To get into a debian release a new port has to go through several stages
1: someone has to create the port, get the basics working etc. 2: the porters must demonstrate it is appropriate for the ftp masters to accept it into unstable (this step often involves flamewars) 3: the port must demonstrate sufficient archive coverage, be sufficiently up to date*, have a usable installer (preferablly a port of the debian installer), have porterboxes and buildds availiable etc to get the release team to accept it into testing 4: the port must get established in testing (this takes time as binaries for the new port can only move to testing when 5: the port must survive the release process
Hurd got over step 2 a long time ago but has never managed step 3. In particular LARGE volumes of the archive still fail to build on hurd.
* By which I mean the binaries in unstable must have been successfully built from the most recent source in unstable. This is important for testing transitions because all ports are supposed to have the same version of each package in testing and the testing scripts enforce this by preventing transition until the packages in unstable are in sync (the testing scripts can be told to ignore this for a given package or architecture but it is not something done lightly)
Afaict the main reason ubuntu exists is because shuttleworth had his ideas on where he wanted to take linux and unlike most people he had the resources to seriously take it in his own direction rather than just fiddling arround at the edges. Further at the time Debian was in a crisis of ever increasing release cycle length with some people seriously wondering if the sarge cycle would ever end so there were a lot of debian "refugees" to give him an initial userbase.
Weight is proportional to the objects mass multiplied by the inverse square of altitude. Centrifugal force is proportional to the objects mass multiplied by altitude. Just to clarify those altitudes are relative to the center of the earth not to the earths surface.
First some definitions. For the purposes of this post
Tope is defined as a construction that is built to withstand tension but will simply bend out of the way if subject to compression. Further it has mass. The mass and strength of a rope may or may not be uniform (for a space elavator it most likely wouldn't be)
A space elavator is a rope running from the earths surface and held up by cetrifugal force.
The nice thing is there is no "you need to support the weight of the rope" non-sense because with centripetal forces, it all equates out and the rope doesn't have to support its own weight. It does kind of.
A mass that is stationary above the equator of a heavy rotating body such as the earth has an effective weight which is the objects weight minus the centrifugal force* acting on it. Weight is proportional to the objects mass multiplied by the inverse square of altitude. Centrifugal force is proportional to the objects mass multiplied by altitude. At some altitude those forces balance out and the effective weight becomes zero, we say an object that is above the equator at this altitude is in geostationary orbit. Above geostationary orbit the effective weight would be negative
This background leads us to the following statements about a space elavator.
1: To keep forces balanced the tension in the rope at any point must be the tension at the ground plus the effective weight below that point minus the effective weight above that point. 2: The rope at the ground must have a tension that is greater than or equal to zero (per defintion that a rope can't resist compressive forces) therefore the overall effective weight of the rope must be negative. This means that the rope (or a weight attatched to it) must extend beyond geostationary orbit. 3: The point of maximum tension in the rope is at geostationary orbit since all mass above geostationary orbit has a negative effective weight and all mass below has a positive effective weight. The rope at geostationary orbit must be able to support the effective weight of the entire run from geostationary orbit to the ground plus the tension at the ground anchor (you don't want this to be zero since it must remain positive even as dynamic changes to the system (such as a climber being attatched to the system) happen.
There is a proportion of people who can't be directly protected by vaccination. Sometimes there are other medical issues that make vaccinating them a bad idea. Sometimes they are too young. Sometimes people are given the vaccine but don't actually develop immunity.
These people are protected in modern society because as long as a sufficient proportion of people are vaccinated it is very difficult for the disease to spread effectively but if the proportion but if too many people start refusing vaccines that can break down.
Because they do still have customers who want/need support, updates (at least enough to keep it running on new hardware as their old hardware dies) licenses etc. That buisness is clearly worth something. How much is debatable but it's almost certainly not zero.
Maybe we could just whip her or cut-off a leg? She won't forget that and her family will get her back faster. Oh no I got it: tattoo her forehead CON-ARTIST!! Hm, but she might be able to remove it someday...
mmm, IMO the job of a punishment system should be threefold.
1: deterrent: If you do X we will do something nasty to you so don't do it. 2: rehabilitation: you commited a crime, you have been punished to keep the deterrent up now we need to make you a productive member of society again. 3: removal: if you have proven you can't be rehabilitated and your crimes are serious enough you need to be removed from society. For some crimes (mass murder) the risk to society is high enough that you need to be removed immediately skiping the rehabilitation step.
Personally I think punishments like whipping can fit into such a system and for some crimes may be a better option than jailtime or fines. Cutting off peoples appendages and/or tattoing CON-ARTIST on them would seem to go against rehabilitation (one thing you really don't want is to release criminals back into society in a way that leaves them with little option but to commit more crime).
Quite simple really you make packets to "long IPs" look like UDP packets to some reserved destination when passing over legacy infrastructure.
Actually some of the IPv6 transition methods aren't far off this, the trouble is 1: they were added as an afterthought and so aren't universally supported/enabled. 2: 6to4 doesn't work behind NAT at all and teredo works against NAT rather than working with it making it unessacerally complex and fragile.
The time when things get excruciatingly slow is when users computers think they have v6 connectivity but either the link is just dropping all packets (some tunnel systems are known to fail in this way) or there is no route to the destination in question and no "destination unreachable" messages coming back either (routing between 6to4/teredo and native v6 seems particularly flaky).
It's not a particularly common situation but afaict being in a situation where IPv6 gives better results than IPv4 is almost unheard of.
the v6 growth curve will still continue to follow its current expontential path for some time before colliding with stragglers in the comming years.
The question is what proportion of users will be "stragglers". So far i've not noticed IPv6 options when configuring any home router. Therefore my guess is there will be a hell of a lot of "stragglers", probablly over 50% of users here in the UK.
If 100% of users can access your website over IPv4 (some of them through a NAT) but only 50% can access it over IPv6 what is the motivation to offer both?
What if I'm a new ISP or hosting provider and I can't get any IP Addresses?? This qualifies as a crisis to me.
You buy a block from an ISP who has "recovered" it from end users by forcing them behind NAT. This won't come cheap but you won't have a lot of choice.
The big difference is that phones and phone exchanges are designed to support dialling with varying numbers of digits*. If you wanted to call me (and I gave you my phone number) you would have to dial 15 digits (3 digit international calling prefix 2 digit country code, 3 digit area code, 7 digit local number). Having an extra digit inserted in your phone numbers, having an area renumbered or having forced dialling of the whole national number introduced is an inconvenience but little more.
Internet addresses OTOH are fixed in size and while there almost certain were ways they could have hacked in longer addresses with minimal breakage they decided to go for a completely new protocol. Further the old API (in particular the call gethostbyname) only supported 32-bit addresses so that had to be changed too requiring application software to be updated.
*there is a limit set by the international standards but it's very high. An international number (including country code but excluding international calling prefix) is allowed to be up to 15 digits. Take off 3 digits for country code (country codes vary from 1-3 digits) and that means each country gets the ability to allocate at least 10^13 phone numbers. Given that the world population is under 10^10 and the highest population of a country with a 3 digit country code is well under 10^9 that is a LOT of phone numbers.
Possible but nontrivial. What you would essentially have to do is every time a DNS request was made allocate a "fake" v4 address from some pool. Then when traffic comes through for that IP perform an address and protocol translation before pushing it out to the v6 network.
I'm not sure if anyone has actually tried to implement such a thing or not.
Comcast, for one. IIRC they are offering tunnels to everyone but only offering native v6 in trial areas. Not sure if that counts.
And comcast are an ISP with far more motivation than most to go IPv6, their "control plane" network has already filled 10.0.0.0/8 and has spilled over into using public IPs! This makes deploying conventional carrier grade NAT rather difficult. From what I can gather they will be using "ds-lite" for providing v4 connectivity to those they can't give a public IP ("ds-lite" has the advantage over traditional nat of not needing a large private IP space within the ISP).
IPv6 will eventually impact every person and business
If your internal networks are never likely to outgrow 10.0.0.0/8 and your only use of the internet is to connect outbound to conventional "server" services and maybe run one or two public servers then any significant impact is likely to be a VERY long way down the road if it happens at all. Maybe you will pay a more for the handful of public IPs you use but it probablly won't be significant in the cost of running a business. Maybe the servers you connect to or the clients wanting to connect to your servers will eventually become v6 only but not for a LONG time. Likewise for end users who only access traditional server services there is likely to be little impact from being switched onto ISP based NAT.
On the other hand if you rely on making outbound connections to equipment on home grade connections and/or using protocols that use NAT traversal* you should be making contingency plans at this point. Most ISPs will probablly offer public v4 IPs for an extra charge but that extra charge may be rather substantial. They may also offer IPv6 but I wouldn't bet on it for the shittier end of ISPs. Similarly if you are planning a large buildout of public IP networks or have large existing blocks that are provider alllocated rather than provider independent you need to consider that things may get very expensive.
Regardless, evaluating a company's IPv6 readiness
IMO what you should really be investigating is what impact IPv4 exhaustion will have on you and your buisness partners. IPv6 may or may not play a part in mitigating the effects that exhaustion.
* Nat traversal works reliably with "full cone", "restricted cone" and "restricted port cone" NATs, doesn't work at all with symetric NATs and is unreliable with "port preservative" NATs. Further with "port preservative" NATs the probability of NAT traversal operating properly goes down as the load on the NAT increases. Linux does port preservative NAT. I'm not sure what most "carrier grade" NATs do.
Guess what has to be done to have IPv4 and IPv6 dual stack? The IP stack of every IPv4 host needs to be updated!
And so does the IP stack of every router in between. And then all the admins running those routers have double the workload because they have to route two completely serperate sets of addresses and then people resort to half baked tunnel solutions which makes website operators reluctant to offer IPv6 because it has a negative impact on their users average browsing experiance (at best they get an experience no better than they had with ipv4, at worst they have to wait for a timeout on every pageload which makes web browsing EXCRUCIATINGLY slow).
With clever design a "long IP" packet could have been made to look like a UDP packet to some "reserved" destination address when passing over legacy infrastructure. That way only the end systems and the default free zone* infrastructure would need to be updated. Existing home routers and ISP/corporate networks could remain untouched. If the design was done right even legacy NATs could have been left untouched.
IPv6 does have some transition systems but teredo is complex and fragile because it works against the NAT rather than with it and 6to4 only works if you have a public IPv4 IP on the box running 6to4.
But the time for argument on that is in the past now, we have no choice but to press ahead with some combination of IPv6 and natted IPv4. There is no way another new protocol will gain significant support in the time we have left before the RIRs start to run out of addresses.
*The default free zone is the group of routers that have no default route and instead have a routing table for the entire internet.
And likely every decade for the rest of the century we will now have problems because devs decided to kick the can down the road rather than fixing the 2 digit year problem properly.
don't know where you heard that, but it's not "apparent" at all. I heard it on/.
assuming the two units have similar antennas and similar processors, of course. Which is a rather big assumption, a cellphone GPS receiver is likely to be an extreme budget unit using whatever antenna they could squeeze in among the antenna(s) for the cellular functionality.
Apparently many cellphone GPS receivers are "tower assisted" and some of them struggle to get a GPS lock without the assistance of a cell tower. Not sure how the iphone stands in this regard.
"If you don't fill up there, you can very well not expect to ever see another human being again for as long as you live." IMO If forgetting to fill up on fuel results in your death then you probably weren't adequately prepared for the trip in the first place. If you are going into an area where you can't walk far enough to get help without needing water then you need to make provisions for how you will survive and either escape or get help if your car breaks down.
Part of the problem is people place blind trust in their GPS units. Just because the GPS unit has found a route doesn't mean it's a route you can drive without taking extra precautions but also IMO there need to be warnings on the ground of routes that should not be driven without special precautions.
Yes some people really are that stupid, I heard on the news of someone who drove into a body of water (not sure if it was a river the lake or the sea) because their GPS didn't indicate it was a ferry link rather than a road. Don't remember whether they died or not but it shows how stupid people can be.
A bigger problem over here in old blighty is articulated lorries getting stuck by driving down roads that are too narrow or otherwise unsuitable. One big problem in this case is it's virtually impossible to turn a lorry on a narrow road. So if the road starts looking bad the choices are to carry on and hope they don't get stuck, try to reverse out (very slow and likely to require a second person) or tow the lorry out.
It doesn't help that in britan we identify our roads based on how important they are in the network, not generally on how big they are.
There's not much you can do about MORONS, one way or another, they may kill themselves.
Everybody talks about how big of a pain in the ass IPv6 is going to be to deploy The big problems are
1: the chicken and egg problem. While everyone can access v4 servers there is little point in adding v6 to them and while a susbstantial portion of servers are v4 only ISPs won't want to make users v6 only. 2: infrastructure equipment, older home routers often don't do IPv6 at all and some older proffesional routers only do v6 in software which is much slower than doing v4 in hardware. 3: application software support. To support IPv6 apps must do certain things differently from with v4, it's not hugely difficult to port an app if you have the source but if you don't it's an extreme PITA to hack support in.
Be honest, is this all another Y2K where everything is just going to be smooth and a bunch of idiots that aren't talented enough to do the work but smart enough to know there is some kind of an issue are going to scream about the sky falling? What is almost certain to happen is that home lusers will be pushed behind ISP level v4 nat to free up IPs for more important/lucrative uses. They may or may not also offer IPv6. If you use anything other than traditional client apps (NAT traversal techniques get flakier the higher the load on the NAT and is incompatible with certain types of NAT so don't rely on apps that use it continuing to work) on a home connection you should expect to have problems and should press your ISP for information on their plans and/or make contingency plans for a situation where the only apps you can run on a home connection are traditional client apps.
Further anyone trying to expand their public IP networks is going to be in for a lot of pain. While ISPs will be able to provide IPs by recovering them from home lusers that is a lot of effort for the ISP and is almost certain to come at a price.
You have a few options for getting IPv6 connectivity if your ISP doesn't support it.
1: 6to4 , this is availiable to anyone who has a public v4 IP and requires no special agreements or accounts or anything like that. Downside is that routing between it and native v6 is less than perfect. 2: teredo, this sucks even more than 6to4 and I won't go into it here. The only reason to use it is if your system doesn't have a public v4 IP and even then a tunnel is probablly better 3: free tunnel providers, freenet6 doesn't even require an account and if you do want an account (main reason for this is if you want to connect a whole network rather than just a single IP) it's no hassle to get one, their performance used to be shit but things seem better now (though still not as good as a native connection). HE require an account but I haven't heard of anyone having problems getting one . sixxs is apparently the fastest but suffer from the "asshat elitists" problem.
The closest equivilent to debian stable on the ubuntu side of the fence would be the LTS releases. They have longer support lifecycles than debian stable and a similar release rate. Sounds good on paper particulally the fact that you get plenty of time to plan your upgrades (unlike debian which releases unpredictably and then gives you only about a year to plan and execute your upgrade)
The downside is the QA. Ubuntu have a largely fixed release cycle (they prioritize releasing on time over releasing right). They produce their LTS releases out of their standard 6 months development cycle (which means they are left with little time to get things stabalised and work the bugs out). They don't have the testing/unstable split that debian do. They also explicitly don't care about the packages in "universe" whereas debian at least in principle care about everything. Debian tend to spend a LONG time stablising and polishing before they release.
Depends whose definition of operating system you are using. The GNU guys use a very wide definition including things like the compiler, shell, command line utils etc but that is far from the only definition.
The problem is while yes you CAN focus sunlight with a mirror to a point where you can cut metal with it to do so you need a rather large and unweildly mirror. Further unless you add a lot of other complications* you are limited to heating things placed between the mirror and the sun. That's fine if you are trying to cut a loose bar not so good if you are trying to cut holes in a large sheet or trying to cut stuff in situ.
Lenses can heat stuff behind the lens but have other problems (size, weight, cost, different behaviour for different wavelengths of light).
* telescopes typically get arround this by having a small mirror near the focus to reflect the light out but telescopes don't have to survive being pointed directly at the sun..
Slashdot Mirror
To get into a debian release a new port has to go through several stages
1: someone has to create the port, get the basics working etc.
2: the porters must demonstrate it is appropriate for the ftp masters to accept it into unstable (this step often involves flamewars)
3: the port must demonstrate sufficient archive coverage, be sufficiently up to date*, have a usable installer (preferablly a port of the debian installer), have porterboxes and buildds availiable etc to get the release team to accept it into testing
4: the port must get established in testing (this takes time as binaries for the new port can only move to testing when
5: the port must survive the release process
Hurd got over step 2 a long time ago but has never managed step 3. In particular LARGE volumes of the archive still fail to build on hurd.
* By which I mean the binaries in unstable must have been successfully built from the most recent source in unstable. This is important for testing transitions because all ports are supposed to have the same version of each package in testing and the testing scripts enforce this by preventing transition until the packages in unstable are in sync (the testing scripts can be told to ignore this for a given package or architecture but it is not something done lightly)
That's the whole reason Ubuntu exists though
Afaict the main reason ubuntu exists is because shuttleworth had his ideas on where he wanted to take linux and unlike most people he had the resources to seriously take it in his own direction rather than just fiddling arround at the edges. Further at the time Debian was in a crisis of ever increasing release cycle length with some people seriously wondering if the sarge cycle would ever end so there were a lot of debian "refugees" to give him an initial userbase.
Weight is proportional to the objects mass multiplied by the inverse square of altitude. Centrifugal force is proportional to the objects mass multiplied by altitude.
Just to clarify those altitudes are relative to the center of the earth not to the earths surface.
First some definitions. For the purposes of this post
Tope is defined as a construction that is built to withstand tension but will simply bend out of the way if subject to compression. Further it has mass. The mass and strength of a rope may or may not be uniform (for a space elavator it most likely wouldn't be)
A space elavator is a rope running from the earths surface and held up by cetrifugal force.
The nice thing is there is no "you need to support the weight of the rope" non-sense because with centripetal forces, it all equates out and the rope doesn't have to support its own weight.
It does kind of.
A mass that is stationary above the equator of a heavy rotating body such as the earth has an effective weight which is the objects weight minus the centrifugal force* acting on it. Weight is proportional to the objects mass multiplied by the inverse square of altitude. Centrifugal force is proportional to the objects mass multiplied by altitude. At some altitude those forces balance out and the effective weight becomes zero, we say an object that is above the equator at this altitude is in geostationary orbit. Above geostationary orbit the effective weight would be negative
This background leads us to the following statements about a space elavator.
1: To keep forces balanced the tension in the rope at any point must be the tension at the ground plus the effective weight below that point minus the effective weight above that point.
2: The rope at the ground must have a tension that is greater than or equal to zero (per defintion that a rope can't resist compressive forces) therefore the overall effective weight of the rope must be negative. This means that the rope (or a weight attatched to it) must extend beyond geostationary orbit.
3: The point of maximum tension in the rope is at geostationary orbit since all mass above geostationary orbit has a negative effective weight and all mass below has a positive effective weight. The rope at geostationary orbit must be able to support the effective weight of the entire run from geostationary orbit to the ground plus the tension at the ground anchor (you don't want this to be zero since it must remain positive even as dynamic changes to the system (such as a climber being attatched to the system) happen.
* http://en.wikipedia.org/wiki/Centrifugal_force#Fictitious_centrifugal_force
There is a proportion of people who can't be directly protected by vaccination. Sometimes there are other medical issues that make vaccinating them a bad idea. Sometimes they are too young. Sometimes people are given the vaccine but don't actually develop immunity.
These people are protected in modern society because as long as a sufficient proportion of people are vaccinated it is very difficult for the disease to spread effectively but if the proportion but if too many people start refusing vaccines that can break down.
Because they do still have customers who want/need support, updates (at least enough to keep it running on new hardware as their old hardware dies) licenses etc. That buisness is clearly worth something. How much is debatable but it's almost certainly not zero.
Maybe we could just whip her or cut-off a leg? She won't forget that and her family will get her back faster. Oh no I got it: tattoo her forehead CON-ARTIST!! Hm, but she might be able to remove it someday...
mmm, IMO the job of a punishment system should be threefold.
1: deterrent: If you do X we will do something nasty to you so don't do it.
2: rehabilitation: you commited a crime, you have been punished to keep the deterrent up now we need to make you a productive member of society again.
3: removal: if you have proven you can't be rehabilitated and your crimes are serious enough you need to be removed from society. For some crimes (mass murder) the risk to society is high enough that you need to be removed immediately skiping the rehabilitation step.
Personally I think punishments like whipping can fit into such a system and for some crimes may be a better option than jailtime or fines. Cutting off peoples appendages and/or tattoing CON-ARTIST on them would seem to go against rehabilitation (one thing you really don't want is to release criminals back into society in a way that leaves them with little option but to commit more crime).
Quite simple really you make packets to "long IPs" look like UDP packets to some reserved destination when passing over legacy infrastructure.
Actually some of the IPv6 transition methods aren't far off this, the trouble is 1: they were added as an afterthought and so aren't universally supported/enabled. 2: 6to4 doesn't work behind NAT at all and teredo works against NAT rather than working with it making it unessacerally complex and fragile.
The time when things get excruciatingly slow is when users computers think they have v6 connectivity but either the link is just dropping all packets (some tunnel systems are known to fail in this way) or there is no route to the destination in question and no "destination unreachable" messages coming back either (routing between 6to4/teredo and native v6 seems particularly flaky).
It's not a particularly common situation but afaict being in a situation where IPv6 gives better results than IPv4 is almost unheard of.
the v6 growth curve will still continue to follow its current expontential path for some time before colliding with stragglers in the comming years.
The question is what proportion of users will be "stragglers". So far i've not noticed IPv6 options when configuring any home router. Therefore my guess is there will be a hell of a lot of "stragglers", probablly over 50% of users here in the UK.
If 100% of users can access your website over IPv4 (some of them through a NAT) but only 50% can access it over IPv6 what is the motivation to offer both?
What if I'm a new ISP or hosting provider and I can't get any IP Addresses?? This qualifies as a crisis to me.
You buy a block from an ISP who has "recovered" it from end users by forcing them behind NAT. This won't come cheap but you won't have a lot of choice.
The big difference is that phones and phone exchanges are designed to support dialling with varying numbers of digits*. If you wanted to call me (and I gave you my phone number) you would have to dial 15 digits (3 digit international calling prefix 2 digit country code, 3 digit area code, 7 digit local number). Having an extra digit inserted in your phone numbers, having an area renumbered or having forced dialling of the whole national number introduced is an inconvenience but little more.
Internet addresses OTOH are fixed in size and while there almost certain were ways they could have hacked in longer addresses with minimal breakage they decided to go for a completely new protocol. Further the old API (in particular the call gethostbyname) only supported 32-bit addresses so that had to be changed too requiring application software to be updated.
*there is a limit set by the international standards but it's very high. An international number (including country code but excluding international calling prefix) is allowed to be up to 15 digits. Take off 3 digits for country code (country codes vary from 1-3 digits) and that means each country gets the ability to allocate at least 10^13 phone numbers. Given that the world population is under 10^10 and the highest population of a country with a 3 digit country code is well under 10^9 that is a LOT of phone numbers.
Possible but nontrivial. What you would essentially have to do is every time a DNS request was made allocate a "fake" v4 address from some pool. Then when traffic comes through for that IP perform an address and protocol translation before pushing it out to the v6 network.
I'm not sure if anyone has actually tried to implement such a thing or not.
Comcast, for one.
IIRC they are offering tunnels to everyone but only offering native v6 in trial areas. Not sure if that counts.
And comcast are an ISP with far more motivation than most to go IPv6, their "control plane" network has already filled 10.0.0.0/8 and has spilled over into using public IPs! This makes deploying conventional carrier grade NAT rather difficult. From what I can gather they will be using "ds-lite" for providing v4 connectivity to those they can't give a public IP ("ds-lite" has the advantage over traditional nat of not needing a large private IP space within the ISP).
http://www.nanog.org/meetings/nanog37/presentations/alain-durand.pdf
IPv6 will eventually impact every person and business
If your internal networks are never likely to outgrow 10.0.0.0/8 and your only use of the internet is to connect outbound to conventional "server" services and maybe run one or two public servers then any significant impact is likely to be a VERY long way down the road if it happens at all. Maybe you will pay a more for the handful of public IPs you use but it probablly won't be significant in the cost of running a business. Maybe the servers you connect to or the clients wanting to connect to your servers will eventually become v6 only but not for a LONG time. Likewise for end users who only access traditional server services there is likely to be little impact from being switched onto ISP based NAT.
On the other hand if you rely on making outbound connections to equipment on home grade connections and/or using protocols that use NAT traversal* you should be making contingency plans at this point. Most ISPs will probablly offer public v4 IPs for an extra charge but that extra charge may be rather substantial. They may also offer IPv6 but I wouldn't bet on it for the shittier end of ISPs. Similarly if you are planning a large buildout of public IP networks or have large existing blocks that are provider alllocated rather than provider independent you need to consider that things may get very expensive.
Regardless, evaluating a company's IPv6 readiness
IMO what you should really be investigating is what impact IPv4 exhaustion will have on you and your buisness partners. IPv6 may or may not play a part in mitigating the effects that exhaustion.
* Nat traversal works reliably with "full cone", "restricted cone" and "restricted port cone" NATs, doesn't work at all with symetric NATs and is unreliable with "port preservative" NATs. Further with "port preservative" NATs the probability of NAT traversal operating properly goes down as the load on the NAT increases. Linux does port preservative NAT. I'm not sure what most "carrier grade" NATs do.
Guess what has to be done to have IPv4 and IPv6 dual stack? The IP stack of every IPv4 host needs to be updated!
And so does the IP stack of every router in between. And then all the admins running those routers have double the workload because they have to route two completely serperate sets of addresses and then people resort to half baked tunnel solutions which makes website operators reluctant to offer IPv6 because it has a negative impact on their users average browsing experiance (at best they get an experience no better than they had with ipv4, at worst they have to wait for a timeout on every pageload which makes web browsing EXCRUCIATINGLY slow).
With clever design a "long IP" packet could have been made to look like a UDP packet to some "reserved" destination address when passing over legacy infrastructure. That way only the end systems and the default free zone* infrastructure would need to be updated. Existing home routers and ISP/corporate networks could remain untouched. If the design was done right even legacy NATs could have been left untouched.
IPv6 does have some transition systems but teredo is complex and fragile because it works against the NAT rather than with it and 6to4 only works if you have a public IPv4 IP on the box running 6to4.
But the time for argument on that is in the past now, we have no choice but to press ahead with some combination of IPv6 and natted IPv4. There is no way another new protocol will gain significant support in the time we have left before the RIRs start to run out of addresses.
*The default free zone is the group of routers that have no default route and instead have a routing table for the entire internet.
And likely every decade for the rest of the century we will now have problems because devs decided to kick the can down the road rather than fixing the 2 digit year problem properly.
don't know where you heard that, but it's not "apparent" at all. /.
I heard it on
assuming the two units have similar antennas and similar processors, of course.
Which is a rather big assumption, a cellphone GPS receiver is likely to be an extreme budget unit using whatever antenna they could squeeze in among the antenna(s) for the cellular functionality.
Apparently many cellphone GPS receivers are "tower assisted" and some of them struggle to get a GPS lock without the assistance of a cell tower. Not sure how the iphone stands in this regard.
"If you don't fill up there, you can very well not expect to ever see another human being again for as long as you live."
IMO If forgetting to fill up on fuel results in your death then you probably weren't adequately prepared for the trip in the first place. If you are going into an area where you can't walk far enough to get help without needing water then you need to make provisions for how you will survive and either escape or get help if your car breaks down.
Part of the problem is people place blind trust in their GPS units. Just because the GPS unit has found a route doesn't mean it's a route you can drive without taking extra precautions but also IMO there need to be warnings on the ground of routes that should not be driven without special precautions.
Yes some people really are that stupid, I heard on the news of someone who drove into a body of water (not sure if it was a river the lake or the sea) because their GPS didn't indicate it was a ferry link rather than a road. Don't remember whether they died or not but it shows how stupid people can be.
A bigger problem over here in old blighty is articulated lorries getting stuck by driving down roads that are too narrow or otherwise unsuitable. One big problem in this case is it's virtually impossible to turn a lorry on a narrow road. So if the road starts looking bad the choices are to carry on and hope they don't get stuck, try to reverse out (very slow and likely to require a second person) or tow the lorry out.
It doesn't help that in britan we identify our roads based on how important they are in the network, not generally on how big they are.
There's not much you can do about MORONS, one way or another, they may kill themselves.
Agreed
Everybody talks about how big of a pain in the ass IPv6 is going to be to deploy
The big problems are
1: the chicken and egg problem. While everyone can access v4 servers there is little point in adding v6 to them and while a susbstantial portion of servers are v4 only ISPs won't want to make users v6 only.
2: infrastructure equipment, older home routers often don't do IPv6 at all and some older proffesional routers only do v6 in software which is much slower than doing v4 in hardware.
3: application software support. To support IPv6 apps must do certain things differently from with v4, it's not hugely difficult to port an app if you have the source but if you don't it's an extreme PITA to hack support in.
Be honest, is this all another Y2K where everything is just going to be smooth and a bunch of idiots that aren't talented enough to do the work but smart enough to know there is some kind of an issue are going to scream about the sky falling?
What is almost certain to happen is that home lusers will be pushed behind ISP level v4 nat to free up IPs for more important/lucrative uses. They may or may not also offer IPv6. If you use anything other than traditional client apps (NAT traversal techniques get flakier the higher the load on the NAT and is incompatible with certain types of NAT so don't rely on apps that use it continuing to work) on a home connection you should expect to have problems and should press your ISP for information on their plans and/or make contingency plans for a situation where the only apps you can run on a home connection are traditional client apps.
Further anyone trying to expand their public IP networks is going to be in for a lot of pain. While ISPs will be able to provide IPs by recovering them from home lusers that is a lot of effort for the ISP and is almost certain to come at a price.
You have a few options for getting IPv6 connectivity if your ISP doesn't support it.
1: 6to4 , this is availiable to anyone who has a public v4 IP and requires no special agreements or accounts or anything like that. Downside is that routing between it and native v6 is less than perfect.
2: teredo, this sucks even more than 6to4 and I won't go into it here. The only reason to use it is if your system doesn't have a public v4 IP and even then a tunnel is probablly better
3: free tunnel providers, freenet6 doesn't even require an account and if you do want an account (main reason for this is if you want to connect a whole network rather than just a single IP) it's no hassle to get one, their performance used to be shit but things seem better now (though still not as good as a native connection). HE require an account but I haven't heard of anyone having problems getting one . sixxs is apparently the fastest but suffer from the "asshat elitists" problem.
The closest equivilent to debian stable on the ubuntu side of the fence would be the LTS releases. They have longer support lifecycles than debian stable and a similar release rate. Sounds good on paper particulally the fact that you get plenty of time to plan your upgrades (unlike debian which releases unpredictably and then gives you only about a year to plan and execute your upgrade)
The downside is the QA. Ubuntu have a largely fixed release cycle (they prioritize releasing on time over releasing right). They produce their LTS releases out of their standard 6 months development cycle (which means they are left with little time to get things stabalised and work the bugs out). They don't have the testing/unstable split that debian do. They also explicitly don't care about the packages in "universe" whereas debian at least in principle care about everything. Debian tend to spend a LONG time stablising and polishing before they release.
Depends whose definition of operating system you are using. The GNU guys use a very wide definition including things like the compiler, shell, command line utils etc but that is far from the only definition.
The problem is while yes you CAN focus sunlight with a mirror to a point where you can cut metal with it to do so you need a rather large and unweildly mirror. Further unless you add a lot of other complications* you are limited to heating things placed between the mirror and the sun. That's fine if you are trying to cut a loose bar not so good if you are trying to cut holes in a large sheet or trying to cut stuff in situ.
Lenses can heat stuff behind the lens but have other problems (size, weight, cost, different behaviour for different wavelengths of light).
* telescopes typically get arround this by having a small mirror near the focus to reflect the light out but telescopes don't have to survive being pointed directly at the sun..