Also (and this is mentioned everywhere in these circles), what you actually launch with may be entirely different than your idea right now anyway.
Don't worry about problems you don't have yet. Good to know about the up coming problems, but focus on making progress, otherwise it's just mental entertainment.
Remember: if a patent holder can show that you've knowingly infringed on a patent, it's much worse than the alternative. Keep your mouth shut until you're absolutely positive that you've got a problem. How do you know? You talk to an attorney.
But you better ask your lawyer on this topic.
Generally speaking, the idea can't be patented, but the implementation can be. Get a copy of the existing patent, and document how your implementation diverges. Have your lawyer see what the potential exposure is.
For that matter, I don't even read paper publications out of places like Microsoft Research because of patent-related issues. Take a look at this, warning about an IBM publication: http://on-ruby.blogspot.com/2008/02/ip-police-line-do-not-cr...
Make a good product, make money, then pay lawyers to sort it out for you. Its all mostly pretend now anyway. Patents really have about as much to do with ideas as Scientology has to do with science.
It also seems these days that the more a company blathers about patent-pending this and proprietary that, the less actual innovation they've done.
1. whether copying was deliberate
2. whether upon notice infringer did good faith investigation (upon notice, did you see if you really were infringing)
3. infringer's behavior as a party to the litigation (did your lawyers place nice in pleadings and motions)
4. infringer's size and financial condition (big guy vs. little guy?)
5. closeness of the case (if the case could go either way, no treble damages)
6. duration of infringer's conduct
7. remedial action by the infringer (when notified of infringement, did you develop a non-infringing work around)
8. infringer's motivation for harm
9. whether infringer attempted to conceal misconduct
The court weighs all of these to determine whether to enforce treble damages. They are not weighed equally, and you don't have to fail a certain number.
You have 2 ways you can go. First, you can just build the idea and see what happens. Don't research whether you are infringing, and just take it as you are working in the same area as someone else.
Or you can read the actual patent and see what is claimed. Engineer around that, and around extensions of that patent that a person of ordinary skill in the art would consider obvious.
In this article I talk about how patent's are effectively nonsense for all but a very few who know how to 'work' the law who also have the fat to do so.