4 ms·
I guess someone calculated that given current rate of sequential assignment of MAC ranges, we won't get to 4 and 6 prefixes in a very long time. That reasoning
by tener 10y ago
I guess someone calculated that given current rate of sequential assignment of MAC ranges, we won't get to 4 and 6 prefixes in a very long time.
That reasoning was broken as it assumed sequential assignment is the only way to go.
- packetized 10y agoUntil fairly recently (in IEEE/OUI terms, ~2012), the distribution has been entirely sequential. FTA: "I believe IEEE changed their strategy to attempt to purposefully higher the chance of collisions with MAC squatters, to encourage people to register and pay the fee"
- bandrami 10y agoNonetheless it was a broken parse: they needed to expend the memory to figure out what the actual protocol is and not jump up from the datalink layer if it's not called for. I guess I'm just saying I really can't blame IEEE here.
- packetized 10y agoOh, I don't blame the IEEE wholly - but I think that they probably changed tack after realizing that a number of manufacturers were simply not paying the fees to actually be assigned a block of OUIs. I view it as kind of a cudgel to be used against vendors, and end-users (read: netops/sysops) are caught in the middle.