[dctoedt]

> patent brain storming camps encouraging various employees to think of and file for patents on just about any topic regardless of the business's involvement in that field

One principal reason is to establish a defensive portfolio, for possible use in counterattacking and as currency in future cross-licensing negotiations. If (let's say) IBM were to threaten sue XYZ Corporation for infringement [0], then XYZ would look through its own portfolio of patents to see which patents the company could use to assert a counterclaim against IBM; if XYZ didn't own any patents, it'd have nothing to trade. [1] It's commonly believed that this is why Microsoft went on a patenting binge in the 1990s: so that it would have something with which to counterattack and trade cross-licenses if, say, IBM sued it for infringement. EDIT: This was documented in a 1991 Bill Gates memo [2].

[0] In the early days of the PC, a certain large computer company had a reputation for (figuratively) dropping two or three stacks of patents on another company's desk and saying, you can have a license to Stack A for X% of your revenue, a license to Stacks A and B for X+Y% of your revenue, and a license to all three stacks for X+Y+Z% of your revenue. The other company would ask, which ones do you think we infringe? The larger company would answer, oh, we have no idea, but now that you know about these patents, if you do infringe, we'll find out eventually, you'll be a willful infringer, and you'll owe us up to treble damages. Some companies decided that settling was the better part of valor.

[1] https://scholarship.law.berkeley.edu/cgi/viewcontent.cgi?art...

[2] https://arstechnica.com/information-technology/2007/03/analy...

[aplorbust]

To finish the story with some original thought, lets ask why this tactic works.

Perhaps it works because of the expense of determining which patent claims could realistically survive a motion to dismiss, i.e., the hourly rate of the patent lawyers hired to do this analysis and the number of hours they take to do it. The larger the stack of potentially assertable claims that need to be considered, the greater the cost. Quantity not quality.

If that process of determination, separating wheat from chaff, filtering out all the nonassertable (junk) claims, was less expensive than simply paying for a licenseFN1, then perhaps this tactic would not work.

FN1. Use imagination if believe this is impossible to achieve.

[dctoedt]

That's correct; many patent claims are written in long, dense paragraphs, without line breaks, subdivision numbers, or other "signposts" to guide the reader. I'm convinced that often this is often more than just laziness or incompetence by the patent attorney — it's intentional obscurantism, to make it more difficult and expensive for others to determine whether or not they infringe and whether or not the claim is in fact legally patentable. [3]

[3] http://www.oncontracts.com/multi-sentence-claims/ (self-cite)

[trhway]

>Perhaps it works because of the expense of determining which patent claims could realistically survive a motion to dismiss, i.e., the hourly rate of the patent lawyers hired to do this analysis and the number of hours they take to do it.

or may be because a patent court case is more like rolling a dice with very significant potential damage. Business likes predictability - license is just a static cost to be factored into the price of the product - vs. high risk gamble. And software patents i think are worst of the worst in that regard (nevermind that i have several of them myself :).