Earlier quoted context omitted.
Can you point to a clear, major innovation software innovation that they are currently being sued for? I can't think of any. I'm also not seeing the difference between "It's nearly impossible to write a nontrivial program that doesn't infringe patents" and "I do not believe I could write any significant program that was entirely patent free."
"Can you point to a clear, major innovation software innovation that they are currently being sued for? I can't think of any." Java ("everything else sucks" - Google), miscellaneous mobile innovations belonging to all sorts of companies via proxy (because they don't actually ship phones). They haven't been sued over VP8 yet but it's likely coming barring licensing. ... "I'm also not seeing the difference between ..."…
VP8: the fact that Google has been using VP8 in WebM for 14 months now, with no clear allegations of wrongdoing, kinda kills your "willful" argument. If Google was fully aware they were infringing on patents, don't you think someone would have brought this up and/or sued them by now? (I could also start ranting about submarine patents, but I'll leave that alone.)
The parent poster also didn't really ask the right question. "Point to a single nontrivial piece of software that doesn't infringe on someone's patents" may indeed be solved by pointing to anything written over 20 years ago, but how is that a useful measure? We're not going to just say that we only need software written by 1991 and everything written in the past 20 years is useless. So sure, maybe "mission accomplished", but that's like saying "mission accomplished" for successfully going to the bathroom. Sure, well done!
The more useful question is -- "can you write a nontrivial piece of software today that infringes no patents?" I don't know the answer to that, but I suspect it's hard to do.