This is the kind of analysis and thinking that would have led to the 1988 version of the internet to scale to about 10,000 users at any one time by the year 2017.
In fact, it is entirely the opposite. Bitcoin's block chain limit to 1MB is akin to specifying that IP addresses shouldn't be more than 16 bit, because otherwise, poor people with 8-bit processors and 64K ram would have to waste too much memory on IP tables. It is bitcoin's kind of thinking that was also inspiring Bill Gates to claim that 640 KB or RAM address space is more than enough for a home PC, ever.
The idea that one should put in small hard limits to avoid the need of computing resources (which is the stance of people wanting small blocks) is what would, indeed, have kept the internet to about 10 000 users. I'm saying exactly the opposite: that there is NO resource problem with bitcoin, and that one is INVENTING a BOGUS resource problem, because EVEN if bitcoin were to go mainstream, by the time it does, the resources needed for that would be available. I'm telling you that bitcoin has OTHER design problems, that 1) most probably will not make it go mainstream and 2) are fundamentally flawed in the economic design (in fact, 1 and 2 are even related). Crippling bitcoin even FURTHER because of a NON-EXISTING resource problem, won't help.
Actually, I am wrong. According to your last paragraph, the internet would have scaled to zero users by 2017.
If the internet were designed like bitcoin, with hard usage limits in it, yes.
Bitcoin needs more Al Gore's, fewer you's.
If bitcoin needs Al Gore's, then that can only confirm my position that it is ill defined
