The original reason for aligning to track boundaries (a track is a cylinder-head pair) is that the first four sectors of MS-DOS' IO.SYS (IBMBIO.SYS) had to be contiguous and on a single track.
You know, I'm fine with that. The problem with WGA is that *in conjunction with making it difficult/expensive for OEMs to include reinstall media* they seem to have completely arbitrary and pernicious rules, which seem designed not so much to thwart piracy but to force legitimate customers to buy another copy.
Yeah, there is a rescue partition. Good thing it does me when the hard disk just exploded.
The tradition of putting a slash through a zero to distinguish it from O is much older than computers... it was standard operating procedure for Morse telegraphists when writing by hand.
If all they want is the tax money, they only need to increase taxes on fuel.
But that would mean SUVs would pay more per unit distance than small cars or hybrids. This would promote energy efficiency, which would make the oil companies mad.
This is an example of tivoization. It's legal with GPLv2. It is illegal with GPLv3. You can argue back and forth if it is moral, but I think you find that it's a very grey area under the circumstances.
Things can suddenly go very wrong. I lost a frequent contributor to one of my projects earlier this year... he apparently passed away within hours of going to the hospital to complain about a backache. The backache was spinal cancer.
The concept of a Branch-Call-Jump (BCJ) filter is well-known in the data compression community, and is a standard part of quite a few deployed compression products. Used as a front end to a conventional compression algorithm -- or, in this case, a binary compression algorithm -- does indeed give significant improvements. The application to binary diff is particularly interesting, since it means you can deal with branches and other references *over* the compressed region, so this is really rather clever.
It is a little old, but the price is excellent... $150 for the general public, and $120 for students. Even though there are cheaper boards on the market, I have found that the combination of features at a low price has been pretty unbeatable, and it is also available from a lot of distributors.
Now, in the interest of full disclosure, I have been focused heavily on retrocomputing projects... they're a lot of fun.
Personally, I would say that Verilog is more C-like: weakly typed, compact, efficient notation, whereas VHDL is much more Ada-like: strongly typed, often verbose, but can catch errors that the other one can't.
In industry, as far as I can tell, Verilog seems to be more used in North America and VHDL in Europe, so that might affect what you care about, too.
VGA is reasonably well documented, although a lot of the quirks aren't. It is, however, a horrible and painful design which had a zillion rarely-used features.
At the rate we're currently socializing and nationalizing everything...
That's rather ironic, given that there is pretty much nothing more socialistic that a resource available for unlimited use and paid for by taxes.
For what it's worth, I really think fuel taxes are the best option for paying for roads, especially since they have the nice side effect of encouraging fuel conservation. Tolls make sense on stretches of road that are extremely expensive (like bridges and tunnels), though.
Power systems change all the time. However, that hasn't kept the user terminal interface to stay virtually unchanged; it is the part with the most inertia and least incentive to change. I wouldn't be surprised if in 50 years we had a standard domestic low-power DC standard for small electronics, but I fully expect it will be in addition to instead of instead of the current 100-240 V terminals.
The two leads need to be supplied with a sine-wave alternating current peaking at 115 Volts" you have no way of knowing that in 50 years they'll be using Volts, AC, two leads, or know what a sine-wave is.
This is pretty silly... unless we assume a total collapse of civilization it seems highly unlikely that 50 years from now we wouldn't be able to decode the corrected form of that statement (which is both incorrect - 115 V is RMS, not peak; and incomplete - it doesn't specify the frequency.) A corrected form should look something like (assuming a relatively modern computer):
The two topmost leads need to be supplied with a 50-60 Hz (300-380 rad/s) sinewave alternating current at 100-240 V RMS (140-340 V peak) from a constant voltage source providing a minimum of 300 W. The bottom lead can optionally be connected to ground for safety.
Odds are relatively good that they'll laugh and plug in into the wall, which still has a NEMA 5-15 or NEMA 5-20 outlet.
What it comes down to is that Comcast wants to pick and choose between two different rulesets. In particular, they want to provide an "information service" (a service at the other end of the Internet), while giving preferrential service using their cable plant to compete against telcos, without telco regulation.
Sorry guys, you get to play one side or the other.
The first mover effect is just another case of hysteresis induced by positive feedback. This is a very common phenomenon present in many physical and other systems; it is hardly a surprise that it would exist in economic systems as well. In the context of the economy, it is simply a reflection that sometimes any standard is better than no standard, even if that standard is absurd.
The essence of a defensive patent pool is explained by this statement from the RPX website:
Members in RPX pay an annual fee to have a license to all patents and associated rights in the aggregation. Over time, companies receive perpetual licenses to all patents purchased in the aggregation and any patents that are sold during their tenure as members. Patents in the aggregation are also available to use in a counter-suit against any non-member who initiates litigation against a member. [my emphasis]
In other words, RPX don't need to assert patents themselves, they are leaving it up to their members/licensees to do so if a nonmember company is suing.
Is this better than a patent troll? If nothing else, it's probably a cheaper patent troll. Now, this is the kind of stuff that companies have been doing internally for a long time; this is just outsourcing.
Poor execution on the hardware side.
Transmeta felt they were taking too many risks on the software side, and adopted a hyper-conservative culture on the hardware side. The result ended up being both late and below target. All the software optimizations in the world could not help push more operations down the pipe than it could actually perform.
The increasing cost of memory performance
As time went on, the cost of x86 decode and scheduling in hardware went down, and the cost of memory performance -- caching systems, and so on -- went up. The VLIW instruction set consumed more icache than the native x86 instruction set.
TSMC meltdown
The best design in the world doesn't help if your fab partner don't deliver for their own design rules.
Worked on?
More like he was hired to sit in an office and be their "star" power.
Nothing could be further from the truth. Out of the five major components of the Crusoe firmware -- the dynamic translator, interpreter, nucleus (mini-OS), virtual I/O, and out-of-line handlers ("microcode"), Linus was the driving force, designer and primary implementor of one (the interpreter.) He eventually transitioned into an "advanced research" role, working on more "far out" projects.
A few things that made Commodore BASIC slow compared to some of the BASICs of the time:
A lot of parsing was done at runtime, rather than at entry time or program startup time. This meant that a lot of work got done over and over rather than once per execution, and also encouraged practices that made programs hard to read, like omitting comments and whitespace.
No support for integer-only arithmetic.
Very few high-level constructs. This didn't just turn programs into a mess of POKE statements, but frequently meant that a lot of them were needed -- all of which involved two floating-point to integer conversions to set a single byte.
Most BASICs at the time would at least tokenize at entry time, and many even converted programs to P-code for execution. It was still much slower than true assembly, but it wasn't anywhere near as bad as the CBM basic. Similarly, most (but again, not all) BASICs of the time supported integer-only arithmetic at least as an option.
I have to admit to being a bit spoiled at the time, since I first learned to program on the Swedish ABC80 computer, which had a very fast BASIC interpreter. Its follower, ABC800, even had a decent collection of high-level programming constructs in its BASIC. Too bad neither had acceptable graphics, nor a reasonable price point for home use (they ended up being sold mostly into schools and small offices.)
I have to say, Commodore BASIC was one of the worst BASIC interpreters available in the "home computer" market. It's pretty clear the C64 was really a game console with a keyboard, so it could be (very successfully, and legitimately so) sold as a computer. If you actually wanted the C64 to do anything interesting, you had the choice of assembly, or BASIC that looked like assembly but ran like crap. As far as I can tell, the C64 BASIC didn't use any of the techniques used by other BASICs from the same era to not run like a total dog.
Don't get me wrong, the C64 was a great machine. It just wasn't a great machine for BASIC programming.
Slashdot Mirror
Google "10Base-5".
The original reason for aligning to track boundaries (a track is a cylinder-head pair) is that the first four sectors of MS-DOS' IO.SYS (IBMBIO.SYS) had to be contiguous and on a single track.
You know, I'm fine with that. The problem with WGA is that *in conjunction with making it difficult/expensive for OEMs to include reinstall media* they seem to have completely arbitrary and pernicious rules, which seem designed not so much to thwart piracy but to force legitimate customers to buy another copy.
Yeah, there is a rescue partition. Good thing it does me when the hard disk just exploded.
The tradition of putting a slash through a zero to distinguish it from O is much older than computers... it was standard operating procedure for Morse telegraphists when writing by hand.
Nonsense. In 64-bit mode you can shift up to 63 bits at a time.
But that would mean SUVs would pay more per unit distance than small cars or hybrids. This would promote energy efficiency, which would make the oil companies mad.
This is an example of tivoization. It's legal with GPLv2. It is illegal with GPLv3. You can argue back and forth if it is moral, but I think you find that it's a very grey area under the circumstances.
Things can suddenly go very wrong. I lost a frequent contributor to one of my projects earlier this year... he apparently passed away within hours of going to the hospital to complain about a backache. The backache was spinal cancer.
The concept of a Branch-Call-Jump (BCJ) filter is well-known in the data compression community, and is a standard part of quite a few deployed compression products. Used as a front end to a conventional compression algorithm -- or, in this case, a binary compression algorithm -- does indeed give significant improvements. The application to binary diff is particularly interesting, since it means you can deal with branches and other references *over* the compressed region, so this is really rather clever.
It is a little old, but the price is excellent... $150 for the general public, and $120 for students. Even though there are cheaper boards on the market, I have found that the combination of features at a low price has been pretty unbeatable, and it is also available from a lot of distributors.
Now, in the interest of full disclosure, I have been focused heavily on retrocomputing projects... they're a lot of fun.
Personally, I would say that Verilog is more C-like: weakly typed, compact, efficient notation, whereas VHDL is much more Ada-like: strongly typed, often verbose, but can catch errors that the other one can't.
In industry, as far as I can tell, Verilog seems to be more used in North America and VHDL in Europe, so that might affect what you care about, too.
Personally, I prefer Verilog.
VGA is reasonably well documented, although a lot of the quirks aren't. It is, however, a horrible and painful design which had a zillion rarely-used features.
That's rather ironic, given that there is pretty much nothing more socialistic that a resource available for unlimited use and paid for by taxes. For what it's worth, I really think fuel taxes are the best option for paying for roads, especially since they have the nice side effect of encouraging fuel conservation. Tolls make sense on stretches of road that are extremely expensive (like bridges and tunnels), though.
Power systems change all the time. However, that hasn't kept the user terminal interface to stay virtually unchanged; it is the part with the most inertia and least incentive to change. I wouldn't be surprised if in 50 years we had a standard domestic low-power DC standard for small electronics, but I fully expect it will be in addition to instead of instead of the current 100-240 V terminals.
This is pretty silly... unless we assume a total collapse of civilization it seems highly unlikely that 50 years from now we wouldn't be able to decode the corrected form of that statement (which is both incorrect - 115 V is RMS, not peak; and incomplete - it doesn't specify the frequency.) A corrected form should look something like (assuming a relatively modern computer):
Odds are relatively good that they'll laugh and plug in into the wall, which still has a NEMA 5-15 or NEMA 5-20 outlet.
What it comes down to is that Comcast wants to pick and choose between two different rulesets. In particular, they want to provide an "information service" (a service at the other end of the Internet), while giving preferrential service using their cable plant to compete against telcos, without telco regulation.
Sorry guys, you get to play one side or the other.
The first mover effect is just another case of hysteresis induced by positive feedback. This is a very common phenomenon present in many physical and other systems; it is hardly a surprise that it would exist in economic systems as well. In the context of the economy, it is simply a reflection that sometimes any standard is better than no standard, even if that standard is absurd.
I guess we now know what generation disk integrated the Maxtor people/facilities. This presumably means Seagate joins Maxtor on the never-again list.
Dump them to hard disk, RAID array, what not; then threat the physical media as transient and/or a backup.
That way you can also index electronically and what not.
In other words, RPX don't need to assert patents themselves, they are leaving it up to their members/licensees to do so if a nonmember company is suing.
Is this better than a patent troll? If nothing else, it's probably a cheaper patent troll. Now, this is the kind of stuff that companies have been doing internally for a long time; this is just outsourcing.
What killed Transmeta was a few things things:
Transmeta felt they were taking too many risks on the software side, and adopted a hyper-conservative culture on the hardware side. The result ended up being both late and below target. All the software optimizations in the world could not help push more operations down the pipe than it could actually perform.
As time went on, the cost of x86 decode and scheduling in hardware went down, and the cost of memory performance -- caching systems, and so on -- went up. The VLIW instruction set consumed more icache than the native x86 instruction set.
The best design in the world doesn't help if your fab partner don't deliver for their own design rules.
Nothing could be further from the truth. Out of the five major components of the Crusoe firmware -- the dynamic translator, interpreter, nucleus (mini-OS), virtual I/O, and out-of-line handlers ("microcode"), Linus was the driving force, designer and primary implementor of one (the interpreter.) He eventually transitioned into an "advanced research" role, working on more "far out" projects.
You might find this link interesting.
A few things that made Commodore BASIC slow compared to some of the BASICs of the time:
Most BASICs at the time would at least tokenize at entry time, and many even converted programs to P-code for execution. It was still much slower than true assembly, but it wasn't anywhere near as bad as the CBM basic. Similarly, most (but again, not all) BASICs of the time supported integer-only arithmetic at least as an option.
I have to admit to being a bit spoiled at the time, since I first learned to program on the Swedish ABC80 computer, which had a very fast BASIC interpreter. Its follower, ABC800, even had a decent collection of high-level programming constructs in its BASIC. Too bad neither had acceptable graphics, nor a reasonable price point for home use (they ended up being sold mostly into schools and small offices.)
I have to say, Commodore BASIC was one of the worst BASIC interpreters available in the "home computer" market. It's pretty clear the C64 was really a game console with a keyboard, so it could be (very successfully, and legitimately so) sold as a computer. If you actually wanted the C64 to do anything interesting, you had the choice of assembly, or BASIC that looked like assembly but ran like crap. As far as I can tell, the C64 BASIC didn't use any of the techniques used by other BASICs from the same era to not run like a total dog.
Don't get me wrong, the C64 was a great machine. It just wasn't a great machine for BASIC programming.
Pelosi probably took impeachment off the table because there is no way it would get the necessary two-thirds majority in the half-Republican Senate.