A couple critiques of the linked piece. First off, the title is somewhat misleading. See this Reddit discussion on the same piece: https://www.reddit.com/r/tec

A couple critiques of the linked piece. First off, the title is somewhat misleading. See this Reddit discussion on the same piece: https://www.reddit.com/r/technology/comments/7gob2w/americans_taxed_400_billion_for_fiber_optic/dqkhyms/ https://www.reddit.com/r/technology/comments/7gob2w/american.... It seems like the "taxes" were really fees that went straight to the telecom companies. It's not as if the government collected these taxes, and then failed to push ISPs to roll out fiber or build public fiber of its own. Secondly, this piece tries to make a number of comparisons to countries that have faster wired internet service, namely South Korea and Singapore. South Korea and Singapore have much smaller populations that are concentrated in urban areas, and are generally smaller geographically. The sheer size of the United States, combined with our propensity for sprawling suburban development, in addition to the higher throughput requirements of a much larger population, would seem to suggest that delivering widespread fiber capability will cost a lot more in the U.S. than in South Korea or Singapore. This is not even to mention the eminent domain issues and federal vs. local government conflicts that South Korea and Singapore do not face to comparable degree. Third, the "recommendation" made at the end of the piece seems to be based on a bit of wishful thinking. > "The Trump administration would be wise to make improving America’s digital infrastructure a priority. The South Korean model, market liberalizations combined with hard investments, could be a viable model." What "liberalizations" would lead to growth in fiber connectivity? For the large numbers of Americans with only one ISP choice (https://arstechnica.com/information-technology/2016/08/us-broadband-still-no-isp-choice-for-many-especially-at-higher-speeds/ https://arstechnica.com/information-technology/2016/08/us-br...), it's not clear what incentive there would be for a competing ISP to come in, make the large investments necessary to physically lay fiber, and then begin competing with the original ISP on price and service. There is tremendous upfront financial risk, and if the thesis of market liberalization is correct, namely that competition drives prices lower and speeds higher, the ISPs laying fiber may well never see a return on their investment. Fourth, I think we should generally question whether massive national investment in fiber even makes sense at this point. Improvements like DOCSIS 3.1 allow for fiber-like speeds over traditional cable lines. At a certain point, we should ask whether gigabit speeds are really all that useful for most people.