Depends. If you have absolutely no version checking code anywhere in your site, then I would say NO; just leave it alone and don't worry. But if you do have version checking of any kind at all, to do anything different based on the version number of any browser, even though IE6 was not one of those, then I'd say YES; add the version check for IE6 and redirect them to Upgrade Your Browser.
If they want to use it internally, that's fine. Just don't use it for general net access. To accomplish that end, web sites should reject attempts to access using IE6 to make it so unworkable as an internet browser that people/companies will at least let their users that should be accessing the internet have an alternate better browser to do so with (never mind what that is). Redirect to a browser upgrade page or something.
I specifically avoid products with such a label because I know that means I can't replace it if it fails. One exception is Mountain Dew's limited edition with real sugar (but that's not something that fails).
Believe me, the industry -is- looking into ways of getting SSDs on to faster buses, but it takes time and some significant rearchitecture. Also, NAND sucks ass, with high block failure rates fresh out of the fab outweighed by sheer density. And it's only going to get worse as lithography gets smaller.
How about the PCIe bus? It's already reasonably mature technology and there's a huge installed base. They can build small cards and huge cards.
I'm looking for an SSD for the OS and programs to reside on, mounted read/only almost all the time (only writing when I need to upgrade it). This does not need sheer density as 16GB will be sufficient (that's GB, not TB). What I want is sheer SPEED. Speed of access and speed of transfer. Single level cells, not multi-level cells, is all that would be needed. And if the current fab tech still gives more space than I need, then mirror the data across the excess space to cover for block failures. Just make sure the access to the emulated drive works in an open way.
The thing to NOT do is build some drive device designed to talk SATA or SAS or some I/O interface, slap it onto a PCIe card, and try to squeeze PCIe bus speeds through that interface. This needs to be as native PCIe as possible. One controller between the PCIe bus and the flash chips is all that is needed. Build it on a x16 PCIe v3.0 and deliver 16 GB/s. I'd be very happy with an x8 at 8GB/s.
"If at any point during the process a human being actually clicked on the link and looked for the infringing content, then they would have realized it wasn't there," said Lipold. "Unfortunately, the bot the IFPI uses to flag piracy isn't that smart... This bot makes a report for security at the IFPI and they forward the whole list on to Google. Google seems to accept their list as is, no questions asked."
The thing is, Google doesn't have a choice.
Actually, Google does have a choice... once they are aware that the party making the complain has a track record of knowingly making flawed and/or fraudulent complaints. Still, this will take a number of documented errors on the part of the IFPI. It is still the IFPI, not Google, that needs to be sued in civil court to recover losses, and pursued for prosecution in criminal court for making repeated fraudulent claims.
My first DoS discovery was in October 1976. On IBM mainframes running VM/CMS, I found I could take down the entire system from an ASCII serial port connection, without even being logged in. At any prompt, including the "LOGON:" prompt (hence why being logged in was not needed), just press the RETURN key followed immediately by the BREAK key.
A couple years later when I obtained the source code to the system (bought it on a reel of tape, from IBM, for $150) I found the bug in the code that caused it. The "CP" kernel went into a loop trying to send a command to the I/O controller to reverse the direction of the half-duplex serial port, which would always fail because it had not received the interrupt informing it of the BREAK status, which it would never get because all I/O interrupts were masked off at that point.
... they could achieve that with absolutely zero collateral damage. However, I highly doubt any government agency, especially the Australian government, could come anywhere close to achieving that. And that is THE reason it should not even be considered, much less attempted.
However, you cannot claim ownership to what already is public domain. Sure, you can make a new work consisting of parts that are public domain. If your work is creative, it can then be copyrighted. But the original public domain parts remain public domain and not subject to royalties.
Follow the "snowflakes" link and look at the bottom of the page:
Copyright/Public domain works
Wilson Bentley did not copyright his photographs and thus they are in the public domain and free to use for any purpose.
HOWEVER
No materials or images from this (or any other) website may be resold in any form (print or electronic).
The Public Domain status does not give you the right to resell material unless you have access to the original source and permission from the owner to reproduce the material. Any published works of Public Domain material is only "Royalty free" if explicitly stated.
WTF? Someone just doesn't understand what Public Domain really is.
They've known this was coming for YEARS. They should have been renumbering W.X.Y.Z to 10.X.Y.Z over the past decade. What hasn't been renumbered, just NAT it and lose the ability to communicate with all of the net until the renumbering is finished.
Find out what companies have/8 allocations without a purpose that is legitimate in 2010. Drive them out of business. Let ARIN grab the IP space back. Or just grab it back anyway (let them keep the first/20 of it).
He probably couldn't figure out the difference between "O" and "0" which are right next to each other. He could have just set up his own top level domain in a self-rooted name server and typed "x", too:-)
This is the same old silly argument that leads to so many bad standards and ideas that get entrenched just because they are first. Firstness does NOT mean best in all ways. Often things that are first are terrible is quite many, usually because so many things were not learned at the time it came out. The point here is to specifically NOT use the argument "because it is already in use" as a counter to the argument "let's change it". Instead, such argument should justify why the change would cost more now, than leaving it the way it is forever. With your argument, we wouldn't even have digital TV. And in the case of OTA TV, the "TONS of stuff" was "ALL the stuff" and digital still won. In the case of H.264 "TONS of stuff" is still just a fraction. It's not like video has expanded to everywhere it will be on the web and internet... not even close. Video is just getting started and what runs H.264 right now is a tiny fraction of all that will be running video in the coming years.
The question is which to use for the growth that video will bring. The dilemma is both H.264 and Theora each do not solve all the issues. And the issues exist both in the decoding as well as encoding realms. So just having both formats available in web pages, or having both formats in the browsers, is not a viable solution. Making hardware chips that do handle Theora for phones is only a partial solution due to the bandwidth issue that phones have (which isn't a big issue for broadband, once the net neutrality rules get established that create strong competition). The answer may well be that the web will use Theora and/or Dirac (which I consider to be a better option than Theora, but it doesn't seem to be getting the press coverage it should), and the phone provides set gateways that convert to H.264.
Greed didn't fix the LZW (for GIF) licensing mess. Why do you think suits will figure it out this time? Remember, the reason people like that have no clue is because they specifically reject clues.
... and also by requiring a royalty based on number of real or potential viewers for each encoding being done... and also by requiring per-encode operation royalties when using already paid for and licensed encoders.
Still, MPEG/H.264 could still succeed because there is a lot of specific demand for it. And it makes economic sense in some situations, such as mobile phones where an H.264 hardware decoder is already present, and in others where the licensee is "too big to fail" and has a royalty cap. So, much effort needs to be done to ensure that an open standard is at least a strong viable option.
Slashdot Mirror
Depends. If you have absolutely no version checking code anywhere in your site, then I would say NO; just leave it alone and don't worry. But if you do have version checking of any kind at all, to do anything different based on the version number of any browser, even though IE6 was not one of those, then I'd say YES; add the version check for IE6 and redirect them to Upgrade Your Browser.
If they want to use it internally, that's fine. Just don't use it for general net access. To accomplish that end, web sites should reject attempts to access using IE6 to make it so unworkable as an internet browser that people/companies will at least let their users that should be accessing the internet have an alternate better browser to do so with (never mind what that is). Redirect to a browser upgrade page or something.
I specifically avoid products with such a label because I know that means I can't replace it if it fails. One exception is Mountain Dew's limited edition with real sugar (but that's not something that fails).
No. I want speed. I want to be able to suck 16GB of the OS out of the SSD and into RAM in 0.03ms. So there :-)
Believe me, the industry -is- looking into ways of getting SSDs on to faster buses, but it takes time and some significant rearchitecture. Also, NAND sucks ass, with high block failure rates fresh out of the fab outweighed by sheer density. And it's only going to get worse as lithography gets smaller.
How about the PCIe bus? It's already reasonably mature technology and there's a huge installed base. They can build small cards and huge cards.
I'm looking for an SSD for the OS and programs to reside on, mounted read/only almost all the time (only writing when I need to upgrade it). This does not need sheer density as 16GB will be sufficient (that's GB, not TB). What I want is sheer SPEED. Speed of access and speed of transfer. Single level cells, not multi-level cells, is all that would be needed. And if the current fab tech still gives more space than I need, then mirror the data across the excess space to cover for block failures. Just make sure the access to the emulated drive works in an open way.
The thing to NOT do is build some drive device designed to talk SATA or SAS or some I/O interface, slap it onto a PCIe card, and try to squeeze PCIe bus speeds through that interface. This needs to be as native PCIe as possible. One controller between the PCIe bus and the flash chips is all that is needed. Build it on a x16 PCIe v3.0 and deliver 16 GB/s. I'd be very happy with an x8 at 8GB/s.
The proper defendant is the IFPI.
"If at any point during the process a human being actually clicked on the link and looked for the infringing content, then they would have realized it wasn't there," said Lipold. "Unfortunately, the bot the IFPI uses to flag piracy isn't that smart... This bot makes a report for security at the IFPI and they forward the whole list on to Google. Google seems to accept their list as is, no questions asked."
The thing is, Google doesn't have a choice.
Actually, Google does have a choice ... once they are aware that the party making the complain has a track record of knowingly making flawed and/or fraudulent complaints. Still, this will take a number of documented errors on the part of the IFPI. It is still the IFPI, not Google, that needs to be sued in civil court to recover losses, and pursued for prosecution in criminal court for making repeated fraudulent claims.
My first DoS discovery was in October 1976. On IBM mainframes running VM/CMS, I found I could take down the entire system from an ASCII serial port connection, without even being logged in. At any prompt, including the "LOGON:" prompt (hence why being logged in was not needed), just press the RETURN key followed immediately by the BREAK key.
A couple years later when I obtained the source code to the system (bought it on a reel of tape, from IBM, for $150) I found the bug in the code that caused it. The "CP" kernel went into a loop trying to send a command to the I/O controller to reverse the direction of the half-duplex serial port, which would always fail because it had not received the interrupt informing it of the BREAK status, which it would never get because all I/O interrupts were masked off at that point.
... they could achieve that with absolutely zero collateral damage. However, I highly doubt any government agency, especially the Australian government, could come anywhere close to achieving that. And that is THE reason it should not even be considered, much less attempted.
... remove the enabling technology from the standards and the browsers. Problem solved.
42
The units depends on what combination you want of:
... and double it.
There, finished it for ya.
However, you cannot claim ownership to what already is public domain. Sure, you can make a new work consisting of parts that are public domain. If your work is creative, it can then be copyrighted. But the original public domain parts remain public domain and not subject to royalties.
Follow the "snowflakes" link and look at the bottom of the page:
Copyright/Public domain works
Wilson Bentley did not copyright his photographs and thus they are in the public domain and free to use for any purpose.
HOWEVER
No materials or images from this (or any other) website may be resold in any form (print or electronic).
The Public Domain status does not give you the right to resell material unless you have access to the original source and permission from the owner to reproduce the material. Any published works of Public Domain material is only "Royalty free" if explicitly stated.
WTF? Someone just doesn't understand what Public Domain really is.
... also block all the people on the list?
They've known this was coming for YEARS. They should have been renumbering W.X.Y.Z to 10.X.Y.Z over the past decade. What hasn't been renumbered, just NAT it and lose the ability to communicate with all of the net until the renumbering is finished.
Find out what companies have /8 allocations without a purpose that is legitimate in 2010. Drive them out of business. Let ARIN grab the IP space back. Or just grab it back anyway (let them keep the first /20 of it).
My work upstream ISP wants to charge lots of extra money to do IPv6. We need a new law.
And that subnet is?
He probably couldn't figure out the difference between "O" and "0" which are right next to each other. He could have just set up his own top level domain in a self-rooted name server and typed "x", too :-)
That's their problem for being stupid. They should have used 10.10.10.10 or 10.4.10.4 or something like that.
This is the same old silly argument that leads to so many bad standards and ideas that get entrenched just because they are first. Firstness does NOT mean best in all ways. Often things that are first are terrible is quite many, usually because so many things were not learned at the time it came out. The point here is to specifically NOT use the argument "because it is already in use" as a counter to the argument "let's change it". Instead, such argument should justify why the change would cost more now, than leaving it the way it is forever. With your argument, we wouldn't even have digital TV. And in the case of OTA TV, the "TONS of stuff" was "ALL the stuff" and digital still won. In the case of H.264 "TONS of stuff" is still just a fraction. It's not like video has expanded to everywhere it will be on the web and internet ... not even close. Video is just getting started and what runs H.264 right now is a tiny fraction of all that will be running video in the coming years.
The question is which to use for the growth that video will bring. The dilemma is both H.264 and Theora each do not solve all the issues. And the issues exist both in the decoding as well as encoding realms. So just having both formats available in web pages, or having both formats in the browsers, is not a viable solution. Making hardware chips that do handle Theora for phones is only a partial solution due to the bandwidth issue that phones have (which isn't a big issue for broadband, once the net neutrality rules get established that create strong competition). The answer may well be that the web will use Theora and/or Dirac (which I consider to be a better option than Theora, but it doesn't seem to be getting the press coverage it should), and the phone provides set gateways that convert to H.264.
At what price (for both decoding and encoding)?
Greed didn't fix the LZW (for GIF) licensing mess. Why do you think suits will figure it out this time? Remember, the reason people like that have no clue is because they specifically reject clues.
... and also by requiring a royalty based on number of real or potential viewers for each encoding being done ... and also by requiring per-encode operation royalties when using already paid for and licensed encoders.
Still, MPEG/H.264 could still succeed because there is a lot of specific demand for it. And it makes economic sense in some situations, such as mobile phones where an H.264 hardware decoder is already present, and in others where the licensee is "too big to fail" and has a royalty cap. So, much effort needs to be done to ensure that an open standard is at least a strong viable option.