Out of curiosity, I went to have a look at AMZN's stock price at the date the article was published, to see how the stock price during the heady heights of the dotcom bubble compared to today.

Interestingly, if you view the entire stock history price chart, the dotcom bubble is barely a blip.

IMO, the right way to look at long-term performance of an investment is on a semi-log chart. (Day or week charts, I prefer linear; anything 3 month or longer, I prefer semi-log.)

Semi-log (where constant percentage changes are constant-sized), it's more than a blip. Linear (where constant dollar changes are constant-sized) charting hides past market moves if the overall bias has been up and exaggerates past market moves if the overall bias has been down. This can lead investors to improperly discount the pain/gain cycle that happened over that time period.

Linear charting: http://bit.ly/2tk5Ulj

Semi log: http://bit.ly/2sJs7ec

Now that I also actually take a look, while his theory has not panned out the way he thought, it is true that Amazon would have been a poor investment for most of the decade following 2000.