No, it just means that they tell the applicant the reason, not joe-blow-random-press. Sklyarov and Katalov should have been told, if they're willing to share.
Its been cited as a key difference between the Mac system architecture and the PC system architecture - different buses for AGP bus data, processormemory, processorPCI, etc.
Okay, I hate to break the news to you, but on PCs, they have all those busses, too.
We've got four main devices to connect on current computers: CPU, AGP card, memory, and PCI bus. Here's how they all connect:
CPU connects via a CPU bus to the 'northbridge' All three current main CPUs use a 64-bit wide data path. The AMD Athlon XP uses a 133MHz (soon to be 166) double clocked, so it is effectively 266MHz (soon to be 333,) for 2.1 or 2.6GB/s of bandwidth The Pentium 4 uses a 100 or 133MHz quadruple clocked, effectively 400 or 533MHz, for 3.2 or 4.2GB/s of bandwidth The G4 uses a 100, 133, or 166MHz bus, single clocked, for a whopping.8, 1, or 1.3 GB/s of bandwidth.
Next is the connection from the northbridge to AGP: 32-bit, 66MHz, quad clocked for 4x AGP, 1 GB/s bandwidth.
Then we have the northbridge to memory bus: SDRAM is 64-bit wide. PC-133 is 133MHz, for 1GB/s, DDR (PC2100 or DDR266) double clocks that to 2.1GB/s. PC2700 (or DDR333) is 166MHz double clocked, for 2.6GB/s of bandwidth. RDRAM is only 16-bit wide, but in Pentium 4 systems, is used in pairs, for a 32-bit data path. It operates at 400 or 533MHz double clocked to 800 or 1066, for 3.2 or 4.2 GB/s of bandwidth (which makes it a perfect match for the Pentium 4.)
Then we have the PCI connection. Most AMD and Intel northbridges use a proprietary connection from the northbridge to the southbridge, which actually contains the PCI bridge. I'm going to not worry about the actual northbridge to PCI connection, and just assume that it's faster than the PCI bus. (For all current chipsets, it is.) PCI operates at either 32-bit or 64-bit, 33MHz or 66MHz. Most PCs still use 32-bit, 33MHz, for 133MB/s of bandwidth. (Gigabit Ethernet uses 125MB/s, and Ultra160 SCSI uses 160MB/s, so we see that putting those on a 32/33 PCI bus is pointless.) PC-architecture-based servers, and the Mac, have used 64-bit, 66MHz PCI for some time, which quadruples that to 533MB/s.
So, let's add it all up: PC2700 memory on the new PowerMacs has 2.6GB/s of bandwidth at its disposal. The processor, can, at most, use half of that, because it's only single clocked. The AGP bus, only 1GB/s of that. So, the two devices with direct connections to memory only use 2.3GB/s of bandwidth, leaving 266MB/s of bandwidth 'spare'. We could still run two Gigabit ethernet controllers at full tilt, and wouldn't yet saturate the memory bus. THAT is what people mean when they say that DDR is wasted on the PowerMac. The processor could be maxing its bus, and still not get data fast enough.
This new 900MHz bus is probably a 450 double clocked (like the Athlon,) or, even more likely, 225 quad clocked (like the Pentium 4.) What I haven't seen is the width of that bus. If it's only 64-bit wide (which would mean it can only get traffic one way at a time, since the processor is 64-bit), it would have an amazing 7.2GB/s of bandwidth at its disposal. If it's 128-bit wide (all current processors have a bus twice as wide as their internal data path) that doubles to an incredible 14.4GB/s of bandwidth.
In fact, it's odd that IBM's own press release says it will only have 6.4GB/s of bandwidth. 6.4GB/s equates to two channels of PC3200 (DDR400 SDRAM) or four channels of PC800 RDRAM. Maybe they just mean that their memory controller is going to be dual channel PC3200...
They'd only get in trouble if the outright lied specifically to either hurt another company, or to help their own company.. (For example, if a/. story said that OSDN had developed a new 128-bit microprocessor that will run at 4GHz, outperform everything out there by a factor of 10, and cost $50 in quantity. THAT would get them in trouble with the SEC.)
oh-by-the-way here's our specs for ATA/66 and USB 2.0 (for which the detailed specs hadn't been finalized, and which didn't start hitting mainstream systems until some 2 years later).
Disclaimer: I used to work for Intel's server motherboard division. I don't think I'm biased, but wanted to get that out of the way.
USB 2.0 still isn't in 'mainstream' systems. I'd give it another 6 months.
It has been more than two years since FireWire came out. The first FireWire Mac was the 'Blue and White' G3 in January 1999, and FireWire cards were an option even before that.
Intel is a member of the IEEE1394 working group, and early in FireWire's life, Intel supported it, only to distance themselves when USB 2.0 was announced.
Intel has Intel-branded motherboards with FireWire onboard.
ATA/66 has nothing to do with FireWire or USB at all... Intel doesn't even dictate ATA standards, although I'm sure they have a lot of clout. (Heck, Maxtor got their 'FastDrives', a.k.a. ATA/133 accepted by the ATA standards board, against Intel's objections...)
Okay, actually read the stories. "According to industry sources..." is what it says. Nowhere is there confirmation from Apple or IBM that Apple has comitted to purchasing them. This is not new, this is just the same news as the last story, only centered on one specific rumor, instead of the main story.
As soon as Apple or IBM officially states that Apple has committed to purchasing these processors, don't title the story 'Apple is Buyer...' since we still aren't sure.
Yeah, I'll admit, I've been expecting it since IBM announced the chip, and I fully expect that Apple will be the main customer. BUT, my belief (or the belief of any 'industry source', without hard proof) doesn't make it a fact.
I'm not asking that you not to rumormonger on it, I'm just asking that it not be presented as fact when it is still just rumor.
(Bah, and now I've forfietted three of my moderator points by posting in a thread I moderated in...:-( It just got me pissed off when I finally noticed that there still isn't any proof.)
*I should note that, not unlike *NIX with XFree86, OS/2 has certain memory requirements that must be met before it'll fly. It came before its time; on a K6-2 with 128MB RAM, it flew with memory 80% free, but by then, it was too late, and NT- an even bigger ball of bloat, being at heart a reimplimentation of OS/2's C sources in C++- was reborn as Win2k about 4 years later.
Hehehe, I happily ran OS/2 2.1 on a 386/16 with 8MB of RAM, and a 200MB hard drive (IBM PS/2 of course, it was technically below the minimum specs for OS/2 even. Hell, it was originally a 286, with a 386 upgrade card in it!!!)
And how is asking to be paid for transmission rights in a similar way to how broadcast radio pays for those rights, "Silencing Internet Radio"?
It's not the asking for payment that is silencing. It's asking for exhorbitantly high payments that is silencing. (The RIAA wants Joe Blow internet broadcaster, sending a stream out of his basement, to pay more in royalties than K-Rock the multimillion dollar radio station.)
Apart from issues of image quality, the immediate feedback of digital, the lighter and faster lenses, greater DOF, and better performance at low light levels mean that you can get many shots with digital that were very hard to get with film.
One: I have yet to see a digital camera beat film in low light level situations. Yeah, the Sony F707 is incredible in low light, but a high quality ISO 1000 film is better. And where long exposure times are okay, ISO 100 film with really long exposures is many times better than the F707.
Although, I think the point of this article is that the new 11MP cameras ARE as good as film.. When the end format is digital. I won't argue that nothing beats film for a print. But if the final destination is digital (or, heck, if it needs to be digital at any point) then these new cameras are better.
Yes, if you're a professional photographer that never digitizes your film, then yes, you have no reason to go digital. But if you're a news reporter, who needs pictures to go digitally to the publisher, then you can finally stop using film. (At least, since I haven't seen the output myself, that's what I'm taking this article to be saying.)
I'm not going to argue with the subjective parts of your argument, because they are, by nature, subjective.
What I am going to dispute is the fact that you state that Firewire is faster than USB-2 (as you put it.) It is not. I won't dispute the 'better' part, but Firewire (as currently available) is 400Mbps, and USB 2.0 (a.k.a USB High Speed) is 480Mbps.
Yes, FireWire has 800Mbps and 1600Mbps in the works, but it's not available yet, as USB 2.0 is. (I have USB 2.0 on both my PC and my Mac. I'm about to get a combo USB2.0 and FireWire hard drive, so I'll let you know which interface is faster.)
Okay, I'm just stating the obvious here, but oh well:
Subscraption: n. pun. A made up word combining 'subscription' and 'scrap'. Meaning: A discontinued plan to offer something by subscription, i.e. Microsoft's software.
I really hope you don't think that was really him... Can you really see a 50+ year old guy with a doctorate saying 'skillz'? Not to mention the fact that he only got a/. account when after this story was posted. Just a dumbass impostering him.
I have to agree, this smells WAY too much like "Oh, look at me, I got canned by a big company that is in the pocket of THE MAN, because I kept saying how good this innocent little penguin OS is...." I'm not anti-Linux by any means, but this is obviously just Bruce trying to get attention. OF COURSE HP doesn't want him spouting off all the time with negative comments about one of HPs biggest partners. ANY employer will request that of their employees. The fact that he stayed at HP of his own will for so long just shows that he was waiting for HP to can him, just so he'd get this reaction.
Obviously, I don't know the 'truth' about why he is no longer there. Maybe he resigned, and is taking advantage of HPs exit system (lots of companies will 'lay you off' if you request it.) Maybe Carly Fiorina herself came to his office with ten security guards and yelled 'OUT, YOU SON OF A B*#^@!!!'. I'm guessing that it's really somewhere in between. Don't believe HPs side, don't believe Bruce's side. But just look at the response in lots of these comments 'Bruce took one for the team' is exactly what someone has said. He's probably just angling for a Linux-related job, and knows he's not going to get as much money if he leaves HP quietly as he will if he appears to have been canned for his beliefs.
Him and RMS are two of the most ego-inflated tech guys out there. (Heck, even Steve Jobs has a smaller ego than RMS....)
Heck, I had the same thing. I quit, of my own accord, because the company wasn't putting me to work the way I thought they would. (They hired me as a 'System Administrator', and I spent more than 90% of my time doing 'PC Help Desk' work.) I went to my boss, said that I didn't get hired for 'PC Help Desk', and unless they took 'PC Help Desk' out of my job description, I was leaving. They said they couldn't afford to do that, so I left. They made it a 'termination due to lack of work'. I quit, it looks like they laid me off. (Looks better on the resume, anyway. Saying 'I quit because the job sucked' isn't the best way to phrase it during an interview.)
I used to work tech support for AT&T WorldNet a number of years ago (seems like a lifetime!) and know how biased it is to be sitting on the end of technical support. If you've heard the problem a dozen times then yes, its happened quite a bit.
But if you ONLY heard it a dozen times it isn't very common. That probably works out to less and 0.001% of issues you've delt with.
Yeah, I always forget about that... Just because it seems like it's common to a tech support guy, doesn't mean it's actually common. Thanks for the reality check.:-p
Personal Telco doesn't own anything. They are a (soon-to-be) nonprofit organization that is a clearinghouse for publiclly shared connections. Connections are shared by individuals (like myself, over my broadband internet connection,) or companies (like WebCriteria, who is providing the connection referred to in this story.) Personal Telco isn't paying for any of the bandwidth, the provider is. If you're within a block of my house, you'll think you're on a Personal Telco connection, but in reality, you're sharing my $37.95/month broadband connection, through my Linksys WiFi router. If you're in Portland's Pioneer Courthouse Square, you may think you're on a Personal Telco connection, but it's really being provided by WebCriteria, Inc, who has a T1 dedicated to two Linksys routers with high gain antennas pointed down at the square from their offices. (They also have two more T1s for use by their employees, as well as a couple Apple AirPort base stations with MAC-filtered, encrypted, and VPNed connections, to cover their floors, for employee use.) So you really have WebCriteria to thank for the free T1 in the square. Other businesses provide similar free connections in other areas of town, as do individuals with DSL/Cable connections. (Yes, most DSL/Cable contracts prohibit this, but at least one DSL ISP in town promotes it.)
It still happens regularly. I used to work for Intel's server motherboard department (not even the networking department,) and about a dozen times I heard from customers who got a new motherboard whose NIC had the same MAC as one of their other boards that was in their server farm. And if it happens that often to one client company, you know there have to be more identical MACs out there that just haven't been put in the same network as their duplicate.
Just because you haven't seen it, doesn't mean it doesn't happen.
Read the story, there is a quote from a former FCC official. The FCC doesn't care any more about this than they care that my neighbor's 2.4GHz wireless phone interferes with my WiFi connection. 2.4GHz is unlicensed and unregulated. Anyone can do anything they want with it. It is completely legal for me to make a 2.4GHz jammer that would overwhelm all 2.4GHz connections within a 4-block radius, and there is nothing the FCC can/would do about it.
Hmmm, maybe I'll go take care of this problem for them. Jam both of them.
Hell, I used to work for the company that provides the free access (the donor company, WebCriteria, not Personal Telco) and I used to 'work' from that Starbucks all the time. (Computer with WiFi NIC, work phone forwarded to my cell phone, ah, what a life.) I know that a couple Starbucks employees knew I was using that connection, becuase I was asked about it by one of them!
One thing to remember is that it is NOTStarbucks that is doing this. It is T-Mobile. Starbucks has a contract with T-Mobile to provide this service. The Starbucks employees do not have any access whatsoever to the WiFi equipment. Don't blame Starbucks, blame T-Mobile. (Disclaimer: My aunt works for Starbucks, but I personally don't like them. I like small local coffee shops more.)
...rather than the usual.25 or.18 produced by Intel and other chip makers.
Uh... Intel hasn't used.25 for microprocessors since 1999, and is about to phase.18 out, now that all their major processors are on.13. (Yes, Itanium 2 is on.18 still. My point exactly.) So how are.25 and.18 'usual'? Heck, even AMD is finally going to.13! Let's see, Motorola is still at.18 with the PPC7550 (aka G4), but UMC and TSMC (the foundrys that make chips for pretty much every one else) are both at.13 now, although.18 is still their volume production.
In fact, I don't know of a single major microprocessor manufacturer that still uses a.25 micron process. Sparc is at.15, but I couldn't find what process HP-PA or MIPS use.
Hey, I bought 10.1 BEFORE July 17, so I still don't qualify. And my friend bought a New iMac a couple months ago, and his OS X Up-To-Date vouchers are still worthless. (If you read the Apple page, only purchases of Macs or boxed OS X BOUGHT AFTER JULY 17 COUNT!!)
Read my next sentence. Heck, Pentium Pro was designed to be ONLY for servers, period. And it only was. But, out of it came the Pentium 2 and Pentium 3. The 'Itanium', formerly known as Merced, was made only for servers. The 'Itanium 2', formerly known as McKinley, is made for servers and workstations. Within a couple years, we'll have a desktop IA64 chip.
Heck, I find it personally hilarious that Intel released a poster for the 386 touting "Finally! Enough horsepower to to voice recognition!" Along with a bunch of other uses that were 'finally' within reach. Almost the exact same uses were used when the Pentium 4 came out.
Slashdot Mirror
No, it just means that they tell the applicant the reason, not joe-blow-random-press. Sklyarov and Katalov should have been told, if they're willing to share.
Okay, I hate to break the news to you, but on PCs, they have all those busses, too.
We've got four main devices to connect on current computers: CPU, AGP card, memory, and PCI bus. Here's how they all connect:
CPU connects via a CPU bus to the 'northbridge' All three current main CPUs use a 64-bit wide data path.
The AMD Athlon XP uses a 133MHz (soon to be 166) double clocked, so it is effectively 266MHz (soon to be 333,) for 2.1 or 2.6GB/s of bandwidth
The Pentium 4 uses a 100 or 133MHz quadruple clocked, effectively 400 or 533MHz, for 3.2 or 4.2GB/s of bandwidth
The G4 uses a 100, 133, or 166MHz bus, single clocked, for a whopping
Next is the connection from the northbridge to AGP: 32-bit, 66MHz, quad clocked for 4x AGP, 1 GB/s bandwidth.
Then we have the northbridge to memory bus: SDRAM is 64-bit wide.
PC-133 is 133MHz, for 1GB/s, DDR (PC2100 or DDR266) double clocks that to 2.1GB/s.
PC2700 (or DDR333) is 166MHz double clocked, for 2.6GB/s of bandwidth.
RDRAM is only 16-bit wide, but in Pentium 4 systems, is used in pairs, for a 32-bit data path. It operates at 400 or 533MHz double clocked to 800 or 1066, for 3.2 or 4.2 GB/s of bandwidth (which makes it a perfect match for the Pentium 4.)
Then we have the PCI connection. Most AMD and Intel northbridges use a proprietary connection from the northbridge to the southbridge, which actually contains the PCI bridge. I'm going to not worry about the actual northbridge to PCI connection, and just assume that it's faster than the PCI bus. (For all current chipsets, it is.)
PCI operates at either 32-bit or 64-bit, 33MHz or 66MHz. Most PCs still use 32-bit, 33MHz, for 133MB/s of bandwidth. (Gigabit Ethernet uses 125MB/s, and Ultra160 SCSI uses 160MB/s, so we see that putting those on a 32/33 PCI bus is pointless.)
PC-architecture-based servers, and the Mac, have used 64-bit, 66MHz PCI for some time, which quadruples that to 533MB/s.
So, let's add it all up: PC2700 memory on the new PowerMacs has 2.6GB/s of bandwidth at its disposal. The processor, can, at most, use half of that, because it's only single clocked. The AGP bus, only 1GB/s of that. So, the two devices with direct connections to memory only use 2.3GB/s of bandwidth, leaving 266MB/s of bandwidth 'spare'. We could still run two Gigabit ethernet controllers at full tilt, and wouldn't yet saturate the memory bus. THAT is what people mean when they say that DDR is wasted on the PowerMac. The processor could be maxing its bus, and still not get data fast enough.
This new 900MHz bus is probably a 450 double clocked (like the Athlon,) or, even more likely, 225 quad clocked (like the Pentium 4.) What I haven't seen is the width of that bus. If it's only 64-bit wide (which would mean it can only get traffic one way at a time, since the processor is 64-bit), it would have an amazing 7.2GB/s of bandwidth at its disposal. If it's 128-bit wide (all current processors have a bus twice as wide as their internal data path) that doubles to an incredible 14.4GB/s of bandwidth.
In fact, it's odd that IBM's own press release says it will only have 6.4GB/s of bandwidth. 6.4GB/s equates to two channels of PC3200 (DDR400 SDRAM) or four channels of PC800 RDRAM. Maybe they just mean that their memory controller is going to be dual channel PC3200...
They'd only get in trouble if the outright lied specifically to either hurt another company, or to help their own company.. (For example, if a /. story said that OSDN had developed a new 128-bit microprocessor that will run at 4GHz, outperform everything out there by a factor of 10, and cost $50 in quantity. THAT would get them in trouble with the SEC.)
Sorry, I don't believe in posting anonymously. If someone thought my post was flame-bait, I want to take the hit, rather than hide behind anonymity.
Disclaimer: I used to work for Intel's server motherboard division. I don't think I'm biased, but wanted to get that out of the way.
Okay, actually read the stories. "According to industry sources..." is what it says. Nowhere is there confirmation from Apple or IBM that Apple has comitted to purchasing them. This is not new, this is just the same news as the last story, only centered on one specific rumor, instead of the main story.
As soon as Apple or IBM officially states that Apple has committed to purchasing these processors, don't title the story 'Apple is Buyer...' since we still aren't sure.
Yeah, I'll admit, I've been expecting it since IBM announced the chip, and I fully expect that Apple will be the main customer. BUT, my belief (or the belief of any 'industry source', without hard proof) doesn't make it a fact.
I'm not asking that you not to rumormonger on it, I'm just asking that it not be presented as fact when it is still just rumor.
(Bah, and now I've forfietted three of my moderator points by posting in a thread I moderated in... :-( It just got me pissed off when I finally noticed that there still isn't any proof.)
Hehehe, I happily ran OS/2 2.1 on a 386/16 with 8MB of RAM, and a 200MB hard drive (IBM PS/2 of course, it was technically below the minimum specs for OS/2 even. Hell, it was originally a 286, with a 386 upgrade card in it!!!)
It's not the asking for payment that is silencing. It's asking for exhorbitantly high payments that is silencing. (The RIAA wants Joe Blow internet broadcaster, sending a stream out of his basement, to pay more in royalties than K-Rock the multimillion dollar radio station.)
Um... I did a few different google searches and couldn't find anything. What search phrase did you use?
One: I have yet to see a digital camera beat film in low light level situations. Yeah, the Sony F707 is incredible in low light, but a high quality ISO 1000 film is better. And where long exposure times are okay, ISO 100 film with really long exposures is many times better than the F707.
Although, I think the point of this article is that the new 11MP cameras ARE as good as film.. When the end format is digital. I won't argue that nothing beats film for a print. But if the final destination is digital (or, heck, if it needs to be digital at any point) then these new cameras are better.
Yes, if you're a professional photographer that never digitizes your film, then yes, you have no reason to go digital. But if you're a news reporter, who needs pictures to go digitally to the publisher, then you can finally stop using film. (At least, since I haven't seen the output myself, that's what I'm taking this article to be saying.)
I'm not going to argue with the subjective parts of your argument, because they are, by nature, subjective.
What I am going to dispute is the fact that you state that Firewire is faster than USB-2 (as you put it.) It is not. I won't dispute the 'better' part, but Firewire (as currently available) is 400Mbps, and USB 2.0 (a.k.a USB High Speed) is 480Mbps.
Yes, FireWire has 800Mbps and 1600Mbps in the works, but it's not available yet, as USB 2.0 is. (I have USB 2.0 on both my PC and my Mac. I'm about to get a combo USB2.0 and FireWire hard drive, so I'll let you know which interface is faster.)
Um... They did. To quote:
Okay, I'm just stating the obvious here, but oh well:
Subscraption: n. pun. A made up word combining 'subscription' and 'scrap'. Meaning: A discontinued plan to offer something by subscription, i.e. Microsoft's software.
I really hope you don't think that was really him... Can you really see a 50+ year old guy with a doctorate saying 'skillz'? Not to mention the fact that he only got a /. account when after this story was posted. Just a dumbass impostering him.
Amen, brotha!
I have to agree, this smells WAY too much like "Oh, look at me, I got canned by a big company that is in the pocket of THE MAN, because I kept saying how good this innocent little penguin OS is...." I'm not anti-Linux by any means, but this is obviously just Bruce trying to get attention. OF COURSE HP doesn't want him spouting off all the time with negative comments about one of HPs biggest partners. ANY employer will request that of their employees. The fact that he stayed at HP of his own will for so long just shows that he was waiting for HP to can him, just so he'd get this reaction.
Obviously, I don't know the 'truth' about why he is no longer there. Maybe he resigned, and is taking advantage of HPs exit system (lots of companies will 'lay you off' if you request it.) Maybe Carly Fiorina herself came to his office with ten security guards and yelled 'OUT, YOU SON OF A B*#^@!!!'. I'm guessing that it's really somewhere in between. Don't believe HPs side, don't believe Bruce's side. But just look at the response in lots of these comments 'Bruce took one for the team' is exactly what someone has said. He's probably just angling for a Linux-related job, and knows he's not going to get as much money if he leaves HP quietly as he will if he appears to have been canned for his beliefs.
Him and RMS are two of the most ego-inflated tech guys out there. (Heck, even Steve Jobs has a smaller ego than RMS....)
Heck, I had the same thing. I quit, of my own accord, because the company wasn't putting me to work the way I thought they would. (They hired me as a 'System Administrator', and I spent more than 90% of my time doing 'PC Help Desk' work.) I went to my boss, said that I didn't get hired for 'PC Help Desk', and unless they took 'PC Help Desk' out of my job description, I was leaving. They said they couldn't afford to do that, so I left. They made it a 'termination due to lack of work'. I quit, it looks like they laid me off. (Looks better on the resume, anyway. Saying 'I quit because the job sucked' isn't the best way to phrase it during an interview.)
Uh, it was filed as 'It's funny, laugh'. Obviously, they realized it was a hoax, and put it up for slashdot readers to ridicule.
Yeah, I always forget about that... Just because it seems like it's common to a tech support guy, doesn't mean it's actually common. Thanks for the reality check. :-p
I'll summarize:
Personal Telco doesn't own anything. They are a (soon-to-be) nonprofit organization that is a clearinghouse for publiclly shared connections. Connections are shared by individuals (like myself, over my broadband internet connection,) or companies (like WebCriteria, who is providing the connection referred to in this story.) Personal Telco isn't paying for any of the bandwidth, the provider is. If you're within a block of my house, you'll think you're on a Personal Telco connection, but in reality, you're sharing my $37.95/month broadband connection, through my Linksys WiFi router. If you're in Portland's Pioneer Courthouse Square, you may think you're on a Personal Telco connection, but it's really being provided by WebCriteria, Inc, who has a T1 dedicated to two Linksys routers with high gain antennas pointed down at the square from their offices. (They also have two more T1s for use by their employees, as well as a couple Apple AirPort base stations with MAC-filtered, encrypted, and VPNed connections, to cover their floors, for employee use.) So you really have WebCriteria to thank for the free T1 in the square. Other businesses provide similar free connections in other areas of town, as do individuals with DSL/Cable connections. (Yes, most DSL/Cable contracts prohibit this, but at least one DSL ISP in town promotes it.)
It still happens regularly. I used to work for Intel's server motherboard department (not even the networking department,) and about a dozen times I heard from customers who got a new motherboard whose NIC had the same MAC as one of their other boards that was in their server farm. And if it happens that often to one client company, you know there have to be more identical MACs out there that just haven't been put in the same network as their duplicate.
Just because you haven't seen it, doesn't mean it doesn't happen.
Read the story, there is a quote from a former FCC official. The FCC doesn't care any more about this than they care that my neighbor's 2.4GHz wireless phone interferes with my WiFi connection. 2.4GHz is unlicensed and unregulated. Anyone can do anything they want with it. It is completely legal for me to make a 2.4GHz jammer that would overwhelm all 2.4GHz connections within a 4-block radius, and there is nothing the FCC can/would do about it.
Hmmm, maybe I'll go take care of this problem for them. Jam both of them.
Hell, I used to work for the company that provides the free access (the donor company, WebCriteria, not Personal Telco) and I used to 'work' from that Starbucks all the time. (Computer with WiFi NIC, work phone forwarded to my cell phone, ah, what a life.) I know that a couple Starbucks employees knew I was using that connection, becuase I was asked about it by one of them!
One thing to remember is that it is NOT Starbucks that is doing this. It is T-Mobile. Starbucks has a contract with T-Mobile to provide this service. The Starbucks employees do not have any access whatsoever to the WiFi equipment. Don't blame Starbucks, blame T-Mobile. (Disclaimer: My aunt works for Starbucks, but I personally don't like them. I like small local coffee shops more.)
Uh... Intel hasn't used .25 for microprocessors since 1999, and is about to phase .18 out, now that all their major processors are on .13. (Yes, Itanium 2 is on .18 still. My point exactly.) So how are .25 and .18 'usual'? Heck, even AMD is finally going to .13! Let's see, Motorola is still at .18 with the PPC7550 (aka G4), but UMC and TSMC (the foundrys that make chips for pretty much every one else) are both at .13 now, although .18 is still their volume production.
In fact, I don't know of a single major microprocessor manufacturer that still uses a .25 micron process. Sparc is at .15, but I couldn't find what process HP-PA or MIPS use.
Hey, I bought 10.1 BEFORE July 17, so I still don't qualify. And my friend bought a New iMac a couple months ago, and his OS X Up-To-Date vouchers are still worthless. (If you read the Apple page, only purchases of Macs or boxed OS X BOUGHT AFTER JULY 17 COUNT!!)
Read my next sentence. Heck, Pentium Pro was designed to be ONLY for servers, period. And it only was. But, out of it came the Pentium 2 and Pentium 3. The 'Itanium', formerly known as Merced, was made only for servers. The 'Itanium 2', formerly known as McKinley, is made for servers and workstations. Within a couple years, we'll have a desktop IA64 chip.
Heck, I find it personally hilarious that Intel released a poster for the 386 touting "Finally! Enough horsepower to to voice recognition!" Along with a bunch of other uses that were 'finally' within reach. Almost the exact same uses were used when the Pentium 4 came out.