I get a bit tired of this: > Xerox failed to turn most of the technologies that it developed at PARC into successful commercial products PARC didn't try -- an

I get a bit tired of this: > Xerox failed to turn most of the technologies that it developed at PARC into successful commercial products PARC didn't try -- and it did make a boatload of licensing fees. PARC from the start followed a policy of licensing intellectual property instead of trying to build it out because that set the expectations of researchers: your ideas will get out, but you'll only get glory and a paycheck and the chance to be a part of a fantastic team. That was essential to the culture and team dynamics. I was on a PARC team that went through the commercialization process when PARC just started licensing IP en masse (e.g., to help large IP holders in their patent battles). PARC researchers are very smart and can envision the future as well as anyone, and if it's possible they can hold something back and then capitalize on it for themselves, they will do so, which completely screws and politicizes the team dynamic of open discussion and development. If you think about the pace of new findings in new territories, slowing down to capture and carry value will be a serious drag, and PARC wouldn't be what it was. I think PARC was right to only monetize patentable ideas (in a parallel stream of lawyers), ignoring opportunities that stem from speed or execution because that's where business uncertainty lies. Indeed, I suspect that in the rapidly-evolved AI space, the companies that can gather the best teams to do the best work in this semi-open fashion will end up with the most (and most valuable) intellectual property. (And if their lawyers are better than PARC's at licensing by recognizing new forms of residual opportunities in AI, it will be beyond profitable.) The captured groups at openai/microsoft, Google, Apple, et al will get tangled in the culture, forces, and technologies driving their patrons. Don't turn your scouts into farmers or builders.