> Huh? The POSIX time specification make it trivial to calculate dates using simple arithmetic
True, but the same property also makes the use of POSIX time for precision timing basically suicidal. POSIX time is a convenient and adequate encoding of civil time, as long as you do not need more than one second of accuracy.
If you want to reliably measure or timestamp anything with more than one second of accuracy you should be using a clock derived or offset from a reliable clock instead. The use of POSIX time for precision timekeeping - even use by such rudimentary applications as 'make' - was defective from the beginning. NTP is equally defective as a consequence.
Since the IETF saw that there was gonna be an industry-wide overhaul in any case, it did this complete overhaul, tossing in everything learnt in the years of IPv4, so that another IP transition won't be likely in the next 50 years, if ever.
By this point, even the luminaries at the IETF have realized that the design for IPv6 as a replacement for IPv4 is fatally flawed. How flawed? Flawed enough that there is a high probability that a worldwide transition to IPv6 will never actually happen.
Now sure, there are technical advantages to a clean slate design, but a clean slate design is also unfortunately almost useless as a replacement for IPv4 in the real world. There is no incremental advantage and extraordinarily high costs to adding a separate numbering plan to an existing network, so no cost conscious organization ever does it unless they are forced to, and probably never will.
At this point I would lay odds on an IPv7 eventually being developed that is a revision of IPv6 with the incorporation of the IPv4 address space in a routeable fashion, and which assigns each IPv4 address a network prefix that an entire subnet of devices may eventually be directly addressed behind, in addition to the default.
Why? Because doing anything else would be one of the biggest wastes of resources the world has ever seen.
Any downsides? An IPv7 router would have bigger routing tables than an IPv6 only router, but the routing tables could be used to route IPv4 packets, and as it is not likely IPv4 is going away anytime soon, the same overhead is there one way or another.
A wide scale deployment of IPv7 would require hardware upgrades in some cases, but for most people it could be deployed silently, without them ever needing to know or care. A simple software update would be all that was necessary, and a few years down the road nearly all IPv4 capable devices would handle the expanded address space in a usable fashion without any renumbering or other configuration changes. That would save billions of dollars a year in unnecessary administration costs worldwide.
There are only eight pages of new rules. The rest is explanation, history, legal justification, and commentary. More here: http://e-pluribusunum.com/2015...
The FCC doesn't have a problem with prioritization per se. The FCC has a problem with paid prioritization, which on a congested network can starve other traffic, and cause other problems, depending on how it is done.
Under any reasonable interpretation of the law, Internet access providers always have been common carriers. The FCC, in a classic example of regulatory capture, simply decided to interpret the law in a relatively perverse manner, by pretending that broadband Internet access providers were "information services" rather than "telecommunications services", which is flat out ridiculous, and the Supreme Court decided to defer to them.
This is the legal definition of telecommunications for example:
The term "telecommunications" means the transmission, between or among points specified by the user, of information of the user's choosing, without change in the form or content of the information as sent and received. (47 USC 153)
Sound familiar? Sounds just like Internet access. How about this one:
The term "telecommunications service" means the offering of telecommunications for a fee directly to the public, or to such classes of users as to be effectively available directly to the public, regardless of the facilities used.
Justice Scalia pointed this out several years ago, but he was in the minority on this one. The justices in the majority said, well it may not make any sense, but we will let the FCC decide. Now rationality has returned to the FCC and they are revisiting the question.
"A utility shall provide a cable television system or any telecommunications carrier with nondiscriminatory access to any pole, duct, conduit, or right-of-way owned or controlled by it." (47 USC 224)
Comcast has this right by virtue of being a "cable television system". The major phone companies have it because they are "telecommunications carriers". But facilities based ISPs like Google Fiber are currently (and incorrectly) classified as neither so they are out of luck until sanity finishes kicking in at the FCC.
XFS is not a copy on write filesystem, by the way. ZFS and BTRFS should work definitely work better, but they might need some internal tweaks to make the best use of it.
The Apple IIgs was dramatically different from all other Apple II models. It was backward compatible, but came with a 16 bit processor (the 65816), much more RAM (256K or more), greatly improved sound and video, and a GUI shell much like that of the Mac, plus color, which nearly all Macs lacked at the time. It was a little underpowered compared to the 68000 based Mac, Amiga, and Atari ST, but a more than respectable upgrade to the Apple II series nonetheless.
As educational / entertainment devices even the older Apple IIs ran circles around the PC until EGA was widely deployed in the late 1980s. PC games inevitably were designed for CGA graphics, with a fixed set of four unimaginative colors at a time. The Apple II was better than that almost ten years earlier, to say nothing of the much less expensive Commodore 64. The PC was intended primarily for business purposes, and it showed.
The Apple II wouldn't be more than a footnote in history if those prices didn't go way down, which they did. By the mid 1980s virtually every school in the country had a classroom full of them.
If you asked the creators, they would probably be embarrassed to call it an operating system at all. Apple DOS didn't handle keyboard support, video support, sound support, or printer support. That was all handled using either the monitor (a BIOS in ROM that was not part of DOS), peripheral card ROMs in some cases, or by direct access to the hardware.
MS-DOS was similar. It handled file I/O and that is it. A disk operating system, not a computer operating system. The BIOS was separate, not controlled by Microsoft, but rather by the PC manufacturer, and in ROM.
A stateful firewall doesn't need to block transport layer protocols it doesn't understand in order to provide a meaningful level of security. All it needs to do is block packets from IP addresses that corresponding interior address has not recently communicated with, with a reasonable time out. UDP is handled much the same way today.
If the developers of stateful IPv6 firewalls do not ship devices with such a reasonable configuration by default, they will block the deployment of new transport protocols indefinitely - at least all those that do not resort to the awkward expedient of running on top of UDP.
Blocking new transport protocols developers can reasonably handle with a standard policy is bad for efficiency, power consumption, latency, user experience, and so on in the long run - TCP is far from ideal as a transport protocol goes. In a number of ways it is outright backwards. If you want to impede the long term development of the Internet, degrading the end-to-end principle unnecessarily is a good place to start.
Work is underway for concurrent multipath transfer for SCTP as well. Also known as CMT-SCTP. There are significant challenges in doing this sort of thing though. SCTP wasn't designed for CMT, and probably needs much more radical changes than the current architects are proposing to do it well.
Changes like subflows with independent sequence numbers and congestion windows, to start with. SCTP is much further ahead in the connection handling and security department, but MPTCP has the odd advantage of resorting to independent subflows to begin with, and if it can handle path failure properly, it might well be ahead in the CMT game, if byte stream semantics are all you need.
On the contrary, SCTP is a transport protocol just like TCP, except with a large number of added features. The main problem with SCTP has nothing to do with SCTP at all. It is that NAT devices do not support any transport protocol that they haven't been programmed for in advance. This makes SCTP next to impossible to deploy on a broad scale - NAT, that wart upon router-kind, is ubiquitous.
TCP would have exactly the same problem if it were a new protocol. A NAT device requires relatively deep knowledge of TCP to support it at all. It play games with both ports and addresses, keeps track of connection state, and so on. Ordinary routers do no such thing. A NAT device is a transport layer proxy by another name.
No, he said that the government breaking up the monopoly was "criminal"
I guess that is why one should quote a line or two of the comment he or she is responding to. The one you were actually responding to was hidden.
And when I say "simple engineering", I mean standard engineering practice. No one in his right mind designs a mission critical system for tens of thousands of people without full redundancy and usually automatic failover as well. A system that size that goes down for more than five minutes is more likely to be the result of engineering malpractice than anything else.
I still have a traditional land line, because it is reliable. I can't remember ever picking it the phone up and not having a dial tone. That would most definitely not be the case with any other kind of service available to residences and small businesses. The PSTN was designed to be reliable, and the people who designed readily available Internet access service designed it by the same standards that one might use for toys. Entertainment devices.
No one cares very much if they can't watch I Love Lucy for a couple of hours, although every television station I am aware of is about three orders of magnitude more reliable than the DSL / cable monopolies as well, if not quite as reliable as the PSTN.
That is quite the red herring there. Did the poster say we should restore the AT&T monopoly? No. What he said (in so many words) is that the systems were engineered to such high standards that they never went down.
The suggestion that we would have to return to monopoly pricing in order for Internet access providers to give reliable service is preposterous. It takes some simple engineering is all. Which the current folks at AT&T (formerly Southeastern Bell) are apparently either incapable of or are prevented from doing. The folks at Comcast and CenturyLink are similar.
If these quasi-monopoly providers can't deliver basic service that doesn't take large areas down for hours on end, they are inviting government regulation to make them engineer their systems so they don't. Not price regulation, engineering standards regulation.
Personally, I appreciate it when I cross a bridge and it doesn't collapse under the offered weight. That doesn't seem to be a particularly high priority at most of the monster ISPs these days - quite the opposite. Lets make the most unreliable service we can get away with seems to be the order of the day.
This is intended for server use. No video output required. The PCIe part is actually optional. I wouldn't expect to see this in workstations anytime soon, not without a major redesign at any rate. The form factor is designed for small, low power processors. The interface is not designed for SMP or anything like that either.
It is not a CPU slot specification. It is a micro-server slot specification, which is much more practical. Think small form factor blade server. The PCIe part is actually optional.
Basically, they are sawing the motherboard in 2, where the CPU and memory are on the daugterboard, and the rest of the components (SATA,USB3, PCIe slots, sound, video outputs) are all that remain on the motherboard
More important, it means that elements of memory and processing that now must be fixed closely together can be separated within a rack, and used as needed for different kinds of tasks.
This statement is in reference to Intel's proposal, which is still vaporware. I seriously doubt they are talking about locating main memory away from the processors. That would more or less be suicidal.
Here is a link to an actual specification. If you read it, you will see that about half of what has been written about this announcement is wildly off base. We are talking micro-servers here - complete with on board cpu, ram, boot eeprom, flash storage, and ethernet. PCIe and SATA connections to the backplane are optional. Think small form factor blade server. http://www.opencompute.org/wp/wp-content/uploads/2013/01/Open_Compute_Project_Micro-Server_Card_Specification_v0.5.pdf
Slashdot Mirror
> Huh? The POSIX time specification make it trivial to calculate dates using simple arithmetic
True, but the same property also makes the use of POSIX time for precision timing basically suicidal. POSIX time is a convenient and adequate encoding of civil time, as long as you do not need more than one second of accuracy.
If you want to reliably measure or timestamp anything with more than one second of accuracy you should be using a clock derived or offset from a reliable clock instead. The use of POSIX time for precision timekeeping - even use by such rudimentary applications as 'make' - was defective from the beginning. NTP is equally defective as a consequence.
By this point, even the luminaries at the IETF have realized that the design for IPv6 as a replacement for IPv4 is fatally flawed. How flawed? Flawed enough that there is a high probability that a worldwide transition to IPv6 will never actually happen.
Now sure, there are technical advantages to a clean slate design, but a clean slate design is also unfortunately almost useless as a replacement for IPv4 in the real world. There is no incremental advantage and extraordinarily high costs to adding a separate numbering plan to an existing network, so no cost conscious organization ever does it unless they are forced to, and probably never will.
At this point I would lay odds on an IPv7 eventually being developed that is a revision of IPv6 with the incorporation of the IPv4 address space in a routeable fashion, and which assigns each IPv4 address a network prefix that an entire subnet of devices may eventually be directly addressed behind, in addition to the default.
Why? Because doing anything else would be one of the biggest wastes of resources the world has ever seen.
Any downsides? An IPv7 router would have bigger routing tables than an IPv6 only router, but the routing tables could be used to route IPv4 packets, and as it is not likely IPv4 is going away anytime soon, the same overhead is there one way or another.
A wide scale deployment of IPv7 would require hardware upgrades in some cases, but for most people it could be deployed silently, without them ever needing to know or care. A simple software update would be all that was necessary, and a few years down the road nearly all IPv4 capable devices would handle the expanded address space in a usable fashion without any renumbering or other configuration changes. That would save billions of dollars a year in unnecessary administration costs worldwide.
There are only eight pages of new rules. The rest is explanation, history, legal justification, and commentary. More here: http://e-pluribusunum.com/2015...
since traffic from isp customers to those edge caches do not get counted against monthly caps
If that is true, the FCC is likely to frown upon that sort of thing.
The FCC doesn't have a problem with prioritization per se. The FCC has a problem with paid prioritization, which on a congested network can starve other traffic, and cause other problems, depending on how it is done.
Under any reasonable interpretation of the law, Internet access providers always have been common carriers. The FCC, in a classic example of regulatory capture, simply decided to interpret the law in a relatively perverse manner, by pretending that broadband Internet access providers were "information services" rather than "telecommunications services", which is flat out ridiculous, and the Supreme Court decided to defer to them.
This is the legal definition of telecommunications for example:
The term "telecommunications" means the transmission, between or among points specified by the user, of information of the user's choosing, without change in the form or content of the information as sent and received. (47 USC 153)
Sound familiar? Sounds just like Internet access. How about this one:
The term "telecommunications service" means the offering of telecommunications for a fee directly to the public, or to such classes of users as to be effectively available directly to the public, regardless of the facilities used.
Justice Scalia pointed this out several years ago, but he was in the minority on this one. The justices in the majority said, well it may not make any sense, but we will let the FCC decide. Now rationality has returned to the FCC and they are revisiting the question.
Federal law states that:
"A utility shall provide a cable television system or any telecommunications carrier with nondiscriminatory access to any pole, duct, conduit, or right-of-way owned or controlled by it." (47 USC 224)
Comcast has this right by virtue of being a "cable television system". The major phone companies have it because they are "telecommunications carriers". But facilities based ISPs like Google Fiber are currently (and incorrectly) classified as neither so they are out of luck until sanity finishes kicking in at the FCC.
XFS is not a copy on write filesystem, by the way. ZFS and BTRFS should work definitely work better, but they might need some internal tweaks to make the best use of it.
Somebody needs a math lesson. 3000 miles * 5280 feet per mile / 78000 = 203 feet. That is a tad more than 40 cm.
The Apple IIgs was dramatically different from all other Apple II models. It was backward compatible, but came with a 16 bit processor (the 65816), much more RAM (256K or more), greatly improved sound and video, and a GUI shell much like that of the Mac, plus color, which nearly all Macs lacked at the time. It was a little underpowered compared to the 68000 based Mac, Amiga, and Atari ST, but a more than respectable upgrade to the Apple II series nonetheless.
As educational / entertainment devices even the older Apple IIs ran circles around the PC until EGA was widely deployed in the late 1980s. PC games inevitably were designed for CGA graphics, with a fixed set of four unimaginative colors at a time. The Apple II was better than that almost ten years earlier, to say nothing of the much less expensive Commodore 64. The PC was intended primarily for business purposes, and it showed.
The Apple II wouldn't be more than a footnote in history if those prices didn't go way down, which they did. By the mid 1980s virtually every school in the country had a classroom full of them.
If you asked the creators, they would probably be embarrassed to call it an operating system at all. Apple DOS didn't handle keyboard support, video support, sound support, or printer support. That was all handled using either the monitor (a BIOS in ROM that was not part of DOS), peripheral card ROMs in some cases, or by direct access to the hardware.
MS-DOS was similar. It handled file I/O and that is it. A disk operating system, not a computer operating system. The BIOS was separate, not controlled by Microsoft, but rather by the PC manufacturer, and in ROM.
He didn't write an OS, he wrote a disk operating system, i.e. a system that operated disks. They called it DOS for a reason.
A stateful firewall doesn't need to block transport layer protocols it doesn't understand in order to provide a meaningful level of security. All it needs to do is block packets from IP addresses that corresponding interior address has not recently communicated with, with a reasonable time out. UDP is handled much the same way today.
If the developers of stateful IPv6 firewalls do not ship devices with such a reasonable configuration by default, they will block the deployment of new transport protocols indefinitely - at least all those that do not resort to the awkward expedient of running on top of UDP.
Blocking new transport protocols developers can reasonably handle with a standard policy is bad for efficiency, power consumption, latency, user experience, and so on in the long run - TCP is far from ideal as a transport protocol goes. In a number of ways it is outright backwards. If you want to impede the long term development of the Internet, degrading the end-to-end principle unnecessarily is a good place to start.
Work is underway for concurrent multipath transfer for SCTP as well. Also known as CMT-SCTP. There are significant challenges in doing this sort of thing though. SCTP wasn't designed for CMT, and probably needs much more radical changes than the current architects are proposing to do it well.
Changes like subflows with independent sequence numbers and congestion windows, to start with. SCTP is much further ahead in the connection handling and security department, but MPTCP has the odd advantage of resorting to independent subflows to begin with, and if it can handle path failure properly, it might well be ahead in the CMT game, if byte stream semantics are all you need.
On the contrary, SCTP is a transport protocol just like TCP, except with a large number of added features. The main problem with SCTP has nothing to do with SCTP at all. It is that NAT devices do not support any transport protocol that they haven't been programmed for in advance. This makes SCTP next to impossible to deploy on a broad scale - NAT, that wart upon router-kind, is ubiquitous.
TCP would have exactly the same problem if it were a new protocol. A NAT device requires relatively deep knowledge of TCP to support it at all. It play games with both ports and addresses, keeps track of connection state, and so on. Ordinary routers do no such thing. A NAT device is a transport layer proxy by another name.
The late 1950s would be more accurate. Computers in the late 1940s predated compilers. If you were extraordinarily lucky, you might have an assembler.
No, he said that the government breaking up the monopoly was "criminal"
I guess that is why one should quote a line or two of the comment he or she is responding to. The one you were actually responding to was hidden.
And when I say "simple engineering", I mean standard engineering practice. No one in his right mind designs a mission critical system for tens of thousands of people without full redundancy and usually automatic failover as well. A system that size that goes down for more than five minutes is more likely to be the result of engineering malpractice than anything else.
I still have a traditional land line, because it is reliable. I can't remember ever picking it the phone up and not having a dial tone. That would most definitely not be the case with any other kind of service available to residences and small businesses. The PSTN was designed to be reliable, and the people who designed readily available Internet access service designed it by the same standards that one might use for toys. Entertainment devices.
No one cares very much if they can't watch I Love Lucy for a couple of hours, although every television station I am aware of is about three orders of magnitude more reliable than the DSL / cable monopolies as well, if not quite as reliable as the PSTN.
That is quite the red herring there. Did the poster say we should restore the AT&T monopoly? No. What he said (in so many words) is that the systems were engineered to such high standards that they never went down.
The suggestion that we would have to return to monopoly pricing in order for Internet access providers to give reliable service is preposterous. It takes some simple engineering is all. Which the current folks at AT&T (formerly Southeastern Bell) are apparently either incapable of or are prevented from doing. The folks at Comcast and CenturyLink are similar.
If these quasi-monopoly providers can't deliver basic service that doesn't take large areas down for hours on end, they are inviting government regulation to make them engineer their systems so they don't. Not price regulation, engineering standards regulation.
Personally, I appreciate it when I cross a bridge and it doesn't collapse under the offered weight. That doesn't seem to be a particularly high priority at most of the monster ISPs these days - quite the opposite. Lets make the most unreliable service we can get away with seems to be the order of the day.
This is intended for server use. No video output required. The PCIe part is actually optional. I wouldn't expect to see this in workstations anytime soon, not without a major redesign at any rate. The form factor is designed for small, low power processors. The interface is not designed for SMP or anything like that either.
It is not a CPU slot specification. It is a micro-server slot specification, which is much more practical. Think small form factor blade server. The PCIe part is actually optional.
Basically, they are sawing the motherboard in 2, where the CPU and memory are on the daugterboard, and the rest of the components (SATA,USB3, PCIe slots, sound, video outputs) are all that remain on the motherboard
It is actually a micro-server architecture. Think small form-factor blade servers with an optional PCIe interconnect, optional remote SATA devices, and one mandatory ethernet interface, all running through what looks like an ordinary PCIe slot, but isn't.
http://www.opencompute.org/wp/wp-content/uploads/2013/01/Open_Compute_Project_Micro-Server_Card_Specification_v0.5.pdf
No. These are micro-servers we are talking about. Two RAM slots typical. Low power energy efficient CPUs too. ARM to start with.
Take a look:
http://www.opencompute.org/wp/wp-content/uploads/2013/01/Open_Compute_Project_Micro-Server_Card_Specification_v0.5.pdf
More important, it means that elements of memory and processing that now must be fixed closely together can be separated within a rack, and used as needed for different kinds of tasks.
This statement is in reference to Intel's proposal, which is still vaporware. I seriously doubt they are talking about locating main memory away from the processors. That would more or less be suicidal.
Facebook's design certainly does no such thing.
http://www.opencompute.org/wp/wp-content/uploads/2013/01/Open_Compute_Project_Micro-Server_Card_Specification_v0.5.pdf
Here is a link to an actual specification. If you read it, you will see that about half of what has been written about this announcement is wildly off base. We are talking micro-servers here - complete with on board cpu, ram, boot eeprom, flash storage, and ethernet. PCIe and SATA connections to the backplane are optional. Think small form factor blade server.
http://www.opencompute.org/wp/wp-content/uploads/2013/01/Open_Compute_Project_Micro-Server_Card_Specification_v0.5.pdf