looks like this is what i was thikning of http://lwn.net/Articles/542642/ http://tools.ietf.org/html/draft-dukkipati-tcpm-tcp-loss-probe-01

the problem with packet loss on the last packets is that tcp/ip only resends packets after a timeout due to lack of ack, or after acknowledging earlier packets multiple times signifying a gap. so if you send 8 packets of data for a typical web page it's way worse if you lose the 8th packet than the 3rd packet. (linux defaults to an initial window size of 10 packets now, windows defaults to 8192 bytes, linux to 14600 bytes. so in linux that would arrive in one round trip time for the data, one one round trip time for the connection setup normally. and in windows three round trips. but when there's packet loss, linux will normally wait for one or three seconds to resend the unacknowledged last packet. i think there's been a recent change to just resend the last packet extra earlier. maybe with the estimated rtt rather than the timeout value.

most of my spam comes from gmail, yahoo, and msn, so i think it's not really the right solution. rbl lists etc are so common now that they just make heaps of accounts on free mail providers. i suppose i could block all free mail on my personal server, but i can't see isp's doing such.

Basically that's the main one. I'm reasonably OS agnostic these days. Maybe apathetic is a better word. Chrome works about the same on Windows and Linux. Skype works better on Windows than Linux. And foobar2000 is better than any music player that I've used on Linux. But ssh/terminals works better on Linux, and Linux pauses less. Currently my solution is to use Synergy, and have two computers, one Windows, one Linux. Kind of the best of both worlds.

Wouldn't it be better for testing to use different phones to test. It may be that one phone works better on one network while the other works better on the other. With DSL this is certainly the case.

So have NSA denied their involvement in taking facebook down today?

So when are youtube, vimeo etc adding support for it?

Verizon and Cogent are most likely exchange a LOT of bandwidth. And the article says that there are 10 peering points. Cogent are at way more than 10 locations in the US alone. I've got no idea how many Verizon are at, but to me it seems reasonable to connect at more locations rather than push all the capacity at those locations. That then means Cogent needs to do some of the backhaul, or have Netflix boxes at more locations. I think this is a reasonable compromise, but I can't find any detailed summaries of the situation. It would also be advisible for Verizon to add Netflix boxes on-site.