Patent applications are secret. What if, in the interval between submission and grant, another application is received covering the same invention, all overlapping claims are denied? The standard is supposed to be "not obvious to someone versed in the state of the art." If there are two submissions for the same thing in a short period of time, that proves that the state of the art is naturally evolving toward the scheme in the application.
The biggest problem would be the ability to ship products in that interval, which is currently practical.
The window could be more explicit, too: application date plus two months, for example.
"Memory is like an orgasm. It's a lot better if you don't have to fake it." -- Seymour Cray, on virtual memory.
This is even more true today, with memory speeds growing faster than disk speeds. Don't use virtual memory for anything that is actually used more than once. I was recently asked to bump the swap on a 128G machine to 1X; imagine how long it would take to access all that memory only once, let alone multiple passes. That machine has a 1-2G swap. Any process that actively used more than 128G wouldn't finish in the lifetime of the project anyway.
It may make sense to enable a lot of swap if you're running Linux and turn off memory overcommit. This is "unused" memory that's required because many programs only run efficiently with overcommit turned on, and you don't lose performance because the memory is never addressed anyway.
Swap in any case should be based on memory size alone only for special purposes like suspend; otherwise, use the largest image you expect minus physical RAM.
"Because a computer can't truly think of a random number" is correct only if you're referring to the "think" part. If you have a newer Intel chipset based PC, you probably can get a few random numbers out of it: http://www.intel.com/design/security/rng/rng.htm
Back when DVD-ROMs were first developed, pleas to include "linking areas," also called "run-in" on magnetic media, were ignored. This meant that there would be no space to write bits to allow the data clock to be recovered. This is not a problem with a continuously written disc, as the "earlier" data can be used.
However, with most technologies, the start of writing can vary hundreds of bits from its nominal location. This means that if you write sector 1, and later write sector 2 on its tail, there is no way to align the data clock to the data in time to read the data of sector 2, since the first bits are used to recover the clock.
DVD-RAM addressed this problem by introducing a new format. There are small embossed address headers to mark where each sector might be, and a small written run-in for each recorded sector. This media is highly susceptible to scratches, since the embossed headers are easily obscured by dust or a scratch. Also, having the embossed headers means that these discs are INCOMPATIBLE with DVD-ROM drives.
DVD-RAM discs also use land/groove recording. This means changes to the servo mechanism just to follow the track. A DVD-ROM drive must add in special reader circuitry to handle the data coding, different block architecture (ROMs use blocks of 16 sectors), different servos for land/groove structure, etc. DVD-RAM is endorsed by the DVD Forum, so is the "official" format.
DVD-RW is also endorsed by the DVD Forum, but for video use. Its intended application for reliable use is to not put data in the first block where you first start writing (it becomes your run-in). The "link point" where writing begins and ends occurs about 1.5 sectors into the 16 sector block. This means that the first 1.5 sectors, while readable, have unrelated data to what you just recorded! Also, some amount of the beginning of what you did write will be unreadable because the data clock needs to be recovered, and the words synchronized to a sync mark. Realistically, the first two sectors will be trashed in any sector that contains a link point. These sectors are recoverable by ECC, but it severely reduces the amount of recovery that can be done on your real data!
DVD-RW uses a shallow, low frequency, wobbled groove with some high frequency "ticks" (occurring at a low frequency) to gain some positional accuracy. These ticks are extremely susceptible to dirt, etc.
Overall, DVD-RW is best used as a re-usable DVD-R.
DVD+RW uses a shallow wobbled groove. However, this wobble is a continuous high frequency, and uses phase modulation to contain address information. The link point is a few bytes from the end of the sector, meaning that only a few bytes are sacrificed to the ECC gods (instead of 4K in the case of DVD-RW), and sufficient data exists to act as the run-in for the newly written sector.
The logical layout of DVD+RW is identical to the logical layout of DVD-ROMs. Any drive that conforms to the DVD-ROM standards will read a DVD+RW disc. Unfortunately, some manufacturers took shortcuts (like assuming a low reflectivity disc is a dual layer disc, rather than reading the proper parameter from the media).
The design characteristics of DVD+RW are such that it works well for randomly written data, and is compatible with DVD-ROM drives.
In short, DVD-RAM only works if the reader was deliberately designed to read it. DVD-RW is readable by properly designed ROM readers, but is a very fragile format. Finally, DVD+RW is readable in properly designed ROM readers, and has the most robust inherent design.
Standards that describe these DVD formats are available from ECMA.
Disclaimer: I work for HP, and used to work in their DVD group. Opinions expressed here are mine, not HPs.
I believe all of these surveys - that overall, the availability of samples is a great promotional tool. I'll bet that RIAA believes this too. However, now everyone will be willing to try "unknown," and worse, unsigned, bands.
RIAA is not worried too much about the piracy, but that is the only legal front they have against Napster. What really scares them is the availability of a distribution channel that they don't control. As another post mentions, they are pushing bands that they can control, not necessarily those that are best. That strategy naturally limits the number of bands that can be made available in a given time frame.
Napster gives artists and consumers a real choice. Artists will now see competition for their work, rather than relying on a lottery system to get signed. Consumers can pre-sort the good stuff from the junk. RIAA knows they've got tons of un-salable junk if better music becomes available.
P.S. For all the rhetoric out of RIAA about respect of copyrights, why are so many artists claiming that the studios do not own many of the on-line distribution rights they claim to own?
Oops - didn't finish:
1201(c)(1) states:
Nothing in this section shall affect rights, remedies, limitations, or defenses to copyright infringement, including fair use, under this title.
Therefore, if any scheme interferes with your fair use rights, you can legally circumvent those schemes. CSS interferes with the ability to make derivative works (i.e. fast forward when you want to), the ability to excerpt for a review, etc.
Since DeCSS is necessary to restore fair use rights, 1201(c)(1) makes it legal.
Slashdot Mirror
What about "only one to file"?
Patent applications are secret. What if, in the interval between submission and grant, another application is received covering the same invention, all overlapping claims are denied? The standard is supposed to be "not obvious to someone versed in the state of the art." If there are two submissions for the same thing in a short period of time, that proves that the state of the art is naturally evolving toward the scheme in the application.
The biggest problem would be the ability to ship products in that interval, which is currently practical.
The window could be more explicit, too: application date plus two months, for example.
The aforementioned practice is patent #9,763,348.
"Memory is like an orgasm. It's a lot better if you don't have to fake it." -- Seymour Cray, on virtual memory.
This is even more true today, with memory speeds growing faster than disk speeds. Don't use virtual memory for anything that is actually used more than once. I was recently asked to bump the swap on a 128G machine to 1X; imagine how long it would take to access all that memory only once, let alone multiple passes. That machine has a 1-2G swap. Any process that actively used more than 128G wouldn't finish in the lifetime of the project anyway.
It may make sense to enable a lot of swap if you're running Linux and turn off memory overcommit. This is "unused" memory that's required because many programs only run efficiently with overcommit turned on, and you don't lose performance because the memory is never addressed anyway.
Swap in any case should be based on memory size alone only for special purposes like suspend; otherwise, use the largest image you expect minus physical RAM.
"Because a computer can't truly think of a random number" is correct only if you're referring to the "think" part. If you have a newer Intel chipset based PC, you probably can get a few random numbers out of it:
http://www.intel.com/design/security/rng/rng.htm
It sure doesn't seem like snake oil:m
http://www.intel.com/design/security/rng/CRIwp.ht
Back when DVD-ROMs were first developed, pleas to include "linking areas," also called "run-in" on magnetic media, were ignored. This meant that there would be no space to write bits to allow the data clock to be recovered. This is not a problem with a continuously written disc, as the "earlier" data can be used.
However, with most technologies, the start of writing can vary hundreds of bits from its nominal location. This means that if you write sector 1, and later write sector 2 on its tail, there is no way to align the data clock to the data in time to read the data of sector 2, since the first bits are used to recover the clock.
DVD-RAM addressed this problem by introducing a new format. There are small embossed address headers to mark where each sector might be, and a small written run-in for each recorded sector. This media is highly susceptible to scratches, since the embossed headers are easily obscured by dust or a scratch. Also, having the embossed headers means that these discs are INCOMPATIBLE with DVD-ROM drives.
DVD-RAM discs also use land/groove recording. This means changes to the servo mechanism just to follow the track. A DVD-ROM drive must add in special reader circuitry to handle the data coding, different block architecture (ROMs use blocks of 16 sectors), different servos for land/groove structure, etc. DVD-RAM is endorsed by the DVD Forum, so is the "official" format.
DVD-RW is also endorsed by the DVD Forum, but for video use. Its intended application for reliable use is to not put data in the first block where you first start writing (it becomes your run-in). The "link point" where writing begins and ends occurs about 1.5 sectors into the 16 sector block. This means that the first 1.5 sectors, while readable, have unrelated data to what you just recorded! Also, some amount of the beginning of what you did write will be unreadable because the data clock needs to be recovered, and the words synchronized to a sync mark. Realistically, the first two sectors will be trashed in any sector that contains a link point. These sectors are recoverable by ECC, but it severely reduces the amount of recovery that can be done on your real data!
DVD-RW uses a shallow, low frequency, wobbled groove with some high frequency "ticks" (occurring at a low frequency) to gain some positional accuracy. These ticks are extremely susceptible to dirt, etc.
Overall, DVD-RW is best used as a re-usable DVD-R.
DVD+RW uses a shallow wobbled groove. However, this wobble is a continuous high frequency, and uses phase modulation to contain address information. The link point is a few bytes from the end of the sector, meaning that only a few bytes are sacrificed to the ECC gods (instead of 4K in the case of DVD-RW), and sufficient data exists to act as the run-in for the newly written sector.
The logical layout of DVD+RW is identical to the logical layout of DVD-ROMs. Any drive that conforms to the DVD-ROM standards will read a DVD+RW disc. Unfortunately, some manufacturers took shortcuts (like assuming a low reflectivity disc is a dual layer disc, rather than reading the proper parameter from the media).
The design characteristics of DVD+RW are such that it works well for randomly written data, and is compatible with DVD-ROM drives.
In short, DVD-RAM only works if the reader was deliberately designed to read it. DVD-RW is readable by properly designed ROM readers, but is a very fragile format. Finally, DVD+RW is readable in properly designed ROM readers, and has the most robust inherent design.Standards that describe these DVD formats are available from ECMA.
Disclaimer: I work for HP, and used to work in their DVD group. Opinions expressed here are mine, not HPs.
...is independent labels.
I believe all of these surveys - that overall, the availability of samples is a great promotional tool. I'll bet that RIAA believes this too. However, now everyone will be willing to try "unknown," and worse, unsigned, bands.
RIAA is not worried too much about the piracy, but that is the only legal front they have against Napster. What really scares them is the availability of a distribution channel that they don't control. As another post mentions, they are pushing bands that they can control, not necessarily those that are best. That strategy naturally limits the number of bands that can be made available in a given time frame.
Napster gives artists and consumers a real choice. Artists will now see competition for their work, rather than relying on a lottery system to get signed. Consumers can pre-sort the good stuff from the junk. RIAA knows they've got tons of un-salable junk if better music becomes available.
P.S. For all the rhetoric out of RIAA about respect of copyrights, why are so many artists claiming that the studios do not own many of the on-line distribution rights they claim to own?
Oops - didn't finish: 1201(c)(1) states: Nothing in this section shall affect rights, remedies, limitations, or defenses to copyright infringement, including fair use, under this title. Therefore, if any scheme interferes with your fair use rights, you can legally circumvent those schemes. CSS interferes with the ability to make derivative works (i.e. fast forward when you want to), the ability to excerpt for a review, etc. Since DeCSS is necessary to restore fair use rights, 1201(c)(1) makes it legal.
The DMCA, in section 1201, includes the anti-circumvention language. 1201(c)(1) states: