It's not very bad advice. I've been building web businesses for 16 years. You have no advantage over me on time.
The end of the dominance of .com was always inevitable, and is already occurring. Sites as varied as ask.fm, twitch.tv, minecraftwiki.net, foods.tv, formspring.me, php.net, about.me, imageshack.us, last.fm, mega.co.nz, rapidgator.net, popads.net, speedtest.net, ustream.tv, prchecker.info, who.is, all.biz, jsfiddle.net, justin.tv, bit.ly, scoop.it, snap.do and vine.co are proving that every day (or already proved it in the past). And that's ignoring the rest of the vast array of .net and .org sites. There will be a lot more big alt sites in the next five years. Instagram didn't start out as instagram.com. Clearly Vine didn't need the .com to succeed at all. Different use case than your typical 1998 .com? You bet, and that's exactly the future.
The proof is already in place: you don't need a .com to succeed any longer.
Among the top 5,000 global sites, .com addresses are in the minority now. Apparently other entrepreneurs and businesses in other countries are also having no problems building global, world class businesses without using .com.
In fact, over the last decade as the web has become dominated by non-English speaking and non-American businesses and users, I'd argue that everybody outside of America is already very accustomed to the use of alternative domains. Just another reason .com is the past.
Saying that something is going to be problematic, doesn't mean that it is actually so.
Only $70,000 for that .com? You mean in other words very few startups could actually afford that. You're talking about an elite class of startups - the top 1% - once you're talking about spending $70k on a domain.