Slashdot Mirror
If This Had Been An Actual Emergency
saridder writes "In an increasing attempt to regulate the Internet like the current PSTN, the US Government has asked the IETF to come up with a system to prioritize government and emergency worker traffic in the event of another disaster, much like the GETS system already in place for the PSTN. It's interesting to follow, because it's only an RFC, so you don't have to follow it. I probably won't be prioritizing government traffic on any of my routers." The story has a link to the ieprep working group if you want to get involved or comment. Perhaps this is a better way than GOVNET.
You could with diffesrv right now (which is what this system will be based off of), but nobody would honor it.
--- RFC 1149 Compliant.
I was able to get streaming video from bbc, but could not hit cnn's website, implying to me that the bottleneck was @ cnn, not with the infrastructure.
"Blake is an idealist, Jenna. He cannot afford to think." - Kerr Avon, Star One, Blakes 7
Come on, you're being gratuitously cynical. Why wouldn't we want our emergency agencies to have access to the latest and greatest information, regardless of the source?
During 1992, I was involved with building the LA Fire Department's new 911 system (uh, that was a debacle but that's another story). The Emergency Operations Center had three or four 12 foot across big screen TVs that could be used to display maps, computer displays, CNN or the local media.
During last week's 9/11 special on CBS, it was commented on how TV viewers and web surfers around the world knew more about what was going on in and around the towers than the firemen in the lobby.
And then consider how many devices, sensors, or applications these folks have to get to that may only have web interfaces....
That's not what RFC means, even though I know you're thinking "Request For Comments."
See the Status of this Memo section at the top of each RFC to determine whether it's an "Internet Standard" or "Internet standards track protocol" or "Experimental Standard" or "Historic" or some other category.
RFC 793 is "only an RFC" but your packets won't be routed if you don't follow it.
Actually all the flags needed to support precedence were defined in RFC 791 many years ago. See http://www.ietf.org/rfc/rfc0791.txt "Type of Service" on page 12.
In theory, packets with non-zero precedence bits would jump to the head of transmission queues for each hop. As far as I know, TOS support has never been implemented in any network -- not even those belonging to the U.S. military.
You use whatever is handy (and working).
Wireless TCP/IP networks might be one of the last things left standing.
Also, TCP/IP networks with too many users will give slow service (until it gets so slow it breaks), whereas phones will completely block any calls above 100% load.
On the flip side, if you have a phone connection and the switches/lines aren't damaged and you aren't preempted (which GETS doesn't do, although it probably should *) you have a much more reliable connection than you would on a TCP/IP network.
*) If all circuits are busy, a GETS call won't get through until someone terminates one of their calls. Granted call terminations happen very often (whenever anyone on or through that switch hangs up) on a large switch - but it is still a delay.
Just because it CAN be done, doesn't mean it should!