I have set up my home wireless so that it only has access to the Internet and other machines on the wireless network. Machines that need to be secure are on a separate wired network.
If I want to access the protected network from a "wireless" machine, I do this though a VPN.
Through the magic of stateful firewall (Netfilter/iptables), the protected machines can access the wireless machines.
Just to be clear about the parameters of this test... I assume the PC from which you sent the request isn't on the same local network as the DNS server?
The machine from which I sent the request is connected to a Comcast residential Cable Internet connection. The server at the other end is a virtual machine in a colo facility somewhere -- not a Comcast facility. And before anyone asks, I tried both tcp and udp requests with the same result (no interception, no transparent proxy).
Are you certain? If they are redirecting the traffic in their network so that one of their DNS servers responds to the query as if it was your DNS server
I'm certain. I sent a query to a DNS server that I control. I ran tcpdump on the DNS server and I could see the packets from my home IP address coming in with the query and the refusal going out (I asked the DNS server that I control to resolve yahoo.com, which it should refuse to do).
Last time I had some spare time in an airport, I found that the T-Mobile hotspot allowed 53/UDP traffic out, so I was thinking of setting up openvpn on port 53 (instead of its usual 1194) in order to access my home machines (without a T-Mobile login). If Comcast intercepts this traffic, my evil plan won't work!
The party who has the burden of proof of proving ownership of copyrights is saying it. Hmmmm. I wonder why. If it was so easy for them to prove, and beyond dispute, why make an issue out of it?
While I have suggested elsewhere that it might be some problem other than outright ownership issues, I can think of a couple of other possible reasons for this motion:
The copyright registration documents are merely the paperwork that indicates that the copyright owners registered their works... which is a necessary precondition to suing for damages. The documents themselves also establish a prima facie presumption that the plaintiffs actually own the copyrights to the works that are allegedly infringed. In other words: Even if this motion to suppress objections fails, the defense is going to have to prove that the plaintiffs do not have rights to the copyrights in question... good luck trying to prove that.
It may not be so black and white. IIRC, if copyrights are not registered within a certain time period, one can only sue for actual damages and not statutory damages. This would make a huge difference to the defendent, since actual damages would be about $10.
And of course, since it's negative towards Microsoft, Slashdot dupes it a few dozen times.
/. dupes stories all the time, why would you assume that this was a malicious act against MS?
Oh, and it was an honest mistake in the first place, not some horrible malicious act.
Microsoft makes deliberate anti-open source moves all the time and is a convicted monopolist, why would you assume that is was a mistake? Or are you a MS employee with inside information?
They could also be keying off the scent of the method of labeling the disks.
You are over-thinking the problem. It's much simpler -- they search for outgoing containers/boxes/whatever that contain DVDs. When they find them, they check the manifest for that shipment. If the DVDs are not listed on the manifest -- they are probably illegally copied DVDs.
Offering different options when plugging into USB is sheer genius.
It's hardly novel -- my Nokia phone has done this for the last 2 or more years and I have no reason to believe other phones haven't been doing it for far longer. The only difference is that my phone doesn't have an "iPod mode".
It's illegal to present it as evidence in a court of law unless you have a Private Detective license.
The motion proposes that MediaSentry's actions are illegal under federal law, irrespective of whether the resulting logs are used in court or not:
C. Federal Electronic Communications Statutes
MediaSentry's activities also constitute criminal violations of two federal
statutes: (1) the Pen Register and Trap and Trace Devices Act, as amended by the USA
PATRIOT Act (the "Pen Register Act"); and (2) the Electronic Communications Privacy Act
of 1986 (the "Wiretap Act").
If I "discover" a new way of processing rubber, that method may have existed before in the abstract universal sense that it was possible to do it. The only thing holding people back was the knowledge.
Isn't that an invention, not a discovery? From Webster:
To Discover, Invent. We discover what existed
before, but remained unknown; we invent by forming
combinations which are either entirely new, or which
attain their end by means unknown before. Columbus
discovered America; Newton discovered the law of
gravitation; Whitney invented the cotton gin; Galileo
invented the telescope.
Perhaps what is really needed is a change to the law which removes the word "discover" and leaves only inventions as patentable.
Not to mention, it would require them to blatantly disregard the patent statute. 35 U.S.C. section 101, "Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter... may obtain a patent therefor[.]"
Doesn't "discovery" require that the subject of the discovery exist already? So, how does one discover something new? For example, when the "New World" was discovered, it's existence had been known for many years by the population who lived there, just not known by Europeans.
There were very good reasons why people used VHDL in the past. Because VHDL was an open language before Verilog, the cost of VHDL tools was historically lower than Verilog tools. Since this cost was much more important to FPGA designers, VHDL tended to dominate the FPGA market.
On ASIC side, the first mainstream commercial synthesis tool was Synopsys and Synopsys chose to support Verilog before supporting VHDL. Amongst all the other NRE costs in designing an ASIC, the added cost of using Verilog tools (instead of VHDL) was not really significant. Also, tools to support large designs advanced initially as Verilog tools.
Fast forward few years and Verilog is now open, the cost differential has now disappeared. However, VHDL had a lot of features related to design validation that were not in Verilog. In VHDL you can read and write files. Such things as configurations are supported, etc.. This type of capability makes it easier to write a testbench in VHDL, while on the Verilog side, additional tools and languages are commonly used.
Fast forwards a few more years to today and now we have System Verilog. This gives Verilog the capabilities that it lacked in comparison to VHDL and probably more. The price of VHDL tools is the same as Verilog tools.
Summary: it's clear that the future does not belong to VHDL. It looks like System Verilog is the future, although there are other contenders. So, if the choice is between VHDL and Verilog -- pick Verilog.
This statement is incorrect. Suppose you want to atomically replace the contents of file "foo". Your application will write a file "foo.tmp", then call rename("foo.tmp", "foo").
And what happens if the process has write permissions for the file, but not the directory?
Why does everyone keep speaking about EXT4 as if it's broken? It's working exactly as designed.
But is the design any good? If the advantage of EXT4 is better performance, how much of that performance improvement will be lost once the applications are fixed?
The money has to come from somewhere. The free news business model has been tried, and kudos to the newspapers for giving it their best shot, but it simply does not work. Screw em? No. Let's not 'screw em.' We need someone to uncover the next Watergate. We need someone to keep an eye on the war profiteers who charge $20 per washing machine load of laundry. We need someone to keep tabs on the polluters and bring it to the public's attention. If it means an exemption to anti-trust laws (that were written before newspapers ended up in this situation) then so be it. A professional news media is too important to be left to die.
Totall agree, we do need this. What we don't need is the rest of the baggage that newspapers want to keep selling to us. We don't need hundreds of newpaper titles, all carrying the same national news -- there is no added value any more in reprinting someone else's reporting.
Unfortunately, we don't currently have the newpapers that we need. The financial crash and the investigation of Madoff, where was that? The same person who gave details to the SEC about why Madoff must be running a Ponzi scheme took his information to a couple of major newspapers -- what did they do with it? Nothing. I did hear an interview of a reporter who had been investigating events that led up to the crash 2-3 years ago -- good reporting, yes? She wrote for UK newspaper.
The problem for the newspaper industry is the same as the buggy whip manufacturers when cars came along -- the demand for their product has dropped dramatically and many of the manufacturers will have to go out of business. Think of it this way, with hundreds of papers, the same function (collating and printing news) is repeated hundres of times each day, with no real added value over and above what the next guy is doing. There is no economic value in this. They had their shot, they made money (lots of money), but their time is now past. Technology and society have moved on.
Light gathering for a digital camera is slightly more complicated. Total light is the speed of the lens but each pixel also has to gather light and the bigger the individual pixel the more light it can gather.
Err, no. The amount of light entering is determined by te size of the lens. That light is spread across the sensor, so, with the same size lens and the same nmber of pixels, the amount of light per pixel (also assuming a fixed amount of time) is the same.
If you don't agre with the above statement, instead of just making bald assertions (as you have done in your posts above), explain why a larger pixel gathers more light if the same amount of light is spread across a larger area.
More pixels in a given area means smaller pixels which means more heat which means more noise in your photo.
Do the sensors really heat up significantly? The sensors don't run at high speeds, so it seems unlikely that they would heat up significantly. Do you have a citation to support your claim?
Bigger pixels can also get more light into each pixel for better low-light performance.
The amount of light entering the camera is determined by the size of the lens. The size of the sensor does not affect this. With a fixed amount of light and a fixed number of pixels, the amount of light per pixel is constant. So, how does a larger sensor collect more light?
Generally, the bigger you can make your signal, the weaker [relatively] the noise will be. The fact of the matter is that small sensors are more noisy, whether you agree or even understand.
Would not the same amount of light falling on a smaller sensor give a larger signal? Instead of just telling me I'm wrong and suggesting that I don't understand, please explain.
If you are telling me that a smaller lens is going to result in less light and hence a smaller signal, that's quite another thing, but I was very specific when I posed my original question -- which was related to the size of the sensor alone and assumed that the lens size remained unchanged.
The pixels are very small; electromagnetic interference within the chip itself becomes a noticeable factor in the total noise of the sensor.
I'm not sure that this is valid. Why should a smaller sensor be more suceptable to EM noise? Even if it were, the smaller size would mean that the distribution of noise between pixels would be smaller, allowing for better correction during the processing of the picture.
From a production standpoint, smaller sensors means more sensors per wafer with a higher failure rate.
Where do you get this from? Why should failure rates be higher? From a production standpoint, yields are generally higher for smaller die sizes.
If each individual pixel is actually larger, some ambient noise (stray high-energy EM, etc.) can be smoothed out or reduced simply because the pixel is too large to be activated by it.
Why? A larger pixel will be like a larger antenna -- larger currents will be induced in the antenna/pixel. The same total amount of light will shine on the pixel if the lens size is unchanged.
Slashdot Mirror
I have set up my home wireless so that it only has access to the Internet and other machines on the wireless network. Machines that need to be secure are on a separate wired network.
If I want to access the protected network from a "wireless" machine, I do this though a VPN.
Through the magic of stateful firewall (Netfilter/iptables), the protected machines can access the wireless machines.
Is there a motion before the court on this point, or time to put such a motion forward before the trial?
The machine from which I sent the request is connected to a Comcast residential Cable Internet connection. The server at the other end is a virtual machine in a colo facility somewhere -- not a Comcast facility. And before anyone asks, I tried both tcp and udp requests with the same result (no interception, no transparent proxy).
I'm certain. I sent a query to a DNS server that I control. I ran tcpdump on the DNS server and I could see the packets from my home IP address coming in with the query and the refusal going out (I asked the DNS server that I control to resolve yahoo.com, which it should refuse to do).
Last time I had some spare time in an airport, I found that the T-Mobile hotspot allowed 53/UDP traffic out, so I was thinking of setting up openvpn on port 53 (instead of its usual 1194) in order to access my home machines (without a T-Mobile login). If Comcast intercepts this traffic, my evil plan won't work!
Me too. I'm also in CA and it is not curently happening.
While I have suggested elsewhere that it might be some problem other than outright ownership issues, I can think of a couple of other possible reasons for this motion:
To distract attention from other issues?
To increase the defendant's legal costs?
It may not be so black and white. IIRC, if copyrights are not registered within a certain time period, one can only sue for actual damages and not statutory damages. This would make a huge difference to the defendent, since actual damages would be about $10.
Microsoft makes deliberate anti-open source moves all the time and is a convicted monopolist, why would you assume that is was a mistake? Or are you a MS employee with inside information?
You are over-thinking the problem. It's much simpler -- they search for outgoing containers/boxes/whatever that contain DVDs. When they find them, they check the manifest for that shipment. If the DVDs are not listed on the manifest -- they are probably illegally copied DVDs.
It's hardly novel -- my Nokia phone has done this for the last 2 or more years and I have no reason to believe other phones haven't been doing it for far longer. The only difference is that my phone doesn't have an "iPod mode".
The motion proposes that MediaSentry's actions are illegal under federal law, irrespective of whether the resulting logs are used in court or not:
So, according to this brief, it is illegal to use tcpdump on packets leaving my own network?
Isn't that an invention, not a discovery? From Webster:
Perhaps what is really needed is a change to the law which removes the word "discover" and leaves only inventions as patentable.
Doesn't "discovery" require that the subject of the discovery exist already? So, how does one discover something new? For example, when the "New World" was discovered, it's existence had been known for many years by the population who lived there, just not known by Europeans.
There were very good reasons why people used VHDL in the past. Because VHDL was an open language before Verilog, the cost of VHDL tools was historically lower than Verilog tools. Since this cost was much more important to FPGA designers, VHDL tended to dominate the FPGA market.
On ASIC side, the first mainstream commercial synthesis tool was Synopsys and Synopsys chose to support Verilog before supporting VHDL. Amongst all the other NRE costs in designing an ASIC, the added cost of using Verilog tools (instead of VHDL) was not really significant. Also, tools to support large designs advanced initially as Verilog tools.
Fast forward few years and Verilog is now open, the cost differential has now disappeared. However, VHDL had a lot of features related to design validation that were not in Verilog. In VHDL you can read and write files. Such things as configurations are supported, etc.. This type of capability makes it easier to write a testbench in VHDL, while on the Verilog side, additional tools and languages are commonly used.
Fast forwards a few more years to today and now we have System Verilog. This gives Verilog the capabilities that it lacked in comparison to VHDL and probably more. The price of VHDL tools is the same as Verilog tools.
Summary: it's clear that the future does not belong to VHDL. It looks like System Verilog is the future, although there are other contenders. So, if the choice is between VHDL and Verilog -- pick Verilog.
And if /tmp is on a different filesystem, this achieves nothing as far as ensuring integrity of the write.
And what happens if the process has write permissions for the file, but not the directory?
But is the design any good? If the advantage of EXT4 is better performance, how much of that performance improvement will be lost once the applications are fixed?
Yeah, apart from the possibly illegal part of changing your user agent string in order to gain access that you would not otherwise have.
Totall agree, we do need this. What we don't need is the rest of the baggage that newspapers want to keep selling to us. We don't need hundreds of newpaper titles, all carrying the same national news -- there is no added value any more in reprinting someone else's reporting.
Unfortunately, we don't currently have the newpapers that we need. The financial crash and the investigation of Madoff, where was that? The same person who gave details to the SEC about why Madoff must be running a Ponzi scheme took his information to a couple of major newspapers -- what did they do with it? Nothing. I did hear an interview of a reporter who had been investigating events that led up to the crash 2-3 years ago -- good reporting, yes? She wrote for UK newspaper.
The problem for the newspaper industry is the same as the buggy whip manufacturers when cars came along -- the demand for their product has dropped dramatically and many of the manufacturers will have to go out of business. Think of it this way, with hundreds of papers, the same function (collating and printing news) is repeated hundres of times each day, with no real added value over and above what the next guy is doing. There is no economic value in this. They had their shot, they made money (lots of money), but their time is now past. Technology and society have moved on.
Err, no. The amount of light entering is determined by te size of the lens. That light is spread across the sensor, so, with the same size lens and the same nmber of pixels, the amount of light per pixel (also assuming a fixed amount of time) is the same.
If you don't agre with the above statement, instead of just making bald assertions (as you have done in your posts above), explain why a larger pixel gathers more light if the same amount of light is spread across a larger area.
Do the sensors really heat up significantly? The sensors don't run at high speeds, so it seems unlikely that they would heat up significantly. Do you have a citation to support your claim?
The amount of light entering the camera is determined by the size of the lens. The size of the sensor does not affect this. With a fixed amount of light and a fixed number of pixels, the amount of light per pixel is constant. So, how does a larger sensor collect more light?
Would not the same amount of light falling on a smaller sensor give a larger signal? Instead of just telling me I'm wrong and suggesting that I don't understand, please explain.
If you are telling me that a smaller lens is going to result in less light and hence a smaller signal, that's quite another thing, but I was very specific when I posed my original question -- which was related to the size of the sensor alone and assumed that the lens size remained unchanged.
I'm not sure that this is valid. Why should a smaller sensor be more suceptable to EM noise? Even if it were, the smaller size would mean that the distribution of noise between pixels would be smaller, allowing for better correction during the processing of the picture.
Where do you get this from? Why should failure rates be higher? From a production standpoint, yields are generally higher for smaller die sizes.
Why? A larger pixel will be like a larger antenna -- larger currents will be induced in the antenna/pixel. The same total amount of light will shine on the pixel if the lens size is unchanged.