During a week long barrage of tests, I was able to sustain my full symmetrical speeds to NewYork during peak hours. This is a little over a 2000 mile round trip for me, according to Google Maps. While doing this test, my up/down was nearly perfectly flat with sustained transfer within +-1% of my allotted rate, while maintaining a 31ms ping +-1ms. zero dropped ping packets over a 10 minute window, while the transfer was going on.
My biggest issue was finding a server that I could test my bandwidth on.
The other cool part about Active Ethernet is that I get a 0.6ms ping to my ISP according to HRPing. My ping to Chicago is now lower than the ping to the first hop in my city when I was on Cable Internet.
"Over Engineering" just means you don't know what you're doing. Typically thinking and claiming you're doing one thing, but you're actually doing another, which gives a bad name for what you're claiming to be doing.
I got Active Gigabit Ethernet fiber from my ISP, which the entire city is getting. I started a ping -t from my work, which has a 10Gb fiber connection, but has a different upstream provider, so the trace route goes from MidWest to Texas back up to MidWest, which is about a 2500 mile round-trip according to Google Maps. After 550k+ pings over a 7 day period, I had about 0.08% packetloss.
So, couldn't find any connection technology that could beat a T1?.. Ha!
While NAT has not been designed to be a security mechanism, it blocks incoming connections, so it can be seen to increase security regarding network attacks against the machine.
Actually it doesn't block incoming connections, the stateful-firewall does. NAT is implemented however the implementer wants to, as it is not a standard. It is a hack that has no security guarantees and only needs to work good enough to sell devices. There is nothing saying that NAT has to work a specific way or needs to cover certain cases. Many implementations of "NAT" have security holes, even on high end enterprise equipment. Why? Because there is no wrong way to implement it.
At this point, it's not the infrastructure that needs to be updated. The backbone of the Internet has been IPv6 for almost a decade now and almost all DSL/Cable hardware is IPv6 native. The only real stuff that needs to get updated is ISPs actually configuring their hardware and end-users having IPv6 capable NAT/Routers.
Some currently theorize that life increases the rate of entropy in system with very low entropy. You also have the whole "literally anything could happen in our universe because of quantum-randomness". Our life could be just a random occurrence with low probability.
I plan on increasing my income by 2x every 18 months. I'll be rich!
Just because one plans for it doesn't mean that they can do it. Being able to maintain this pace without some huge increase in R&D is what is so incredible.
When someone makes a branch, does some changes, then submits the diff via git, what license is that patch under? I assume that git doesn't explicitly attach a licence agreement to every diff submitted.
Then that begs the question. If by default, all patches do not have a licence, are they assumed to be public domain? Assuming all of these patches are public domain, if I take enough patches going back far enough, I will have pretty much the entire source-code. This means I could just aggregate the patches together and claim the program is public domain.
Modifications to GPL must remain GPL, so unless the government is not working with GPL code, how could they know how to create the patch in the first place?
The problem is GPL. Linux is popular. Yes, the government is big enough that they could issue their own fixes, but any work that the government does MUST be compatible with public domain because it is tax payer money paying for that work. Guess what, GPL is not compatible. MIT/BSD/etc are.
Want the government to use OpenSource, get rid of the restrictions on GPL.
There are so many government funded research projects that start on BSD/etc because of the GPL restrictions. I guess that's helping BSD.
but I can't use a desktop environment that causes my laptop fan to run constantly
If the system is not doing anything productive, then it should not be using system resources. There is no reason for his fans to spin up unless he's actually doing something.
Let me fix that: If you purchased an "unlimited" rope that was rated for "any car", and you tried to haul a fully-loaded dump-truck..... Unlimited "with in reason".
But no definition of what a "server" is. There is a huge grey area, which this person went past, but in general, many every day applications run both as servers and clients.
My ISP say "no servers of any kind", but they says BitTorrent is fine.. well, wtf? Please define "servers". P2P is fine, SSH is fine, remote VPN is fine.. wtf? They're all "servers" by technical definitions.
FTTH is almost always cheaper, it just requires a bit more up-front, but it only takes a few years to pay it off.
The only reason companies don't do FTTH is because it makes for a bad quarter report compared to less capital intensive upgrades of old-copper that allows the ISP to charge more and increase revenue.
Slashdot Mirror
During a week long barrage of tests, I was able to sustain my full symmetrical speeds to NewYork during peak hours. This is a little over a 2000 mile round trip for me, according to Google Maps. While doing this test, my up/down was nearly perfectly flat with sustained transfer within +-1% of my allotted rate, while maintaining a 31ms ping +-1ms. zero dropped ping packets over a 10 minute window, while the transfer was going on.
My biggest issue was finding a server that I could test my bandwidth on.
The other cool part about Active Ethernet is that I get a 0.6ms ping to my ISP according to HRPing. My ping to Chicago is now lower than the ping to the first hop in my city when I was on Cable Internet.
"Over Engineering" just means you don't know what you're doing. Typically thinking and claiming you're doing one thing, but you're actually doing another, which gives a bad name for what you're claiming to be doing.
Don't you know? Most upstream providers have Layer7 firewalls that can drop non-player EvE logins. The secret is the magic pixie dust.
Many even have kids in the background. I get "brb, gotta put the kids to sleep" and other stuff.
Their entire network is down.
My 1Gb fiber connection is good for 20km. Kind of cool. Got one of these in my basement http://calix.com/images/products/p-series/calix_716GE-I.png
Some of the newer models can handle 40km or even 80km.
I got Active Gigabit Ethernet fiber from my ISP, which the entire city is getting. I started a ping -t from my work, which has a 10Gb fiber connection, but has a different upstream provider, so the trace route goes from MidWest to Texas back up to MidWest, which is about a 2500 mile round-trip according to Google Maps. After 550k+ pings over a 7 day period, I had about 0.08% packetloss.
So, couldn't find any connection technology that could beat a T1?.. Ha!
While NAT has not been designed to be a security mechanism, it blocks incoming connections, so it can be seen to increase security regarding network attacks against the machine.
Actually it doesn't block incoming connections, the stateful-firewall does. NAT is implemented however the implementer wants to, as it is not a standard. It is a hack that has no security guarantees and only needs to work good enough to sell devices. There is nothing saying that NAT has to work a specific way or needs to cover certain cases. Many implementations of "NAT" have security holes, even on high end enterprise equipment. Why? Because there is no wrong way to implement it.
USA made the Internet, USA brags about being the "best". Anything less than 1st is "lagging".
At this point, it's not the infrastructure that needs to be updated. The backbone of the Internet has been IPv6 for almost a decade now and almost all DSL/Cable hardware is IPv6 native. The only real stuff that needs to get updated is ISPs actually configuring their hardware and end-users having IPv6 capable NAT/Routers.
Some currently theorize that life increases the rate of entropy in system with very low entropy. You also have the whole "literally anything could happen in our universe because of quantum-randomness". Our life could be just a random occurrence with low probability.
Numbers don't matter, only rates do.
Just as long as you remembered to "park" your harddrive before shutting down. You didn't want to head-crashing.
I plan on increasing my income by 2x every 18 months. I'll be rich!
Just because one plans for it doesn't mean that they can do it. Being able to maintain this pace without some huge increase in R&D is what is so incredible.
When someone makes a branch, does some changes, then submits the diff via git, what license is that patch under? I assume that git doesn't explicitly attach a licence agreement to every diff submitted.
Then that begs the question. If by default, all patches do not have a licence, are they assumed to be public domain? Assuming all of these patches are public domain, if I take enough patches going back far enough, I will have pretty much the entire source-code. This means I could just aggregate the patches together and claim the program is public domain.
Sounds like a very slippery slope.
Modifications to GPL must remain GPL, so unless the government is not working with GPL code, how could they know how to create the patch in the first place?
The problem is GPL. Linux is popular. Yes, the government is big enough that they could issue their own fixes, but any work that the government does MUST be compatible with public domain because it is tax payer money paying for that work. Guess what, GPL is not compatible. MIT/BSD/etc are.
Want the government to use OpenSource, get rid of the restrictions on GPL.
There are so many government funded research projects that start on BSD/etc because of the GPL restrictions. I guess that's helping BSD.
but I can't use a desktop environment that causes my laptop fan to run constantly
If the system is not doing anything productive, then it should not be using system resources. There is no reason for his fans to spin up unless he's actually doing something.
Not "mini", just "ultra-short".
Nothing in this universe is unlimited, so we might as well get rid of the word.
Let me fix that: If you purchased an "unlimited" rope that was rated for "any car", and you tried to haul a fully-loaded dump-truck..... Unlimited "with in reason".
But no definition of what a "server" is. There is a huge grey area, which this person went past, but in general, many every day applications run both as servers and clients.
My ISP say "no servers of any kind", but they says BitTorrent is fine.. well, wtf? Please define "servers". P2P is fine, SSH is fine, remote VPN is fine.. wtf? They're all "servers" by technical definitions.
http://en.wikipedia.org/wiki/Asynchronous_Transfer_Mode They had a reason
FTTH is almost always cheaper, it just requires a bit more up-front, but it only takes a few years to pay it off.
The only reason companies don't do FTTH is because it makes for a bad quarter report compared to less capital intensive upgrades of old-copper that allows the ISP to charge more and increase revenue.
For what it's worth, it's not providing any more bandwidth than the old technique, which had 80 channels at 10Gbps each.
Post:
which was not able to carry 10Gbps channels using older technology