This is what I got from comcast.
Important Speed Increase Update
Just when you thought the internet couldn't get any faster, Comcast High-Speed Internet brings it to you with more blazing speeds than ever before.
During the previous weeks Comcast has been upgrading all of our High-Speed Internet Service customers from 3Mbps downstream and 256Kbps upstream to speeds as fast as 4Mbps downstream and 384Kbps upstream - at no additional charge.
All residential Comcast High-Speed Internet markets have now been upgraded to the 4Mbps/384Kbps service.
*drool*
I got cable in August 1999 from Multimedia Cablevision, a regional company.
They provided 10mbit up/down
Cox bought Multimedia in mid 2001. Later, cox dropped all connections to 3mbit down/384 kbit up.
Last year we got an upgrade to 5mbit down/768kbit up.
But 1000 times faster?
*drools again*
Hellooooo home-based webserver I ran in 1999-2001. I miss you dearly.
sounds like all I'd have to do is enter my office with a thought in my mind and ZAP, before I hit the keyboard - there it would be...
somethings are just too fast
When will the speed limit be made law to protect us from ourselves?
Of course this will mean new hardware is needed and...you don't think they'll be dishing this out for free do you? Hell no. We'll see new "tiered" levels of service...just the same as we do with cable TV. I'm perfectly satisfied with the speed I get now. As long as they don't try to screw me over and FORCE me to upgrade or raise my rates, I'm happy. But then, they're not in business to make me happy, are they?
Actually, one thing I would like to see more than increases in raw speed would be an improvement in the mail-bounce architecture. At present, mail bounces are delivered using normal 'forward' mail protocol and consequently there is no guarantee that the recipient of a bounce will have anything to do with the original sender.
My proposed improvement would be to have every machine that handles a message include a tag that would either include the IP of the source and some authentication information (so that, if the machine sees the message again, it can tell whether it in fact created it), or else some identifying information about the original sender, also including identification information.
That way, if a message bounced, it could be routed back via the same pathway as it came; since machines would only pass along bounces if they had handled the original messages, this would prevent bounces from harrassing innocent people who had nothing to do with the messages being bounced. In addition and more importantly, by preventing the delivery of bogus bounce messages, such a protocol could increase the likelihood that undeliverable mail would actually result in a bounce message to the originator [at present, many spam filters discard messages without bouncing them to avoid the hazard of bouncing to spoofed addresses; if bouncing didn't carry such risk, the filters could let the message originators know that their messages didn't reach the intended recipients].
Yea, but what about their servers? And dial up users are going to find the net completly unusable.