I don't think that I agree with the conclusions of that article. There was an
article in the Wall Street Journal (paywall) that said that it was believed that the interval for Moore's law might be slowing to something closer to ever 2.5 years:
“The last two technology transitions have signaled that our cadence today is closer to 2½ years than two,” Intel Chief Executive Brian Krzanich said during a conference call with analysts to discuss second-quarter results.
Reality check. Moore's Law is bullshit, proven false (as if it were ever scientific). LOL try again. Do you realize that Moore's Law didn't actually apply to bandwidth? It regarded processor speeds, or overall processing power. And again, it's dead, finished, irrelevant so the point is moot. Take your "technology gets better, don't worry!" talk to the kindergarten class.
Moore's law applied/applies to how many transistors can fit on a chip, however the principles behind moore's law generally apply to other kinds of technology as well. Moore's law is also generally accepted to be slowing, but is far from dead. On a cost (to the consumer) per unit basis, technology has been increasing at an exponential rate for decades, and I do not see any sign of significant slowing down of this growth.
AFAIK bitcoin was created as a p2p cash in the first place and i hope in the future it gets closer to this purpose, and i think it can get there too if the number of users increase and the volatility decreases.
That's a rather complex problem (volatility) that is not directly related to the debate. Bitcoin should be able to always accommodate new users; as long as it continues functioning the same way that it does. The problem that we can run into is that suddenly 1 million people (arbitrary number) want to buy Bitcoin and create 1 TX in a day. This would create a 'heavy' backlog because the system is only able to handle 3 TPS at the moment.
This is why we need everything (to be ready and prepared):--snip--
2) Block size increase
--snip--
Quoted for reference.