> Thanks to the trifecta of ubiquitous smartphones, pervasive Internet and no-holds-barred data mining, I'd argue that the tech landscape we inhabit today is ve

> Thanks to the trifecta of ubiquitous smartphones, pervasive Internet and no-holds-barred data mining, I'd argue that the tech landscape we inhabit today is very different from anything we've ever encountered. Say a startup does create a better product than Google or Facebook, it can't charge for it, because everything is free. It certainly can't monetize it better than Google or Facebook because they have infinitely more data on users than any startup can even fathom. It has to go through Google's and Facebook's "gates" to find and retain customers. Those precious few that do manage to pull off the impossible, like Whatsapp, will be quickly bought over by Google or Facebook. Thus closing off any gaps that existed in their defense and reducing the possibility of independent competition even more. I feel like you just described the exact circumstances in which Google and Facebook came up in the world. Competing against huge entrenched interests, couldn't charge for their products, and remember that Brin and Page tried to get acquired and Zuckerberg refused many offers. If history had gone a little differently and had Zuckerberg changed his mind, both companies would be little more than footnotes today, absorbed into some existing behemoth. Meanwhile computing infrastructure today is far far cheaper at gigantic scales. I don't think we can conclude much from the fact that very few companies become hugely successful these days; that's always been true. > Merely that we need to do a comprehensive forward-looking analysis of Google's dominance, extrapolating from present data and trends. Using past data alone strikes me as lazy or self-serving present data and trends are a direct result of the past, and while things do change, it's not clear that today is fundamentally different than 10 years ago and those aren't all still instructive examples (while remembering we're hopelessly under the foot of biased sampling here)