require minimum of two lines or lg inside lg

currently <lg> asks for <zeroOrMore> of group of things. would you be happy if we just made that <oneOrMore>? the choice is:

<ref name="model.divTop"/>
<ref name="model.lLike"/>
<ref name="model.stageLike"/>
<ref name="lg"/>
<ref name="model.global"/>

(which is actually fairly weird, but no matter).

Or do you mean that an <lg> must have 2 <l>, come hell or high water, whatever else it has?

I am sure someone (probably Paul) can find an example of some poetic text in which there is a stanza, indicated by number, heading, etc, containing only a single line. Something like the following is also surely very common :

<lg type="stanza"> <l>... </l><l> ... </l>
etc </lg>

<lg type="chorus">
<l>Hey ho rumbelow &c.</l>
</lg>

So although I agree it might appear like poor practice, I don't think we should suppress it.

(Far poorer, in my opinion, is the practice of requiring <lg> when there is no actual grouping of <l>s at all -- analogous to the use of <div> for an undivided text)

I agree that a "line group" (stanza or verse) might contain only a single line, however counter-intuitive this might seem. It's the name of the element that causes this apparent oddity. I think reject this proposal.

I'm with Gaby: reject. A single-line line-group is a legitimate structure, and part of its poetic impact may come from the fact that although the overall structure would suggest a multiline stanza or group, this instance has only one.

Since the text and the schema are out of sync, I think we should change the docs to "contains a group of verse line(s) functioning as a formal unit" or similar

I don't mind if we reject my proposal to have the schema force an agreement with the current prose. But then we *MUST* correct the prose to no longer have the restriction that it suggests. 'a group of verse lines' is plural. If we change this to 'a group of verse line(s)' then the implication is that we can have 'a group of verse line' which is not only poor grammar but seems nonsensical to me.

What about if this said "contains one or more verse lines functioning as a formal unit" and we change it to one-or-more in the schema?

-James

Following the decision of Council in Ann Arbor, "MH will write a Schematron rule requiring that <lg> contains <l> , <lg> , or <gap>", the rule has been added. I'm keeping this ticket open to remind me to write a test for it, and through doing that, document the testing framework in a section of the "How to edit the Guidelines" document.

Forgot to add: that was done in rev 10522.

Closing since appears to be completed