Patent examiners can do their thing. Alice gives them a tool now too.
Yes, and no... Patent Examiners are bound under the requirements of due process to present a prima facie case for why an application is not patentable, as the initial burden rests on the Office. How do you provide a prima facie case that an idea is abstract? It's a conclusion, not an argument supported by evidence, as the Courts have admitted when their evidence is "I know it when I see it".
How do you define "actually inventive"?
Here are the questions I ask when contemplating patent filings, post-Alice, for a software method (or computer implemented method):
* Can I reasonably determine the bits and pieces you put together a specific solution to a specific problem based on your claims? should avoid a 101 issue.
* Do the claims give me all of the pieces of the puzzle or does it give me a flowchart?
* And, to entirely avoid an Alice question, are you using generic bits of technology for their ordinary purpose to solve an old problem the old way?
"Good" answers to these questions should avoid a 101 issue.
Quite possibly, though it fails to answer my question about your definition of "actually inventive". It also points to part of the problem with Alice, since your first question is really about 112 written description, your second question is really about 112 enablement and unclaimed essential matter, and your third question (as you note) is really about 103 obviousness. Now, I agree, that if you meet 103 and 112, Justice Thomas would likely not "know an [abstract idea] when he sees it" and find the application invalid under 101, and maybe that's a fine answer from a pragmatic standpoint, but it's a terrible one from a jurisprudence standpoint.