two strong male voices still singing the melody. Only when the congregation is very comfortable with the setting would I feel free to have the choir sing in full 4-part harmony, and with any descant(s). This simply reflects a sound layered approach for a congregation to learn a new work.

@mrcopper... This has long been my sense too. I suspect you may be hitting at the original poster's intention, and the "conflict of interest" he speaks of.