Usenet Growth Reveals Need for 64-bit Based Article Numbering

[Beck38]

Tripped across this the other day on GN; further proof (if any was needed) in the explosion of usenet.

http://www.giganews.com/news/article/64-bit-Usenet.html

[MultiForce]

I like the "18,446,744,073,709,551,615 (over eighteen quintillion)" part :D

[Beck38]

I like the "18,446,744,073,709,551,615 (over eighteen quintillion)" part :D

Yep; then again, I remember all the arguments (at the time) with IPv4 and the address space there, that ABSOLUTELY NO WAY would the total addresses in the whole UNIVERSE eat up all the spaces.

Of course, this was when the TOP people in the field (read: cloistered academic types with no real grip on the 'real world') were saying that, at a maximum, 1 home in 1000 would ever have a computer, and out of those 1 in another 1000 would actually have a connection to 'the internetwork'.

Now, of course, the entire planet is (in fits and starts) looking dead in the face of the conversion to IPv6. And the major reason given for the (now small) address space in IPv4? It saved a couple KBytes of memory space in the protocol stack. But that was considered a big deal, cost $20 in memory in the mid-70's.

I used to run a usenet server from my home about 14 years ago, message newsgroups and such, way before even jpeg groups really got going. Interconnection via ISDN. I even tried satellite feeds for a while. But even then, it was obvious (with the message traffic) that the article numbering system was going to eventually get jammed up.