Sure, the cards might have been resold, but they are branded cisco items bearing the entire cisco interface and functionality - somehow I doubt outright fake chipsets and devices like this can be produced by anyone other than cisco themselves.
Whether or not this is what happened in this particular case, I don't know. But in general, the issue is not that someone has taken the time to reverse-engineer a complete product and produce it again from the ground up. The "fake" hardware likely comes from any combination of several places:
Chip vendors often have huge inventories of chips that failed testing, but are otherwise marginally functional. Some of
these chips could be branded and sold by an unscrupulous factory (hehe, that sounds funny:-) as legitimate parts. Or more likely, they can be sold to the guy in the next point:
Factories in 3rd-world or offshore countries (cheap labour) can and do produce legitimate hardware items for some of the big-name companies, of which Cisco is one. That is, they produce the legit hardware by day. After hours, or for a portion of the day, they can use the exact same process, exact same tooling, etc. to produce the knock-offs. These are then distributed through some other means.
Finally, the contract manufacturers (factories in the above point) will have many products that failed QA, but a marginally
functional. These also can get sold as counterfeit gear.
So the reverse-engineering is not so much the issue (although I am sure there is some degree of that). But as another poster mentioned, if you have the expertise to completely reverse-engineer something and reproduce it, you should go into business yourself selling a competing product that is much cheaper:-)
... have viewed the PPT presentation of Paul Allen's yacht, circulating a while back. Or other PPT things that get mailed around almost as often as the funny video clips we send each other.
And this is not a virus: I choose to send these to my friends:-(
The computing equivalent - of regularly applying updates - really ought to be just as obvious.
Ummm... nope. Why should it be obvious? Most people have no idea what is involved in constructing a complex piece of software. This is essentially what you said in the very next sentence:
[...] the equivalent understanding - that almost no sufficiently-complex code is perfect first time - hasn't made it into the collective subconscious.
The car analogy is a good one. I expect things to stop working eventually on my car (needs oil, new tires, belts, etc.).
Similarly, I expect things to eventually stop working on my computer (new hard disk, power supply, fans, etc). But I don't expect the software to stop working, or to need "regular maintenance". The fact that most software does, in fact, need regular updates only to fix security holes is not expected by the average net user.
Bill Gates has achieved what most people only dream of in terms of their life's ambitions. What do most people want? Money? Sure, but that is not the end of everything. Most (normal) people actually want to make a contribution to society/the world; to leave a legacy, if you will. (Granted, Bill has already done that.)
So when you have succeeded beyond your wildest ambitions, then what? Gates cannot actually spend his money on himself fast enough. There comes a point when you start to want to spend it on your legacy instead. Hence, the charity funding. But this is still ambition.
(Of course, I wish more people would reach that stage.)
I disagree with his grouping. The vast majority of the six billion doesn't give a shit about software bugs. They're primarily concerned with their ability to exchange services and products for money and vice versa, and if they do have free time, they don't spend it fretting about XP's bug count.
No kidding. I want to be able to USE a software product in the way that it is intended (even advertised) to be used. When a product ships with defects that make me wonder what kind of spaghetti-code mess is behind the pretty GUI, I do get just a little frustrated. I am a programmer, but I would classify myself more as Group 2 than Group 1.
The author is thinking like a lot of PHBs in large corporations do. But there *is* another side. It *is* possible to develop high-quality software with confidence, especially in medium-sized applications. I know lots of programmers who really should not be writing core software, because they produce bug-prone, non-robust code.
The Solaris Web Start command-line interface has been replaced with suninstall to improve usability.
Did anyone else read this as s-uninstall? I was wondering why it was so important to include an uninstall option right away, and to feature it so prominently in the article.:-)
FYI, I recently did a test using a Xilinx FPGA to see how fast I could clock a signal from one register to an adjacent register (i.e. no combinatorial logic in-between the flops). The top speed on V2Pro (the latest and greatest from Xilinx --.13 process and all that) was about 500 MHz in a -6 speed grade.
(I am going from memory here, it *may* have been as high as 600 MHz.) In any case, like the man says, CPU speeds are a long ways away. The *top* top speed you can get for any real-world logic is going to be about 400 MHz, and that only for very specialized applications.
Re:Down with specialized pieces
on
Lego Addictions
·
· Score: 1
Even online sales don't sell sets of assorted
bricks. Go to lego.com and try to find a package with assorted bricks in different colors. It doesn't exist. (The best I have found was assorted black bricks: 1x2, 1x4, 1x8, 2x2, 2x4, etc. And the shipping and handling cost is more than the cost of the bricks.)
What really happens with the sets you can buy today at WalMart is as follows:
Daddy takes 6-year-old Robby to the store.
Robby sees lego sets with cool themes that he
has recently seen in movies.
Daddy, on the other hand, sees a tub of assorted bricks packaged
in a boring plastic case.
Guess which one gets purchased? (...even though daddy knows that Robby will *eventually* get more use out of the tub of bricks than he will out of the Harry Potter castle.)
The tone of the article makes it sound like Linux is losing bad. Certainly the little blurb on/. tries to make it sound bad for Linux...
But you can't compare revenue generated by a FREE operating system with revenue generated by a rather
costly operating system. The goals are completely different.
Last year the Pentagon awarded Raytheon an 18-year contract valued at $3 billion to produce more than 10,000 missiles for both the Navy and the Air Force.
This sounds sort of like the tale of NASA spending millions to develop a pen that works in zero-G environments. The Russian's, when confronted with the same problem, used a pencil.
It sounds strange to rely on a missile that has a tight turning radius when all you have to do is swivel the launch rails. Surely the problem of how to swivel the launch rails is easier to solve than how to make a missile turn better.
That being said, there are many other benefits to a tight-turn-radius missile.
If, as the article says, this is to be used in printers and other small embedded systems, there is still a need for serial and parallel ports. Many *many* devices out there still use these things. Same goes for the large headers which take up space above the memory connectors.
Gates directed 7,000 programmers to spend February scouring the Windows operating system for openings hackers might exploit to steal data or shut down systems.
Wow, 7000 programmers! I bet they figure out how to close the barn door.
They claim that a city will do well if they install a broadband communications network that connects citizens, local businesses and the global marketplace.
Sure, but there is another factor that is important for high-tech companies that Gartner is missing: skilled workers.
I work for a company that has one of its offices in Edmonton, AB, Canada. Why Edmonton? Let's see:
Canadian dollar paychecks for employees
cheaper costs for business infrastructure
University of Alberta in Edmonton provides quality engineering talent.
And yes, high-speed internet.
The third point there is very important. Yes, it is possible to attract SV engineers to Edmonton, but it takes a *lot* of effort and incentive. One major problem is that a SV engineer may sell a bungalow for $600,000 USD. In Edmonton, you can get a mansion for $200,000 USD; there are no houses to be bought for $600K. So the SV engineer suddenly has $400K to pay taxes on.
So, the vast majority of people in R&D are local-area Canadians.
But they are really in two different market segments. Yes, they are both embedded, but the main focus of the high-end embedded MIPS chips is performance without the high power consumption.
ARM has excellent performance for some applications, but for many embedded systems, using an ARM is simply not an option.
Speaking as someone who does digital design: I would *never* overclock a chip on a system that I wanted to be reliable unless I knew that the manufacturer was deliberately marketing their chips at a lower speed than they were capable of. There are just too many ways that this can bite you.
The main problem is that you just don't know when you have gone over the line. Overclocking might be suitable in most cases except that one critical path which doesn't get executed very much.
That being said, for getting the latest gaming system, overclock to your heart's content. Who cares if the game crashes once in a while?
The labels see signs of a similar death spiral in the United States. Sales of CD singles are off 41 percent, compared with the same time last year, and album sales are effectively flat -- up less than 1 percent from a year ago, according to SoundScan, a market research firm that tracks retail music sales.
Hello!??? Economy just *might* have something to do with this.
In Germany alone, one survey by market researcher GfK found that blank CD sales jumped 129 percent this year. Purchases of pre-recorded music dropped 2.2 percent in the same period.
Indeed, blank CDs now outsell recorded discs in Europe and Canada, according to one label executive.
So, blank CD purchases are up by over 100%, and they outsell recorded CDs, yet recorded music sales have only dropped by 2.2%.
Gosh! Could it be that people by blank CDs for purposes other than pirating music?
Nahh. We should pay extra for each blank CD we buy, because we need to reimburse the music industry for pirated music.
"Second, we know we need to continue to focus in on our relationship with our customers. This is an area where we need to be ever vigilant. Certainly, as Bill talked about, we have opportunities for improvement in security, in virus protection, in the way we license and sell our products, and the reminders on that are always in front of us." --Steve Ballmer
I think this pretty much sums up a lot of what is wrong with Microsoft:
1) Security
2) The way they license and sell products.
At least they are realizing that market opinion is starting to go against them, and are trying to change this. I don't love Microsoft, but if they started to change their licensing tactics, I would be more inclined to buy their products.
Not a chance. You *can* do what you want in high-tech, and have fun doing it.
And... it's a *lot* better than school. You actually get paid decent money. There are less deadlines, less overtime hours, and less stress. I don't miss school one bit.
But then, I was never the "career" student.
But in general, start doing something you like doing, even if it pays less. If you like doing it, chances are you will be good at it. If you are good at it, chances are you will be noticed. If you are noticed, chances are your pay will go up too.
FibreChannel sucks for doing TCP/IP, but is really quite good at SCSI (called FCP).
But in terms of doing optical *on* the motherboard (or directly to the processor), this is a very long way off. Even if we could find a cheap way of converting electrical signals to optical signals in a processor chip or ASIC, the issue of manufacturing the motherboard PCB with optical traces is a big one.
The other thing to note here is that we would likely not see optical busses. Rather, a single multi-gigabit serial channel would be more cost-effective. But copper also has a lot of room to grow in this regard...
Slashdot Mirror
Whether or not this is what happened in this particular case, I don't know. But in general, the issue is not that someone has taken the time to reverse-engineer a complete product and produce it again from the ground up. The "fake" hardware likely comes from any combination of several places:
Uhhh... no, everyone is not a criminal. That is why we have a difference between criminal and civil (torte?) law.
Dimensions are 3.5" height, 2" (or 1.5", I forget) width, and 0.5" thickness.
I was thinking of moderating these, but I couldn't find the "-1 whoosh" mod.
And this is not a virus: I choose to send these to my friends :-(
The computing equivalent - of regularly applying updates - really ought to be just as obvious.
Ummm... nope. Why should it be obvious? Most people have no idea what is involved in constructing a complex piece of software. This is essentially what you said in the very next sentence:
[...] the equivalent understanding - that almost no sufficiently-complex code is perfect first time - hasn't made it into the collective subconscious.
The car analogy is a good one. I expect things to stop working eventually on my car (needs oil, new tires, belts, etc.). Similarly, I expect things to eventually stop working on my computer (new hard disk, power supply, fans, etc). But I don't expect the software to stop working, or to need "regular maintenance". The fact that most software does, in fact, need regular updates only to fix security holes is not expected by the average net user.
Gates is still ambitious.
Bill Gates has achieved what most people only dream of in terms of their life's ambitions. What do most people want? Money? Sure, but that is not the end of everything. Most (normal) people actually want to make a contribution to society/the world; to leave a legacy, if you will. (Granted, Bill has already done that.)
So when you have succeeded beyond your wildest ambitions, then what? Gates cannot actually spend his money on himself fast enough. There comes a point when you start to want to spend it on your legacy instead. Hence, the charity funding. But this is still ambition.
(Of course, I wish more people would reach that stage.)
The author is thinking like a lot of PHBs in large corporations do. But there *is* another side. It *is* possible to develop high-quality software with confidence, especially in medium-sized applications. I know lots of programmers who really should not be writing core software, because they produce bug-prone, non-robust code.
Did anyone else read this as s-uninstall? I was wondering why it was so important to include an uninstall option right away, and to feature it so prominently in the article. :-)
FYI, I recently did a test using a Xilinx FPGA to see how fast I could clock a signal from one register to an adjacent register (i.e. no combinatorial logic in-between the flops). The top speed on V2Pro (the latest and greatest from Xilinx -- .13 process and all that) was about 500 MHz in a -6 speed grade.
(I am going from memory here, it *may* have been as high as 600 MHz.) In any case, like the man says, CPU speeds are a long ways away. The *top* top speed you can get for any real-world logic is going to be about 400 MHz, and that only for very specialized applications.
What really happens with the sets you can buy today at WalMart is as follows:
- Daddy takes 6-year-old Robby to the store.
- Robby sees lego sets with cool themes that he
has recently seen in movies.
- Daddy, on the other hand, sees a tub of assorted bricks packaged
in a boring plastic case.
Guess which one gets purchased? (...even though daddy knows that Robby will *eventually* get more use out of the tub of bricks than he will out of the Harry Potter castle.)But you can't compare revenue generated by a FREE operating system with revenue generated by a rather costly operating system. The goals are completely different.
Last year the Pentagon awarded Raytheon an 18-year contract valued at $3 billion to produce more than 10,000 missiles for both the Navy and the Air Force.
This sounds sort of like the tale of NASA spending millions to develop a pen that works in zero-G environments. The Russian's, when confronted with the same problem, used a pencil.
It sounds strange to rely on a missile that has a tight turning radius when all you have to do is swivel the launch rails. Surely the problem of how to swivel the launch rails is easier to solve than how to make a missile turn better.
That being said, there are many other benefits to a tight-turn-radius missile.
If, as the article says, this is to be used in printers and other small embedded systems, there is still a need for serial and parallel ports. Many *many* devices out there still use these things. Same goes for the large headers which take up space above the memory connectors.
Gates directed 7,000 programmers to spend February scouring the Windows operating system for openings hackers might exploit to steal data or shut down systems.
Wow, 7000 programmers! I bet they figure out how to close the barn door.
Sure, but there is another factor that is important for high-tech companies that Gartner is missing: skilled workers.
I work for a company that has one of its offices in Edmonton, AB, Canada. Why Edmonton? Let's see:
The third point there is very important. Yes, it is possible to attract SV engineers to Edmonton, but it takes a *lot* of effort and incentive. One major problem is that a SV engineer may sell a bungalow for $600,000 USD. In Edmonton, you can get a mansion for $200,000 USD; there are no houses to be bought for $600K. So the SV engineer suddenly has $400K to pay taxes on.
So, the vast majority of people in R&D are local-area Canadians.
But they are really in two different market segments. Yes, they are both embedded, but the main focus of the high-end embedded MIPS chips is performance without the high power consumption.
ARM has excellent performance for some applications, but for many embedded systems, using an ARM is simply not an option.
Speaking as someone who does digital design: I would *never* overclock a chip on a system that I wanted to be reliable unless I knew that the manufacturer was deliberately marketing their chips at a lower speed than they were capable of. There are just too many ways that this can bite you.
The main problem is that you just don't know when you have gone over the line. Overclocking might be suitable in most cases except that one critical path which doesn't get executed very much.
That being said, for getting the latest gaming system, overclock to your heart's content. Who cares if the game crashes once in a while?
The labels see signs of a similar death spiral in the United States. Sales of CD singles are off 41 percent, compared with the same time last year, and album sales are effectively flat -- up less than 1 percent from a year ago, according to SoundScan, a market research firm that tracks retail music sales.
Hello!??? Economy just *might* have something to do with this.
In Germany alone, one survey by market researcher GfK found that blank CD sales jumped 129 percent this year. Purchases of pre-recorded music dropped 2.2 percent in the same period.
Indeed, blank CDs now outsell recorded discs in Europe and Canada, according to one label executive.
So, blank CD purchases are up by over 100%, and they outsell recorded CDs, yet recorded music sales have only dropped by 2.2%.
Gosh! Could it be that people by blank CDs for purposes other than pirating music?
Nahh. We should pay extra for each blank CD we buy, because we need to reimburse the music industry for pirated music.
"Second, we know we need to continue to focus in on our relationship with our customers. This is an area where we need to be ever vigilant. Certainly, as Bill talked about, we have opportunities for improvement in security, in virus protection, in the way we license and sell our products, and the reminders on that are always in front of us." --Steve Ballmer
I think this pretty much sums up a lot of what is wrong with Microsoft:
1) Security
2) The way they license and sell products.
At least they are realizing that market opinion is starting to go against them, and are trying to change this. I don't love Microsoft, but if they started to change their licensing tactics, I would be more inclined to buy their products.
Not a chance. You *can* do what you want in high-tech, and have fun doing it.
And... it's a *lot* better than school. You actually get paid decent money. There are less deadlines, less overtime hours, and less stress. I don't miss school one bit.
But then, I was never the "career" student.
But in general, start doing something you like doing, even if it pays less. If you like doing it, chances are you will be good at it. If you are good at it, chances are you will be noticed. If you are noticed, chances are your pay will go up too.
FibreChannel sucks for doing TCP/IP, but is really quite good at SCSI (called FCP).
But in terms of doing optical *on* the motherboard (or directly to the processor), this is a very long way off. Even if we could find a cheap way of converting electrical signals to optical signals in a processor chip or ASIC, the issue of manufacturing the motherboard PCB with optical traces is a big one.
The other thing to note here is that we would likely not see optical busses. Rather, a single multi-gigabit serial channel would be more cost-effective. But copper also has a lot of room to grow in this regard...