Quote from: rebuilder on March 03, 2016, 05:54:05 PM

AlexGR: So, in the 90's, did the IETF dictate a cap on e-mail size? No? Somehow it survived, though.

No, but we actually had disk quotas on our unix shells and everyone understood the reason for it because we were more technically minded. A mass mailbombing on the list of users stored at the /etc/passwd file would not only fill all our mailboxes, it could actually fill the entire server and crash it.

Quote

I repeat: I'd like an example of a centrally planned production quota winning out in efficiency over a free-market supply-and-demand model.

It depends on what you mean "production". We are talking about services here. The closest thing that comes to mind, in terms of central planned services and quotas, is national infrastructure.

In my country, for example, the main roads and train lines that connect the various cities are singular. There aren't 5 roads or 5 train lines that overlap the same distance and "compete" between themselves. There aren't 3 power line networks that compete. It is well understood that a centrally planned infrastructure in things like roads, train lines, water pipes, gas pipes, power lines, etc etc, cannot be subject to competition. It is highly redundant, increases costs, decreases efficiency, etc etc. Whatever competition might exist, is in layers beyond the infrastructure itself.

Quote from: BitconAssociation on March 03, 2016, 05:58:32 PM

As such attack vectors became apparent, successful mailbox providers realized the folly of allowing "a few kb / a few mb mailboxes," and limited mailbox size to a few hundred bites.

Initially it might have started uncapped - I actually remember having one that was uncapped, although I didn't use that one - until they had to impose the few kbytes limits due to practical considerations.

Quote

Email account caps grew and grew.

...as technology allowed it.