As I see it, the key issue is about allowing users to identify themselves. When I went to see Mr. Spolsky speak about SE, the first and biggest point he made is that when someone visits the front page of an SE site, they can immediately see whether the site is for them. They look at the questions, and they think, "Yes, these are the kind of questions I am interested in," or alternatively, "I have no idea what these people are talking about." Obviously, this isn't to be taken too narrowly: nobody suggests that questions about phones put off tablet users.
Looking at it that way, the debate over non-programming questions that only a developer would ask (which includes publishing questions) has these pros and cons:-
Arguments for allowing
It would attract developers (most of whom are also users) to the site. They would be useful for answering some of the harder user questions as well. Some of the users interested in such questions are already ASE users, so the change would make the site more useful to existing users.
Occasionally, even developer-oriented questions are useful to end users. Sometimes it satisfies one's curiosity to see what happens "behind the scenes". Better understanding the relationship between app publishers and Google can help advanced users to understand quirks in the way apps interact and why they're written in particular ways. We try to spot such questions and see their usefulness from a user point-of-view, but I wouldn't claim that we're 100% successful: I'm sure there are closed questions that could be useful but were phrased in the wrong way. In addition, there may be many more that simply aren't asked because of the policy. Allowing developer-oriented questions may well improve the site for non-programmers too.
Arguments for disallowing (i.e. keeping the status quo)
Visitors who aren't developers would see the questions written from a developer's perspective, be unable to identify with them, and thus feel less like the site is for them. How likely this is would depend on how many such questions we get, but don't forget that a feedback mechanism is at work: the more non-programming developer questions we get, the larger a fraction of the site will be developers, so the more non-programming developer questions we'll get, &c.
There's a smaller risk that showing ASE as a place where developer questions and user questions meet will increase the number of other off-topic developer questions, such as polling for user suggestions, spam advertising apps, and localized questions that only Google support can answer ("can I upload app x").
Balancing the arguments
So which side you're on really depends on how you weigh the risk of driving away users against the chance of attracting new users and becoming more useful to existing users; and the burden of new kinds of bad questions against the proportion of questions that turn out to be good for non-programmers.
To weigh these up, you also have to consider that there are a lot more non-programmer Android users to drive away than there are Android programmers to attract; OTOH, the programmers are more likely to already be SE users, so they're more likely to find ASE in the first place.
Whatever we decide here, I think these arguments apply almost verbatim to other kinds of non-programming developer questions. It might be worth thinking of the publishing questions as a "pilot" for allowing other kinds in future.