The cubic with speed is only in a wind tunnel. If they car looks decent, it's back end won't be the best shape to have minimal drag which means you have a bubble of parasitic drag behind the car. Factor in cross winds and landscape related issues around the car and you will find that there are ranges where increases in speed mean on minor increases in drag. Throw in the efficiency curve of a typical high compression engine/transmission ratios and the compounding effects of how much extra fuel gets burned by the compound interest like effect of stopping at too many red lights and you may find that your cars most efficient speed may be faster than you think.
The problem is that the sats that are measuring the magnetic field are way too low to be very useful and the data seems to show they are missing massive changes for years. One example is the two N poles that are forming west of Perth. One of them showed up about the time rain stopped and the other had to have been there about the same amount of time based on related fields but showed up a few years later. I expect that what is going on at the peak of the magnetosphere has a significant effect on global weather patterns but most of our info on it comes from satellite that were on their way to out other jobs and you can't really park a satellite there.
Funny thing is that none of my production code base even runs under 2.6. I'm moving stuff from a very old server to new hardware and so far I've had to move 2.1,2.2,.2.3 and 2.4 over and some stuff broke when using the newest version of some of the old version. The result is now I have to spend lots of time maintaining programs that should not have to be maintained. I have never seen a project written in Python that meets its time or financial budget and stuff like this makes me want to ban the language from our systems completely. That also seems to be a tend with major open source projects that also never seem to get finished. There is a perpetual tweaking that must be done to keep things working and that is so wrong. Throw in the security issues and the maintenance costs of python code and there is no positive return on investment management point of view. Remember stability is good. Loading thousands of unauditable packages is bad.
Yet that plastic doesn't give you permission to drive on the roads as it is only the record of a past grant of permission. A bit in a database somewhere is what gets checked if anyone needs to know if you have a license. If the bit says no, the plastic is useless (except as an ID). So which is the real license? The database or the plastic?
PCI came form SETCo's specs (which I was involved with) and they still haven't caught up yet. The original idea is you need to run something like tripwire on your Unix systems and anti-virus stuff on your PCs and Macs (OS 10 didn't exist at that time). If your Unix systems are processing mail or file transfers and you have the ability to run anti-virus on the mail then you should. The idea was not to run Norton AV on your Unix OS.
PCI was set up as an attempt to get merchants that don't care about security to consider it with out it being too hard and forcing them away from card payments. They pretend to have no conflict of interest yet the only way to get certified is by proving you do have a conflict of interest. Many of the ideas for standards tends to be controlled by people who want to micromanage how things are done (which means you can use their product). Some of the ideas tend to get implemented poorly just in a way to be "PCI compliant" without the full consideration of the real security issues at hand.
The most interesting thing about the system is that the purpose of the auditors is to be the fall guys and take the liability. If a clearing house looses a bunch of card details, they won't be around long enough for anyone to sue which is why the auditors need so much insurance.
I've been setting up machines with no swap partitions for a few years. Swap partitions have a bad habit if collecting secure info you may have assumed was just in ram. All modern operating systems allow to you use a file or other blank space as swap means you don't need a dedicated partition. There is also the issues that if your starting to swap, where does it end? If your swapping on a machine with 4 or 8 gig of ram, will an extra gig help fix what ever is broken or just make the machine very slow until it gets around to telling the runaway program that there is no more memory. In the case of no swap, that tends to happen much faster. The only reason I see for swap partitions is that the OS will need a place to dump debug info if it crashes and the swap partition has traditionally been used for that.
The hardware based controllers (TRS and PC) had a sensor to look for the hole. That was only used for formatting because you just told the chip what sector you were trying to read and it collected it for you as the disk went around. The formatting could allow 218, 256 or 512 byte sectors (or even other sizes too) and there were a great deal of wasted space between the sectors. When I was playing with 80 track double sided disks, I found I could get massive capacities if I was willing to rewrite the entire track every time I wanted to write one sector by simply not putting as much junk in between each sector. Once you start writing a track at a time, you look for the hole and start writing and hope the hole doesn't show up before your done. The sector order didn't matter.
Even easier... You just tell it that its a 16 track disk and don't even format the rest. The PC uses the 1st track and the C64 uses 18 on. The PC uses 512 byte sectors while TRS-80 used 256 byte sectors. They TRS80 and IBM PC both used nearly the same controller chip and the Apple ][, V20/c64 did the decoding in software. There were disks that worked with both the COCO (also used track 18 for dir) and Microware's OS9.
Since I known 30+ people in the list I would have to say you are very wrong. While I don't know the author of Word, it would not surprise me if the 1st few versions weren't 90% the work of one person and I've used word on a unix based 3b2.
About the only thing that it doesn't do is the whole "This volume needs a primitive file system at this location on its idea of the disk" which nearly every system needs to boot.
The solution I'm looking for is the Solaris Kernel, FreeBSD system, optional OS X interface and something like apt-get that plays nice with ports. I hate when I have to install several packages to load one small command line tool and most of the times ports makes that far easier to deal with than any other packing system.
So you don't reallocate the/8 in pools/20 or/19 and you allocate them just as/28 or smaller. We are not running out of addresses, we are running out of routes and IPv6 doesn't fix that.
Getting rid of broadcast addresses and network addresses would go a long way in reducing IP usage. Its amazing how much brand new equipment can't deal with/30 and/31 addressing.
--and this will modded down by IPv6 fnaboys so please metamoderate
You logic assumes few people want to be dual homed which is a very bad assumption for any modern network. Once that happens you need router tables. The sparseness of the routing tables doesn't change how many entries there are unless your not really building a routing table but just want to build an association list. Once again IPv6 doesn't fix the real problem which is existing routers can't cope with the number of unique routes they need to hold. This is why places with a large number of existing IPv4 and IPv6 are ditching small entries for both. Try to route to a dual homed/24 in the US or Europe from Japan and you will see many of them are simply consolidated away even though they are directly connected to the same upstream providers.
A modern 4 interface router that sees the world as a 16 million/24 needs as much cache ram as a modern CPU tied into an FPGA and the whole problem goes away and everyone could have several upstream ISPs and it would all just work.
We are not running out of IPv4 addresses, but we are running out of IPv4 routes. IPv6 isn't going to fix that and in fact doubles the number of routes needed.
We are not out of IPv4 addresses. We are out of unique IPv4 routes. IPv6 doesn't help with that situation.
The proper solution (which I proposed back around '93) was allocate address only in/24 blocks and treat all address blocks as/24 which means every router needs to deal with 16.7 million routes. Using content addressable memory (as used in cache tag ram) and an FPG, routing could be done at wire speeds back then.
Does anyone have a list of the patents numbers? Some of the space based GPS stuff already has prior art and I expect all 6 of them can be broken before the auction.
US Government research belongs to the citizens of the US. If they want to license these patents internationally, great but they need to go to great lengths to ensure that US citizens don't pay twice.
Maybe the EFF should see if they can get this reconsidered.
Its not just the tube energy, but all the other bits in the lights that were intended to turn on about 250 times a year with a 10 to 20 year life span. If your start turning them on 10 times a day, you may have to replace them in less than a year.
Your description of the data structures needed seems to be covered in every OOD book... too bad the air traffic control system isn't concerned about planes but air space. I'll try to point out why this problem is much harder than it looks with some other info. The old system is based on allocating slots which most people don't seem to understand. Say your doing a low flight from STL to MKC (250 miles or so). There will be several slots allocated. You tell ATC when you expect to take off and the speed and the whole system allocates a take off slot, a departure slot, two en-route slots, an arrival area slot, rival slot and and landing slot. There there will be a slot from the landing airport to the alternate with whoever many slots it needs. Once a plan is accepted, each of those slots gets reserved. The system was based on the idea of what if several things go wrong. From the time a pilot takes off, the system assumes they will lose their radio or radar or several other things at any point. If nothing breaks, they can close the reservation early. If everything is working then the controller can adjust the slots so optimise things but things still won't break if things go wrong. In the case of route I've mentioned above, the slots will be from the 1st airport to the VOR on the west side of town. Then a slot on the air highway v12 will be allocated at a specific altitude to the VOR about 120 miles away. Years ago planes flew close to VORs but tended to be a mile or so to the side. Now with GPS based autopilots, the planes are often within a wingspan of being over the exact spot. There are also more issues with allocating climbing and descending slots so they don't interferer with other allocations. The backup plan for when the ATC computer dies is the controller has a number of paper tags that get moved around on slots to indicate where different aircraft are.
I figure less than 1 out 100 people have what it takes to be a very good programmer. The foundations for them will be different than the foundations needed for other students and the wrong ones can create a sort of brain damage that will take years to unlearn and I'm not sure some bad thought processes can ever be unlearned.
I would start the 1st week or two off with a very basic system of what the computer is doing.... i.e. moving numbers around. Go find a computer book from the 1950s for ideas on how to do that. Next build on the ideas that all software builds on complex layers of other software. Show them assembly code (but not x86 ick), basic, C, logo, and lisp for a start. Explain why they all have their uses and see how they react. Show them that the simplest problem involves many levels of depths and then explain how different groups each have their own area like networking, cpu design, programers, operators and maintenance coders.
For intro to programming, the IDE makes more of a difference than the language. Its one reason VB was used a while back since it had an easy to learn interface and they could write simple programs that did simple things and focus on that and not how to get something compiled. Most of todays IDEs are so full featured they can be hard to use for the level you are looking at. Just try not to pick a language that encourages programming brain damage (like Basic in all of its forms)
Slashdot Mirror
The cubic with speed is only in a wind tunnel. If they car looks decent, it's back end won't be the best shape to have minimal drag which means you have a bubble of parasitic drag behind the car. Factor in cross winds and landscape related issues around the car and you will find that there are ranges where increases in speed mean on minor increases in drag. Throw in the efficiency curve of a typical high compression engine/transmission ratios and the compounding effects of how much extra fuel gets burned by the compound interest like effect of stopping at too many red lights and you may find that your cars most efficient speed may be faster than you think.
The problem is that the sats that are measuring the magnetic field are way too low to be very useful and the data seems to show they are missing massive changes for years. One example is the two N poles that are forming west of Perth. One of them showed up about the time rain stopped and the other had to have been there about the same amount of time based on related fields but showed up a few years later. I expect that what is going on at the peak of the magnetosphere has a significant effect on global weather patterns but most of our info on it comes from satellite that were on their way to out other jobs and you can't really park a satellite there.
Funny thing is that none of my production code base even runs under 2.6. I'm moving stuff from a very old server to new hardware and so far I've had to move 2.1,2.2,.2.3 and 2.4 over and some stuff broke when using the newest version of some of the old version. The result is now I have to spend lots of time maintaining programs that should not have to be maintained. I have never seen a project written in Python that meets its time or financial budget and stuff like this makes me want to ban the language from our systems completely. That also seems to be a tend with major open source projects that also never seem to get finished. There is a perpetual tweaking that must be done to keep things working and that is so wrong. Throw in the security issues and the maintenance costs of python code and there is no positive return on investment management point of view. Remember stability is good. Loading thousands of unauditable packages is bad.
Yet that plastic doesn't give you permission to drive on the roads as it is only the record of a past grant of permission. A bit in a database somewhere is what gets checked if anyone needs to know if you have a license. If the bit says no, the plastic is useless (except as an ID). So which is the real license? The database or the plastic?
PCI came form SETCo's specs (which I was involved with) and they still haven't caught up yet. The original idea is you need to run something like tripwire on your Unix systems and anti-virus stuff on your PCs and Macs (OS 10 didn't exist at that time). If your Unix systems are processing mail or file transfers and you have the ability to run anti-virus on the mail then you should. The idea was not to run Norton AV on your Unix OS.
The license is the governments permission and the certificate is the bit of paper saying so.
PCI was set up as an attempt to get merchants that don't care about security to consider it with out it being too hard and forcing them away from card payments. They pretend to have no conflict of interest yet the only way to get certified is by proving you do have a conflict of interest. Many of the ideas for standards tends to be controlled by people who want to micromanage how things are done (which means you can use their product). Some of the ideas tend to get implemented poorly just in a way to be "PCI compliant" without the full consideration of the real security issues at hand.
The most interesting thing about the system is that the purpose of the auditors is to be the fall guys and take the liability. If a clearing house looses a bunch of card details, they won't be around long enough for anyone to sue which is why the auditors need so much insurance.
I've been setting up machines with no swap partitions for a few years. Swap partitions have a bad habit if collecting secure info you may have assumed was just in ram. All modern operating systems allow to you use a file or other blank space as swap means you don't need a dedicated partition. There is also the issues that if your starting to swap, where does it end? If your swapping on a machine with 4 or 8 gig of ram, will an extra gig help fix what ever is broken or just make the machine very slow until it gets around to telling the runaway program that there is no more memory. In the case of no swap, that tends to happen much faster. The only reason I see for swap partitions is that the OS will need a place to dump debug info if it crashes and the swap partition has traditionally been used for that.
I don't remember doom running on the punch card machine.
The hardware based controllers (TRS and PC) had a sensor to look for the hole. That was only used for formatting because you just told the chip what sector you were trying to read and it collected it for you as the disk went around. The formatting could allow 218, 256 or 512 byte sectors (or even other sizes too) and there were a great deal of wasted space between the sectors. When I was playing with 80 track double sided disks, I found I could get massive capacities if I was willing to rewrite the entire track every time I wanted to write one sector by simply not putting as much junk in between each sector. Once you start writing a track at a time, you look for the hole and start writing and hope the hole doesn't show up before your done. The sector order didn't matter.
The step size on the 35 and 40 were the same. The track size on the 80 was 1/2 what it was on the 40. Track 0 on all disks was at the same location.
Even easier...
You just tell it that its a 16 track disk and don't even format the rest. The PC uses the 1st track and the C64 uses 18 on.
The PC uses 512 byte sectors while TRS-80 used 256 byte sectors. They TRS80 and IBM PC both used nearly the same controller chip and the
Apple ][, V20/c64 did the decoding in software.
There were disks that worked with both the COCO (also used track 18 for dir) and Microware's OS9.
Since I known 30+ people in the list I would have to say you are very wrong. While I don't know the author of Word, it would not surprise me if the 1st few versions weren't 90% the work of one person and I've used word on a unix based 3b2.
About the only thing that it doesn't do is the whole "This volume needs a primitive file system at this location on its idea of the disk" which nearly every system needs to boot.
The solution I'm looking for is the Solaris Kernel, FreeBSD system, optional OS X interface and something like apt-get that plays nice with ports. I hate when I have to install several packages to load one small command line tool and most of the times ports makes that far easier to deal with than any other packing system.
Remember for short lived variables its i,j for integers and x,y,z for floating point unless your doing 3d work and then its a,b, and c.
So you don't reallocate the /8 in pools /20 or /19 and you allocate them just as /28 or smaller.
We are not running out of addresses, we are running out of routes and IPv6 doesn't fix that.
Getting rid of broadcast addresses and network addresses would go a long way in reducing IP usage. /30 and /31 addressing.
Its amazing how much brand new equipment can't deal with
--and this will modded down by IPv6 fnaboys so please metamoderate
You logic assumes few people want to be dual homed which is a very bad assumption for any modern network. Once that happens you need router tables. /24 in the US or Europe from Japan and you will see many of them are simply consolidated away even though they are directly connected to the same upstream providers.
The sparseness of the routing tables doesn't change how many entries there are unless your not really building a routing table but just want to build an association list. Once again IPv6 doesn't fix the real problem which is existing routers can't cope with the number of unique routes they need to hold. This is why places with a large number of existing IPv4 and IPv6 are ditching small entries for both. Try to route to a dual homed
A modern 4 interface router that sees the world as a 16 million /24 needs as much cache ram as a modern CPU tied into an FPGA and the whole problem goes away and everyone could have several upstream ISPs and it would all just work.
We are not running out of IPv4 addresses, but we are running out of IPv4 routes. IPv6 isn't going to fix that and in fact doubles the number of routes needed.
We are not out of IPv4 addresses. We are out of unique IPv4 routes. IPv6 doesn't help with that situation.
The proper solution (which I proposed back around '93) was allocate address only in /24 blocks and treat all address blocks as /24 which means every router needs to deal with 16.7 million routes. Using content addressable memory (as used in cache tag ram) and an FPG, routing could be done at wire speeds back then.
Does anyone have a list of the patents numbers?
Some of the space based GPS stuff already has prior art and I expect all 6 of them can be broken before the auction.
US Government research belongs to the citizens of the US. If they want to license these patents internationally, great but they need to go to great lengths to ensure that US citizens don't pay twice.
Maybe the EFF should see if they can get this reconsidered.
Its not just the tube energy, but all the other bits in the lights that were intended to turn on about 250 times a year with a 10 to 20 year life span. If your start turning them on 10 times a day, you may have to replace them in less than a year.
Your description of the data structures needed seems to be covered in every OOD book... too bad the air traffic control system isn't concerned about planes but air space.
I'll try to point out why this problem is much harder than it looks with some other info.
The old system is based on allocating slots which most people don't seem to understand. Say your doing a low flight from STL to MKC (250 miles or so). There will be several slots allocated. You tell ATC when you expect to take off and the speed and the whole system allocates a take off slot, a departure slot, two en-route slots, an arrival area slot, rival slot and and landing slot. There there will be a slot from the landing airport to the alternate with whoever many slots it needs. Once a plan is accepted, each of those slots gets reserved. The system was based on the idea of what if several things go wrong. From the time a pilot takes off, the system assumes they will lose their radio or radar or several other things at any point. If nothing breaks, they can close the reservation early. If everything is working then the controller can adjust the slots so optimise things but things still won't break if things go wrong. In the case of route I've mentioned above, the slots will be from the 1st airport to the VOR on the west side of town. Then a slot on the air highway v12 will be allocated at a specific altitude to the VOR about 120 miles away. Years ago planes flew close to VORs but tended to be a mile or so to the side. Now with GPS based autopilots, the planes are often within a wingspan of being over the exact spot. There are also more issues with allocating climbing and descending slots so they don't interferer with other allocations. The backup plan for when the ATC computer dies is the controller has a number of paper tags that get moved around on slots to indicate where different aircraft are.
Its not about pushing socialism, its about introducing new trade tariffs.
I figure less than 1 out 100 people have what it takes to be a very good programmer. The foundations for them will be different than the foundations needed for other students and the wrong ones can create a sort of brain damage that will take years to unlearn and I'm not sure some bad thought processes can ever be unlearned.
I would start the 1st week or two off with a very basic system of what the computer is doing.... i.e. moving numbers around. Go find a computer book from the 1950s for ideas on how to do that. Next build on the ideas that all software builds on complex layers of other software. Show them assembly code (but not x86 ick), basic, C, logo, and lisp for a start. Explain why they all have their uses and see how they react. Show them that the simplest problem involves many levels of depths and then explain how different groups each have their own area like networking, cpu design, programers, operators and maintenance coders.
For intro to programming, the IDE makes more of a difference than the language. Its one reason VB was used a while back since it had an easy to learn interface and they could write simple programs that did simple things and focus on that and not how to get something compiled. Most of todays IDEs are so full featured they can be hard to use for the level you are looking at. Just try not to pick a language that encourages programming brain damage (like Basic in all of its forms)