I was almost certain[1] that Craigslist was working on PadMapper like interface and it was the reason they sued. They should have done the right thing and hire/buy one person company PadMapper.
On top of that, even if PadMapper had not been slimy, I don't see why it's "right" to pay a sum for something you're perfectly capable of building yourself for less time, money, effort, and headache than what an acquisition would cost.
That is, it isn't so much what Craigslist "needs" to do to be in the right, here, but it's what they could do to be in the very right. To be better than any other random classifieds company, to generate viral PR, to be something you'd be happy to say you use, etc.
2. One cannot "steal" publicly accessible data.
3. A search interface with resulting links to the relevant post on CL; if PadMapper was a "competitor" then so is google.
Well, since you've clearly been over this more carefully than most of us, maybe you could be troubled just a bit to elaborate on these questions one more time:
* Is it clear that the kind of data that craigslist published can in fact be "owned"? Because no matter what their terms claim, the legal analyses I've read aren't positively conclusive.
* Why is what Padmapper did any different than a search engine? Contrary to some assertions, PadMapper was not republishing listings wholesale -- they published digests (making the ownership issue even less tenable) along with hyperlinks, which sounds a lot like a search engine to me (pretty much what they are). Or even a typical anchor tag itself: the contents of such a tag are generally a digest, the attribute presents the hypertext.
* If you agree these activities are something people shouldn't be able to do without permission, are you fine with a web where search indexing or even linking is essentially by agreement only? This is not an academic question, btw -- there have been lawsuits over linking: http://www.salon.com/1999/08/12/deep_links/