The requirements were that the sites had legacy T1's and similar and were being upgraded to fiber. Therefore the router had to have both legacy interfaces and high performance. That combination is awfully expensive and the 3945 is not an unreasonable choice.
It would have been much cheaper If the requirements had allowed for temporarily having two routers on the sites until the legacy T1's were taken down or alternatively allowed for an extra visit to the site to replace the router.
Trying to avoid an extra trip to each site is not stupid. Requiring both legacy and high speed interfaces is not stupid. Going for a unified platform is not stupid. However, a joint meeting with the pre-qualified bidders would likely have revealed the potential cost savings of making a compromise on the requirements. Alternatively, an independent consultant with just a little experience in the area should have spotted it.
The same thing happens in many of bids, not just in the IT sector. Seemingly reasonable requirements together mean that only very few vendors can bid and that they need their most expensive solutions to handle it.
If you read the article, this is a buffer overflow in the VM itself, overwriting internal VM structures. In previous cases you'd be correct, but this is an actual JVM flaw.
It is likely that there are similar vulnerabilities in other VMs. People generally do not worry about them, because they are not made for untrusted code. You can crash the Python VM with python -c "from ctypes import string_at; string_at(0xDEADBEEF)". That is fine, because Python does not have sandboxing.
To get the traffic to the box each ISP just listens to a BGP announcement from the Cleanfeed "ISP". Then all of the ISP routers automatically do the right (wrong) thing, no further configuration needed. To get the return traffic from the content provider, the ISP needs to do source-based routing (policy routing). This means adding configuration to at least each edge router which might receive traffic from one of those banned content providers, and in many cases the core routers need to be configured for this as well. Maintaining this configuration on all the routers is a nightmare and source-based routing is generally difficult to get right and to troubleshoot when it goes wrong.
You are not going to get ISPs to do source-based routing just by putting a bit of pressure on them. That will only happen if there is no legal way to avoid it, and there will be massive lobbying from the industry against such a law.
Back when I worked with transparent proxies in the 90's, keeping the source address intact was a fairly standard feature...
Anyway, their setup depends on the source address changing. Otherwise return routing would miss the inspection-box. So even if their transparent proxy/DPI box has the "keep source address intact" option, they cannot enable it.
Oh, if only they had read your comment before sending their lawyers to the Supreme Court. It is unfortunate that they picked lawyers who didn't know anything about proper protocol. Victory would have been assured if they had picked a couple of Slashdotters at random instead.
He says exactly the opposite, TFS is just wrong as usual.
It is the SCANNER which supposedly removes the need for enhanced pat-downs, so supposedly everyone would have to go through those without the scanner. Good luck with that.
Most likely the evil twin is just routing back to your wi-fi network.
That wouldn't really work. The point of the evil twin network is to steal your WPA key / login. It would be difficult to pull off the attack quickly enough to actually be able to route traffic back to your wi-fi network at the same instant.
The Evil Twin network likely doesn't have Internet access. Even if it does, it is probably using one of the other nearby wifi networks for connectivity.
Actually it has pretty much everywhere to radiate the heat. It is the non-radiative heat transfer which is missing in space.
Since we are talking Western-style revolvers, I would guess that it is unlikely that they will fire enough bullets to make heat a significant problem. How quickly can you reload in space?
Admittedly I have never fired a handgun or travelled in space.
The solar panels and wind turbines require materials that are not available on a scale that would allow those sources of energy to ever meet our current needs, let alone future needs.
This is simply not true. I have no idea where you are getting it from.
A wind turbine is simply a bunch of fibre glass, a gear, and a generator. Fibre glass is abundant, gears just require rather commonly available metals, and the generator is often a standard electromagnetic generator. You can win a few percent extra power and possibly save on the gear by going to permanent magnets, but 5% at the margin isn't going to determine whether we can meet the energy needs of the world -- and the "rare earths" needed for permanent magnets are not actually very rare.
Solar cells are made from a myriad of materials. Some of those will scale almost unlimited, some likely won't.
However, your views luckily do not matter. Wind power is now cost competitive in South America without subsidies, even if fossil fuels do not have to pay for any of the damage they cause. In just a few years that will be true in much larger parts of the world. Simple economics will kill off coal-fired power plants.
Water gets broken into oxygen and hydrogen all the time and combined with carbon and other atoms for all sorts of interesting molecules. The fact that water itself cannot escape does not necessarily mean that it cannot disappear in other ways.
I would guess the same as you though, that the amount of water is increasing, incredibly slowly.
That is actually the only benefit of corn ethanol. It ensures an oversupply of corn compared to consumption for food in average years, and in drought years you can just stop producing ethanol.
It would admittedly be cheaper and less wasteful to mandate that the government buys 20% of the corn production in non-drought years and buries it in the ground...
Since WW2 Brazil has been using home grown ethanol as a fuel because they either couldn't get oil (I'm told this is what diesel is made from) or didn't want to pay high prices for it.
Brazil AFAIK made ethanol from sugar cane. Sugar cane is an excellent choice for ethanol production; it is one of the most efficient plants when it comes to photosynthesis and it produces lots of sugar which is easy to turn into ethanol. Ethanol from sugar cane should have no problem producing more energy than is consumed.
Corn is just fairly crap all around when it comes to ethanol production.
If you set overcommit aggressively enough or use a sufficiently old Linux kernel, you will be able to malloc() 3GB on any 32 bit system. Assuming you stay with the default 3/1GB memory split.
Any testing of overcommit done on 32 bit systems is a bit useless. 32 bit systems are pretty much embedded-only.
As an experiment, I wrote a little progrem that malloc'ed 200MB chunks of memory. I ran this on a Linux box with 2GB of RAM and all the SWAP disabled. The program could malloc 3GB of RAM before the allocation requests failed.
You were running on 32 bit? You will hit the same limit whether you have 256MB or 16GB then.
Yes, UEFI has lots of interesting features. Carefully coded by the BIOS programmers we know and love so much. The ones who failed even the Visual Basic course and so had to settle for BIOS programming.
It is the terminal emulator part of the console which is going to user space (optionally). The console itself will not go away, so your serial port console is safe.
The built-in terminal emulator is pretty good for a 90's piece of software, but it is showing its age and the limitations are difficult to fix in kernel space.
For some reason people think that pharmacological patents make more sense than other patents. In reality, it is probably the sector which sees most harm from patents. A software patent rarely kills people, while pharmacological patents do -- and they do not save even remotely as many people as they kill.
Slashdot Mirror
The requirements were that the sites had legacy T1's and similar and were being upgraded to fiber. Therefore the router had to have both legacy interfaces and high performance. That combination is awfully expensive and the 3945 is not an unreasonable choice.
It would have been much cheaper If the requirements had allowed for temporarily having two routers on the sites until the legacy T1's were taken down or alternatively allowed for an extra visit to the site to replace the router.
Trying to avoid an extra trip to each site is not stupid. Requiring both legacy and high speed interfaces is not stupid. Going for a unified platform is not stupid. However, a joint meeting with the pre-qualified bidders would likely have revealed the potential cost savings of making a compromise on the requirements. Alternatively, an independent consultant with just a little experience in the area should have spotted it.
The same thing happens in many of bids, not just in the IT sector. Seemingly reasonable requirements together mean that only very few vendors can bid and that they need their most expensive solutions to handle it.
If you read the article, this is a buffer overflow in the VM itself, overwriting internal VM structures. In previous cases you'd be correct, but this is an actual JVM flaw.
It is likely that there are similar vulnerabilities in other VMs. People generally do not worry about them, because they are not made for untrusted code. You can crash the Python VM with python -c "from ctypes import string_at; string_at(0xDEADBEEF)". That is fine, because Python does not have sandboxing.
You can, but it gets extremely tricky.
To get the traffic to the box each ISP just listens to a BGP announcement from the Cleanfeed "ISP". Then all of the ISP routers automatically do the right (wrong) thing, no further configuration needed. To get the return traffic from the content provider, the ISP needs to do source-based routing (policy routing). This means adding configuration to at least each edge router which might receive traffic from one of those banned content providers, and in many cases the core routers need to be configured for this as well. Maintaining this configuration on all the routers is a nightmare and source-based routing is generally difficult to get right and to troubleshoot when it goes wrong.
You are not going to get ISPs to do source-based routing just by putting a bit of pressure on them. That will only happen if there is no legal way to avoid it, and there will be massive lobbying from the industry against such a law.
Back when I worked with transparent proxies in the 90's, keeping the source address intact was a fairly standard feature...
Anyway, their setup depends on the source address changing. Otherwise return routing would miss the inspection-box. So even if their transparent proxy/DPI box has the "keep source address intact" option, they cannot enable it.
"Cleanfeed". Built for blocking child porn, of course. Traffic to specific IP addresses is redirected to a deep inspection system.
I am afraid you got the last line wrong.
Gov: We spy on Americans in secret.
Me: Stop spying on me
Gov: You can't prove that we did
Gov: *middle finger*
Oh, if only they had read your comment before sending their lawyers to the Supreme Court. It is unfortunate that they picked lawyers who didn't know anything about proper protocol. Victory would have been assured if they had picked a couple of Slashdotters at random instead.
He says exactly the opposite, TFS is just wrong as usual.
It is the SCANNER which supposedly removes the need for enhanced pat-downs, so supposedly everyone would have to go through those without the scanner. Good luck with that.
Most likely the evil twin is just routing back to your wi-fi network.
That wouldn't really work. The point of the evil twin network is to steal your WPA key / login. It would be difficult to pull off the attack quickly enough to actually be able to route traffic back to your wi-fi network at the same instant.
The Evil Twin network likely doesn't have Internet access. Even if it does, it is probably using one of the other nearby wifi networks for connectivity.
Actually it has pretty much everywhere to radiate the heat. It is the non-radiative heat transfer which is missing in space.
Since we are talking Western-style revolvers, I would guess that it is unlikely that they will fire enough bullets to make heat a significant problem. How quickly can you reload in space?
Admittedly I have never fired a handgun or travelled in space.
The solar panels and wind turbines require materials that are not available on a scale that would allow those sources of energy to ever meet our current needs, let alone future needs.
This is simply not true. I have no idea where you are getting it from.
A wind turbine is simply a bunch of fibre glass, a gear, and a generator. Fibre glass is abundant, gears just require rather commonly available metals, and the generator is often a standard electromagnetic generator. You can win a few percent extra power and possibly save on the gear by going to permanent magnets, but 5% at the margin isn't going to determine whether we can meet the energy needs of the world -- and the "rare earths" needed for permanent magnets are not actually very rare.
Solar cells are made from a myriad of materials. Some of those will scale almost unlimited, some likely won't.
However, your views luckily do not matter. Wind power is now cost competitive in South America without subsidies, even if fossil fuels do not have to pay for any of the damage they cause. In just a few years that will be true in much larger parts of the world. Simple economics will kill off coal-fired power plants.
Water gets broken into oxygen and hydrogen all the time and combined with carbon and other atoms for all sorts of interesting molecules. The fact that water itself cannot escape does not necessarily mean that it cannot disappear in other ways.
I would guess the same as you though, that the amount of water is increasing, incredibly slowly.
So fix the bloody tools showing file sizes to show them correctly.
alias ls="ls --si"; alias du="du --si".
Sugar from beets is completely economically unviable. The only reason it is still grown is stupid protectionism.
That is actually the only benefit of corn ethanol. It ensures an oversupply of corn compared to consumption for food in average years, and in drought years you can just stop producing ethanol.
It would admittedly be cheaper and less wasteful to mandate that the government buys 20% of the corn production in non-drought years and buries it in the ground...
Since WW2 Brazil has been using home grown ethanol as a fuel because they either couldn't get oil (I'm told this is what diesel is made from) or didn't want to pay high prices for it.
Brazil AFAIK made ethanol from sugar cane. Sugar cane is an excellent choice for ethanol production; it is one of the most efficient plants when it comes to photosynthesis and it produces lots of sugar which is easy to turn into ethanol. Ethanol from sugar cane should have no problem producing more energy than is consumed.
Corn is just fairly crap all around when it comes to ethanol production.
I am sure there is a BIOS writer somewhere who is a proficient assembly guru. I have never had the chance to use any system he wrote code for.
If you set overcommit aggressively enough or use a sufficiently old Linux kernel, you will be able to malloc() 3GB on any 32 bit system. Assuming you stay with the default 3/1GB memory split.
Any testing of overcommit done on 32 bit systems is a bit useless. 32 bit systems are pretty much embedded-only.
As an experiment, I wrote a little progrem that malloc'ed 200MB chunks of memory. I ran this on a Linux box with 2GB of RAM and all the SWAP disabled. The program could malloc 3GB of RAM before the allocation requests failed.
You were running on 32 bit? You will hit the same limit whether you have 256MB or 16GB then.
Yes, UEFI has lots of interesting features. Carefully coded by the BIOS programmers we know and love so much. The ones who failed even the Visual Basic course and so had to settle for BIOS programming.
Laptop? They have PCMCIA to serial port adapters.
Yes, and floppy disk drives. Laptops don't even have Mini-PCI-Express these days, and they certainly don't have PCMCIA.
It is the terminal emulator part of the console which is going to user space (optionally). The console itself will not go away, so your serial port console is safe.
The built-in terminal emulator is pretty good for a 90's piece of software, but it is showing its age and the limitations are difficult to fix in kernel space.
For some reason people think that pharmacological patents make more sense than other patents. In reality, it is probably the sector which sees most harm from patents. A software patent rarely kills people, while pharmacological patents do -- and they do not save even remotely as many people as they kill.
However, does the data 1101 hold up to a copyright claim in court?
Of course it does. If it is a part of a copyrighted file, it absolutely holds up. See What colour are your bits?.
You can try to get away with calling it fair use. Good luck.
(Yes, it's absolutely ridiculous that bits have colour. I am not saying that the law is moral or even sane.)