You don't honestly think the clone market was doing anything other than cannibalizing the Mac market, do you? Most post-mortem analyses of the Mac clone market I've seen point to less than 10% of clones being purchased by people who had never owned a Mac before. Of course, it wasn't long after Apple killed the clone market, including Motorola's own line of clones, that Motorola began phasing out Macs in use in their own business.
If you read Motorola's website, it's quite clear that they don't care about the desktop market at all. All their documentation and marketing for PowerPC processors (which is all tucked away from their main products) advocates their use in embedded systems. You'll find nary a mention of Apple as a successful customer. The fact is that Motorla doesn't give a damn about Apple.
As much as I like the technical merits of PowerPC processor family, I've been pushed over the past year into the firm belief that Apple should once again make the move to a different processor family or to begin designing a strategy that can continue to fund OS development without depending on hardware sales. I'm convinced that the failed AIM alliance is nothing but an albatross around Apple's neck. Maybe if Exponential's technology hadn't have been inferior to the PPC 750, we'd have a company dedicated to the Mac for survival right now that could keep Apple going.
In the end, the G4 fiasco was a sign that Apple has tied its destiny to the whims of two companies who are content to let them die. Motorola will fiddle while Cupertino burns, and that great titan IBM, will simply shrug.
(Is that enough pointless and melodramatic literary references for you all? <g>)
Ah, apparently I just haven't been looking hard enough. Thanks for pointing that out. It's a bit more than I can justify spending right now, but I'm glad to know that something like it is finally available.
None of Aqua is based on Adobe's software anymore. The primary motivation for moving from Display PS to Aqua's PDF-based system was to completely avoid licensing fees from Adobe. These fees were one reason that NeXTSTEP and OPENSTEP were so expensive. PDF, however, is an open format.
Also, rewriting a GUI for Darwin would be pointless, since the Aqua GUI and the Carbon & Cocoa APIs are foundation of every reason to buy Mac OS X. They define the user experience. Using Darwin on it's own just gets you a version of BSD/Mach with a weird filesystem layout and native support for HFS & HFS+ volumes.
Also, the "theming engine" referred to is simply the Appearance Manager. It was originally intended to support multiple Mac OS interfaces, but Jobs ironically decided that allowing users to drastically change the interface would be confusing. (Poor, poor Kaleidoscope users. They must be so confused.) This product is simply a tool to create native Mac OS 8.1+ themes for use in the Appearance manager. I don't really understand what Apple's problem is with it.
Otherwise, your points are all valid, especially the point about jumping up and down and shouting at Apple.
From the blurb up top:
Apparently the chemical is used to try to stop weight loss...
It doesn't help weight loss. It helps remdy/prevent weight loss. The receptors and the naturally produced chemicals in the brain that trigger them were evolved to encourage someone who had been starving to eat extra amounts of food to try to build back what had been damaged during starvation.
Perhaps I jumped down your throat unfairly. It seemed as if you were trying to say that it was silly to link marijuana to weight loss, as if they were saying that those chemicals were having the opposite effect. In retrospect, I'm not sure exactly what it was you were saying, but it sounded like you had things backwards and were being a smartass because of it. Sorry if I jumped down your throat without reason. I get a little irritated sometimes about the way that about 90% of the comment in any Science article are by people who didn't even bother to read the article.
4: Or maybe how they claim PowerPC processors kick the crap out of x86 processors. Remember how they used to say the G4/450 was double the speed of a Pentium III/450? Somebody explain to me how a G4/733 (with an extended pipeline as the P4 has) can be double the speed of Pentium 4 1.5 ghz if that was the case. Do the math people....
Well, you're kind of off-base about the pipeline.
MPC7400 -- 4-stage pipeline (Fetch, decode, execute, and writeback)
MPC7410 -- 4-stage pipeline (Ditto...)
MPC7450 -- 7-stage pipeline (Didn't find the names)
All this information can be found on Motorola's website in their technical specs for the processors. I'm very unhappy about them moving to a 7-stage pipeline, since that small, simple pipeline has been a key to competing with x86 processors. However, to say that they are just as over-extended as the Pentium IV is ridiculous. The Pentium IV has a 20-stage pipeline! That can be a 20-cycle gap in execution when you have a branch mispredicition. Sure, it can dispatch 6 instructions at once, compared to the MPC74XX's 3 at once, but when 50% of your instructions are memory-bound load/store operations, it doesn't really help that much in the average case. This why even the Athlon doesn't bother with more than 3 even with its 15-stage pipeline.
Granted, Apple is exaggerating by saying that just because certain operations (all SIMD-heavy Photoshop filter) complete at half the time of a top-of-the-line Pentium system, that the PowerMac G4 is always going to be twice as fast. However, the fact remains that for those operations, a chip clocked at half the speed does perform twice as fast. It does go to show that MHz as a rating of performance is just an imaginary figure used to blow magical marketing smoke. Apple just decided to Blow Smoke Different.
3: Remember the whole G4 fiasco? I wonder how many people actually bought one of those 350 mhz G4 processors....
Um, no, I don't remember it. What are you talking about?
I'll take this one.
I remember this clearly because I was looking to purchasing my PowerMac G4 around that time. Originally, the first PowerMac G4s were going to be shipped with 400, 450, and 500 MHz processors. However, the thrice-damned Motorola was well into the swing of giving Moore's Law the finger, and they couldn't produce enough 500 MHz processors to meet the demand. Apple made a move that many, including myself, considered ill-advised at the time. Since they couldn't sell 500 MHz PMG4's, they retroactively adjusted the entire line -50 MHz for the same price. 400 MHz machines became 350 MHz machines for the exact same price. They simply waved their hand over all orders and changed them. I think I remember them giving special condolance offers to people who already had orders in the system, but everyone afterwards had to pay the same for less. It was a total rip-off.
I ended up with a 400 MHz (middle-of-the-line) machine, with the intention to upgrade it with a dual-processor card (which has yet to materialize), but I feel a little upset in that I could've gotten that processor for cheaper before the change.
What was the logic behind creating a whitespace-based syntax rule? And why do you feel it is good, please refrain from the readability answer because that is all I get from those people I know who know Python.
I fail to see why there would be any other reason. Furthermore, I fail to see why there should even need to be a better reason. Why do you have whitespace at all? There are only 2 real answers: easy parser writing and human readability.
Python's style makes it easy to see blocks of logic. It also forces you to think about how your code is organized by exposing these blocks to you at all times. Braces, parentheses, brackets, etc. are easy to lose track of in complex single-line statements. You have to spend too much time thinking about whether or not you've got your puncuation matched up properly. Python eliminates this confusion by exposing logical blocks. Besides, properly formatted and readable C code should already be spaced out like a Python program. Python just eliminates the redundant punctuation.
Typically, the people who complain the loudest about enforcing spacing in syntax are the same people who write those tangled, dense, single-line statements in C and Perl that inspired their respected obfuscated code contests. You don't need the ability to cram 5 lines of Python in 1 line of Perl. It just hurts maintainability, and there's really no compelling argument for keeping source code dense and compact anymore if it doesn't add speed and remove bloat. (Forgive me if I have unfairly tarred you with this brush, but this has been my general experience.)
(In response to another post:)
Also, I've never seen a source-control system mess with the spacing of a file before. That's just odd. Be consistent with using either spaces OR tabs and your Python code will be much easier to store. I'm not saying it doesn't happen. I'm just saying that bugs in certain tools that weren't written with Python in mind shouldn't be a black mark against the whole language.
I can't believe you used Motif and X Windows as examples of good procedural APIs. They're horrible to program in! If you want a good, clean procedural GUI API, look at GTK.
However, give me Java anyday for writing GUI apps. The well-constructed OO model and exception handling in Java's GUI framework is simply a thing of joyous beauty. It makes writing event-driven windowed apps a joy. Motif and X Windows (especially when you start worrying about internationalization) are not anywhere nearly as well designed. Sure, they're old, large, and widely used. That doesn't mean that they work well or that they can be considered "mature." Both are examples of hacked together standards attempting to please all the people who had implemented proprietary solutions before. ICCCM alone makes X Windows a hideous beast. There is simply no comparison.
Wow. I'm impressed! Not only did you manage to come up with an intelligent reply without reading the linked story, you did it without even fully reading the Slashdot summary! Thumbs up to you, man!
It's not that much. I believe it's less than a dollar, so it's probably already figured into that "roughly a dollar" estimate I read somewhere during the whole licensing flak. Plus, most of the most energetic supporters of Firewire, like Sony, get the license fee waved anyway. It's really a nonissue.
The real price issue is that the device controllers for Firewire are a bit more complex than USB device controllers because the protocol is more advanced and because the devices should be able to act on their own on the network. Also, a device can talk to multiple devices on a Firewire network, not just the main computer. For your computer or a video camera, it's not that big of a deal. For a hard drive, though, it requires a bit of additional logic to handle transfers which are abnormal for a hard disk. I mean, how many hard drives talk to multiple machines normally without some mediating agent handling the communication?
This is where the costs add up. USB 2.0 saves a good bit of that by making the CPU do a good bit of the work and by not allowing device-to-device communications without the CPU involved. USB 2.0 is like a Winmodem. It's cheap and gets the job done, but it taxes resources better spent doing other work.
Darn. I always had hoped it was true. It's just too good of a story. (I'm a former S2 Systems employee, myself, from back before the Stratus merger with Ascend communications.)
'd like to point out that our systems are already fairly dependent on the processor... seeing as how THAT'S WHAT MAKES OUR SYSTEMS WORK!
Then, are you opposed to the use of graphics cards? They allow you to farm out computation of 3D rendering to keep the load off the processor. Are you opposed to sound cards which keep sound processing off of the CPU? You must also be in favor of Winmodems, too, if this is your stance.
The fact is that the more needless tasks you take off of your CPU, the more your CPU is free to do real work. USB requires the CPU to be involved when it doesn't have to be. Firewire frees a device from dependency on the CPU. There doesn't even have to be a computer running for two Firewire devices to talk to each other. Try saying the same thing about USB.
USB works on multiple platforms, including PPC. Please explain how USB locks us into Intel, as you imply?
Well, this is a straw man argument as I never said that you get locked into Intel processors with this. USB 2.0 just makes you more dependent on a faster CPU for performance. Slower machines will perform more poorly when using high bandwidth USB 2.0 devices. When you need to take in a DV stream over the USB 2.0 port and do some video rendering on it, the contention for CPU resources makes have a faster (more expensive) CPU necessary. It's all very simple.
Oh, really. I've got 3 Firewire connectors on my motherboard -- two external 6-pin connectors and one internal 6-pin connector. They cost less than $1 to include on a motherboard, after all. Most PC manufacturers just don't do it right now, but it's getting more common. Compaq has had computers with Firewire built-in for years.
Oh, you want to get a hacked up version of Final Fantasy I? Get the translated ROM for Final Fantasy IIj. It basically reuses 90% of the graphics from FF1. The plot's a little weak, but it's far better than the first. Watch out, though, the monsters are tough as nails, and you have to spend a lot of time "leveling up" to get anywhere. The advancement system was pretty neat, though. No levels, just boosting attributes through use. They don't pull punches either. If you aren't supposed to be in an area in the game yet, the monsters there WILL kill you very quickly.
Seriously, though, try the game out, and you'll see instantly why I think it seems like a hacked remake. The graphics are just dead on.
Actually, in the class sense of the word, it is a troll. However, no one really uses troll to mean a good intellectual post meant to trick people into a flamewar. Nowdays, it just means some 14 year old posting a bunch of profanity laced racist slurs or an inane, worn-out joke over and over again, like the old grits routine and the recent revival of the Beowulf cluster nonsense.
I guess no one remembers to good old days, when newsgroup discussions were filled with people from academia, who actually put some thought into baiting people. Race baiting and name calling is easy. Making people honestly believe you think something really stupid or irritating is a lot harder. No one remembers when a troll was a thing of art instead of the infantile behavior it is today.
I mourn for the loss of the intellectual troll. They were the court jesters of the Internet. Nowdays, trolls are just a bunch of immature vandals who go about urinating all over message boards because they have no respect for the sense of community that the long-time members of that site once enjoyed.
In a way, I feel like a herald for the aliens in Independence Day. You move into a new community on the Internet, and you find a rich variety of well-meaning individuals who are interested in little more than discussing their ideas. In this small preserve of civility, you can feel free to express yourself. Then, the ravagers come and destroy the community you once loved. They don't read old discussions for topics well hashed-out. They don't obey the rules of behavior. In fact, they spurn them openly. You then have no choice but to move on, saddened by the loss of another haven on the Internet. You find a new place to enjoy, and then the cycle repeats.
I guess I've rambled long enough. It just makes me nostalgic to see a post like this. To see what intellectual provocation looks like once again. I doubt I'll see it again on Slashdot in a long time.
No kidding. It's an obvious troll, in the classic sense of a post designed to fool people into a flamewar. It's actually one of the few beautiful, well thought out trolls I've seen on Slashdot. Most people don't even bother to think something out as far as this. It's all goatse.cx this and Beowulf cluster that, with a mix of profanity, racism, and obscene ascii art thrown in.
It's good to see the rare intellectual troll every now and then. This post should've only have been marked Funny (+1) or Troll (-1), with maybe a couple of Overrated/Underrated moderation thrown in.
Oh no. Oh, Lord, no. I'm posting about moderation, something I consider the cardinal sin of offtopic kvetching. Oh, well. I'm still counting down from the Day of the Karma Cap.
The annoying thing though is that, although firewire might be better, most motherboards don't have them on-board. So USB 2.0 would be nice for those of us (ahem, all of us) that have USB ports on our computers.
While the connectors are the same, the circuitry needed to recognize USB 2.0 isn't there. Surely, you don't expect to plug an UltraATA/100 drive into an EIDE controller and expect to get full UltraATA/100 speeds out of it? This isn't just a matter of updating software. The hardware controller device has to recognize the protocol, which it won't magically do. Since the purpose of these controllers is to be cheap, don't expect them to be firmware-upgradeable either.
First of all, they don't use the same connector at all, so there's no use trying to make them compatible. Second, a system with good Firewire support is makes it just as easy to plugin a Firewire device and go as it is for a USB device.
This has nothing to do with politicking, for once, and everything to do with superior, mature technical standards. Despite all the touting of USB by Intel, MS is actually going with the better standard. USB 2.0 is not as ready to go as Firewire, which MS has had *years* to get running. I'm not surprised that MS has Firewire support first. It's been around since 1986 with the latest revision to the standard happening in 1993. USB 2.0 hadn't even been started on until Apple started pushing Firewire in their own products and it gained media attention. This is why Windows XP has support for Firewire and not USB 2.0. It has nothing to do with moneyed interests. If it did, then you can be sure it would've been Intel's standard going in instead.
This has nothing to do with USB 1.0. It's still being supported as the hot-pluggable low-bandwidth device interface it was intended to be.
This is about USB 2.0, the hacked extended version that support devices at 480 Mb/s. It was intended to be a Firewire killer, but doesn't have the same technical merits. Intel started crowing about coming up with it soon about the time Firewire started to get some momentum going. It effectively killed the momentum for Firewire for awhile. I say let USB 2.0 die. It's an inferior protocol that's intended to further slave your computer to the processor instead of farm that functionality out.
USB 1.0 isn't going anywhere. It's still good for what it was intended for.
Actually, I hope it is the death knell. The original USB standard was not intended to be extended like this. It was meant for simple, low bandwidth devices, and the protocol specs show it. While USB 2.0 defeats some of the problems, such as evenly slicing your bandwidth among devices no matter what they need, it's still a processor arbitrated bus. It's just an attempt by Intel for further system dependency on the processor at the expense of performance. It helps their bottom line to make us more dependent on their hardware at our own expense.
Plus, the FUD marketing plan that they used just when Firewire started to get some public interest turned me off instantly. "Oh, no, don't use Firewire! We'll have this standard up and running in a few years, and then Firewire will be dead because ours is slightly faster than their current standard. Never mind that speed bump Firewire has planned. Listen to our vaporware instead!"
Firewire is elegant. It doesn't suffer from the star topology bottlenecks of USB 2.0. It's not processor arbitrated and does not require and active computer to be useful. It supports isochronous transfers. It makes a good networking protocol. It is in every way superior, and Intel and, apparently, Microsoft both know it. I hope that this is the knife in USB 2.0 that kills it. Bring on the Firewire devices, I say.
I remember hearing a rumor back at my last job about a Stratus VOS machine that the same thing happened to in a bank in South America somewhere. You see, Stratus machines are fault-tolerant, and are intened to be ran for years without ever going down. A nifty feature of it is the ability to virtually partition single machines into fully seperated processing areas called modules. A module can also be on another machine, allowing you to combine multiple Status boxen into a single machine.
Well, this bank was remodelling their computer room when they found the Stratus box chugging along. They actually logged a support call with Stratus technical support to find out what the machine was. It turns out that the machine they had discovered had been module 2 of their production system all along. Apparently it had been buried in the wall several years before and had been forgotten about after some changes in staff over the years. Can you imagine the support call? "Hi, we found this extra machine of ours. What is it?" They'd still been paying maintenance on it the whole time!
I always thought the tale was just an urban legend among Stratus employees. I mean, who would be so stupid as to wall up a quarter of a million dollar machine? Apparently, this kind of thing just happens.
I'm not familiar with proper pronounciation of pinyin. How do you pronounce Wong? How is it different from Wang? Does this have something to do with the 4 pitches for pronouncing syllables?
Slashdot Mirror
You don't honestly think the clone market was doing anything other than cannibalizing the Mac market, do you? Most post-mortem analyses of the Mac clone market I've seen point to less than 10% of clones being purchased by people who had never owned a Mac before. Of course, it wasn't long after Apple killed the clone market, including Motorola's own line of clones, that Motorola began phasing out Macs in use in their own business.
If you read Motorola's website, it's quite clear that they don't care about the desktop market at all. All their documentation and marketing for PowerPC processors (which is all tucked away from their main products) advocates their use in embedded systems. You'll find nary a mention of Apple as a successful customer. The fact is that Motorla doesn't give a damn about Apple.
As much as I like the technical merits of PowerPC processor family, I've been pushed over the past year into the firm belief that Apple should once again make the move to a different processor family or to begin designing a strategy that can continue to fund OS development without depending on hardware sales. I'm convinced that the failed AIM alliance is nothing but an albatross around Apple's neck. Maybe if Exponential's technology hadn't have been inferior to the PPC 750, we'd have a company dedicated to the Mac for survival right now that could keep Apple going.
In the end, the G4 fiasco was a sign that Apple has tied its destiny to the whims of two companies who are content to let them die. Motorola will fiddle while Cupertino burns, and that great titan IBM, will simply shrug.
(Is that enough pointless and melodramatic literary references for you all? <g>)
Ah, apparently I just haven't been looking hard enough. Thanks for pointing that out. It's a bit more than I can justify spending right now, but I'm glad to know that something like it is finally available.
None of Aqua is based on Adobe's software anymore. The primary motivation for moving from Display PS to Aqua's PDF-based system was to completely avoid licensing fees from Adobe. These fees were one reason that NeXTSTEP and OPENSTEP were so expensive. PDF, however, is an open format.
Also, rewriting a GUI for Darwin would be pointless, since the Aqua GUI and the Carbon & Cocoa APIs are foundation of every reason to buy Mac OS X. They define the user experience. Using Darwin on it's own just gets you a version of BSD/Mach with a weird filesystem layout and native support for HFS & HFS+ volumes.
Also, the "theming engine" referred to is simply the Appearance Manager. It was originally intended to support multiple Mac OS interfaces, but Jobs ironically decided that allowing users to drastically change the interface would be confusing. (Poor, poor Kaleidoscope users. They must be so confused.) This product is simply a tool to create native Mac OS 8.1+ themes for use in the Appearance manager. I don't really understand what Apple's problem is with it.
Otherwise, your points are all valid, especially the point about jumping up and down and shouting at Apple.
From the blurb up top:
Apparently the chemical is used to try to stop weight loss...
It doesn't help weight loss. It helps remdy/prevent weight loss. The receptors and the naturally produced chemicals in the brain that trigger them were evolved to encourage someone who had been starving to eat extra amounts of food to try to build back what had been damaged during starvation.
Perhaps I jumped down your throat unfairly. It seemed as if you were trying to say that it was silly to link marijuana to weight loss, as if they were saying that those chemicals were having the opposite effect. In retrospect, I'm not sure exactly what it was you were saying, but it sounded like you had things backwards and were being a smartass because of it. Sorry if I jumped down your throat without reason. I get a little irritated sometimes about the way that about 90% of the comment in any Science article are by people who didn't even bother to read the article.
4: Or maybe how they claim PowerPC processors kick the crap out of x86 processors. Remember how they used to say the G4/450 was double the speed of a Pentium III/450? Somebody explain to me how a G4/733 (with an extended pipeline as the P4 has) can be double the speed of Pentium 4 1.5 ghz if that was the case. Do the math people....
Well, you're kind of off-base about the pipeline.
MPC7400 -- 4-stage pipeline (Fetch, decode, execute, and writeback)
MPC7410 -- 4-stage pipeline (Ditto...)
MPC7450 -- 7-stage pipeline (Didn't find the names)
All this information can be found on Motorola's website in their technical specs for the processors. I'm very unhappy about them moving to a 7-stage pipeline, since that small, simple pipeline has been a key to competing with x86 processors. However, to say that they are just as over-extended as the Pentium IV is ridiculous. The Pentium IV has a 20-stage pipeline! That can be a 20-cycle gap in execution when you have a branch mispredicition. Sure, it can dispatch 6 instructions at once, compared to the MPC74XX's 3 at once, but when 50% of your instructions are memory-bound load/store operations, it doesn't really help that much in the average case. This why even the Athlon doesn't bother with more than 3 even with its 15-stage pipeline.
Granted, Apple is exaggerating by saying that just because certain operations (all SIMD-heavy Photoshop filter) complete at half the time of a top-of-the-line Pentium system, that the PowerMac G4 is always going to be twice as fast. However, the fact remains that for those operations, a chip clocked at half the speed does perform twice as fast. It does go to show that MHz as a rating of performance is just an imaginary figure used to blow magical marketing smoke. Apple just decided to Blow Smoke Different.
I'll take this one.
I remember this clearly because I was looking to purchasing my PowerMac G4 around that time. Originally, the first PowerMac G4s were going to be shipped with 400, 450, and 500 MHz processors. However, the thrice-damned Motorola was well into the swing of giving Moore's Law the finger, and they couldn't produce enough 500 MHz processors to meet the demand. Apple made a move that many, including myself, considered ill-advised at the time. Since they couldn't sell 500 MHz PMG4's, they retroactively adjusted the entire line -50 MHz for the same price. 400 MHz machines became 350 MHz machines for the exact same price. They simply waved their hand over all orders and changed them. I think I remember them giving special condolance offers to people who already had orders in the system, but everyone afterwards had to pay the same for less. It was a total rip-off.
I ended up with a 400 MHz (middle-of-the-line) machine, with the intention to upgrade it with a dual-processor card (which has yet to materialize), but I feel a little upset in that I could've gotten that processor for cheaper before the change.
Anyway, that's what he was talking about.
What was the logic behind creating a whitespace-based syntax rule? And why do you feel it is good, please refrain from the readability answer because that is all I get from those people I know who know Python.
I fail to see why there would be any other reason. Furthermore, I fail to see why there should even need to be a better reason. Why do you have whitespace at all? There are only 2 real answers: easy parser writing and human readability.
Python's style makes it easy to see blocks of logic. It also forces you to think about how your code is organized by exposing these blocks to you at all times. Braces, parentheses, brackets, etc. are easy to lose track of in complex single-line statements. You have to spend too much time thinking about whether or not you've got your puncuation matched up properly. Python eliminates this confusion by exposing logical blocks. Besides, properly formatted and readable C code should already be spaced out like a Python program. Python just eliminates the redundant punctuation.
Typically, the people who complain the loudest about enforcing spacing in syntax are the same people who write those tangled, dense, single-line statements in C and Perl that inspired their respected obfuscated code contests. You don't need the ability to cram 5 lines of Python in 1 line of Perl. It just hurts maintainability, and there's really no compelling argument for keeping source code dense and compact anymore if it doesn't add speed and remove bloat. (Forgive me if I have unfairly tarred you with this brush, but this has been my general experience.)
(In response to another post:)
Also, I've never seen a source-control system mess with the spacing of a file before. That's just odd. Be consistent with using either spaces OR tabs and your Python code will be much easier to store. I'm not saying it doesn't happen. I'm just saying that bugs in certain tools that weren't written with Python in mind shouldn't be a black mark against the whole language.
I can't believe you used Motif and X Windows as examples of good procedural APIs. They're horrible to program in! If you want a good, clean procedural GUI API, look at GTK.
However, give me Java anyday for writing GUI apps. The well-constructed OO model and exception handling in Java's GUI framework is simply a thing of joyous beauty. It makes writing event-driven windowed apps a joy. Motif and X Windows (especially when you start worrying about internationalization) are not anywhere nearly as well designed. Sure, they're old, large, and widely used. That doesn't mean that they work well or that they can be considered "mature." Both are examples of hacked together standards attempting to please all the people who had implemented proprietary solutions before. ICCCM alone makes X Windows a hideous beast. There is simply no comparison.
More sarcasm:
Wow. I'm impressed! Not only did you manage to come up with an intelligent reply without reading the linked story, you did it without even fully reading the Slashdot summary! Thumbs up to you, man!
It's not that much. I believe it's less than a dollar, so it's probably already figured into that "roughly a dollar" estimate I read somewhere during the whole licensing flak. Plus, most of the most energetic supporters of Firewire, like Sony, get the license fee waved anyway. It's really a nonissue.
The real price issue is that the device controllers for Firewire are a bit more complex than USB device controllers because the protocol is more advanced and because the devices should be able to act on their own on the network. Also, a device can talk to multiple devices on a Firewire network, not just the main computer. For your computer or a video camera, it's not that big of a deal. For a hard drive, though, it requires a bit of additional logic to handle transfers which are abnormal for a hard disk. I mean, how many hard drives talk to multiple machines normally without some mediating agent handling the communication?
This is where the costs add up. USB 2.0 saves a good bit of that by making the CPU do a good bit of the work and by not allowing device-to-device communications without the CPU involved. USB 2.0 is like a Winmodem. It's cheap and gets the job done, but it taxes resources better spent doing other work.
Darn. I always had hoped it was true. It's just too good of a story. (I'm a former S2 Systems employee, myself, from back before the Stratus merger with Ascend communications.)
'd like to point out that our systems are already fairly dependent on the processor... seeing as how THAT'S WHAT MAKES OUR SYSTEMS WORK!
Then, are you opposed to the use of graphics cards? They allow you to farm out computation of 3D rendering to keep the load off the processor. Are you opposed to sound cards which keep sound processing off of the CPU? You must also be in favor of Winmodems, too, if this is your stance.
The fact is that the more needless tasks you take off of your CPU, the more your CPU is free to do real work. USB requires the CPU to be involved when it doesn't have to be. Firewire frees a device from dependency on the CPU. There doesn't even have to be a computer running for two Firewire devices to talk to each other. Try saying the same thing about USB.
USB works on multiple platforms, including PPC. Please explain how USB locks us into Intel, as you imply?
Well, this is a straw man argument as I never said that you get locked into Intel processors with this. USB 2.0 just makes you more dependent on a faster CPU for performance. Slower machines will perform more poorly when using high bandwidth USB 2.0 devices. When you need to take in a DV stream over the USB 2.0 port and do some video rendering on it, the contention for CPU resources makes have a faster (more expensive) CPU necessary. It's all very simple.
Oh, really. I've got 3 Firewire connectors on my motherboard -- two external 6-pin connectors and one internal 6-pin connector. They cost less than $1 to include on a motherboard, after all. Most PC manufacturers just don't do it right now, but it's getting more common. Compaq has had computers with Firewire built-in for years.
Oh, you want to get a hacked up version of Final Fantasy I? Get the translated ROM for Final Fantasy IIj. It basically reuses 90% of the graphics from FF1. The plot's a little weak, but it's far better than the first. Watch out, though, the monsters are tough as nails, and you have to spend a lot of time "leveling up" to get anywhere. The advancement system was pretty neat, though. No levels, just boosting attributes through use. They don't pull punches either. If you aren't supposed to be in an area in the game yet, the monsters there WILL kill you very quickly.
Seriously, though, try the game out, and you'll see instantly why I think it seems like a hacked remake. The graphics are just dead on.
Actually, in the class sense of the word, it is a troll. However, no one really uses troll to mean a good intellectual post meant to trick people into a flamewar. Nowdays, it just means some 14 year old posting a bunch of profanity laced racist slurs or an inane, worn-out joke over and over again, like the old grits routine and the recent revival of the Beowulf cluster nonsense.
I guess no one remembers to good old days, when newsgroup discussions were filled with people from academia, who actually put some thought into baiting people. Race baiting and name calling is easy. Making people honestly believe you think something really stupid or irritating is a lot harder. No one remembers when a troll was a thing of art instead of the infantile behavior it is today.
I mourn for the loss of the intellectual troll. They were the court jesters of the Internet. Nowdays, trolls are just a bunch of immature vandals who go about urinating all over message boards because they have no respect for the sense of community that the long-time members of that site once enjoyed.
In a way, I feel like a herald for the aliens in Independence Day. You move into a new community on the Internet, and you find a rich variety of well-meaning individuals who are interested in little more than discussing their ideas. In this small preserve of civility, you can feel free to express yourself. Then, the ravagers come and destroy the community you once loved. They don't read old discussions for topics well hashed-out. They don't obey the rules of behavior. In fact, they spurn them openly. You then have no choice but to move on, saddened by the loss of another haven on the Internet. You find a new place to enjoy, and then the cycle repeats.
I guess I've rambled long enough. It just makes me nostalgic to see a post like this. To see what intellectual provocation looks like once again. I doubt I'll see it again on Slashdot in a long time.
No kidding. It's an obvious troll, in the classic sense of a post designed to fool people into a flamewar. It's actually one of the few beautiful, well thought out trolls I've seen on Slashdot. Most people don't even bother to think something out as far as this. It's all goatse.cx this and Beowulf cluster that, with a mix of profanity, racism, and obscene ascii art thrown in.
It's good to see the rare intellectual troll every now and then. This post should've only have been marked Funny (+1) or Troll (-1), with maybe a couple of Overrated/Underrated moderation thrown in.
Oh no. Oh, Lord, no. I'm posting about moderation, something I consider the cardinal sin of offtopic kvetching. Oh, well. I'm still counting down from the Day of the Karma Cap.
Good point. I had forgotten about that. I wonder if the Slashdot editors knew that or not.
Otherwise, this article could qualify as a troll to stir up USB 2.0/Firewire zealots, like myself. <g>
Just being pedantic.
The annoying thing though is that, although firewire might be better, most motherboards don't have them on-board. So USB 2.0 would be nice for those of us (ahem, all of us) that have USB ports on our computers.
While the connectors are the same, the circuitry needed to recognize USB 2.0 isn't there. Surely, you don't expect to plug an UltraATA/100 drive into an EIDE controller and expect to get full UltraATA/100 speeds out of it? This isn't just a matter of updating software. The hardware controller device has to recognize the protocol, which it won't magically do. Since the purpose of these controllers is to be cheap, don't expect them to be firmware-upgradeable either.
First of all, they don't use the same connector at all, so there's no use trying to make them compatible. Second, a system with good Firewire support is makes it just as easy to plugin a Firewire device and go as it is for a USB device.
This has nothing to do with politicking, for once, and everything to do with superior, mature technical standards. Despite all the touting of USB by Intel, MS is actually going with the better standard. USB 2.0 is not as ready to go as Firewire, which MS has had *years* to get running. I'm not surprised that MS has Firewire support first. It's been around since 1986 with the latest revision to the standard happening in 1993. USB 2.0 hadn't even been started on until Apple started pushing Firewire in their own products and it gained media attention. This is why Windows XP has support for Firewire and not USB 2.0. It has nothing to do with moneyed interests. If it did, then you can be sure it would've been Intel's standard going in instead.
This has nothing to do with USB 1.0. It's still being supported as the hot-pluggable low-bandwidth device interface it was intended to be.
This is about USB 2.0, the hacked extended version that support devices at 480 Mb/s. It was intended to be a Firewire killer, but doesn't have the same technical merits. Intel started crowing about coming up with it soon about the time Firewire started to get some momentum going. It effectively killed the momentum for Firewire for awhile. I say let USB 2.0 die. It's an inferior protocol that's intended to further slave your computer to the processor instead of farm that functionality out.
USB 1.0 isn't going anywhere. It's still good for what it was intended for.
Actually, I hope it is the death knell. The original USB standard was not intended to be extended like this. It was meant for simple, low bandwidth devices, and the protocol specs show it. While USB 2.0 defeats some of the problems, such as evenly slicing your bandwidth among devices no matter what they need, it's still a processor arbitrated bus. It's just an attempt by Intel for further system dependency on the processor at the expense of performance. It helps their bottom line to make us more dependent on their hardware at our own expense.
Plus, the FUD marketing plan that they used just when Firewire started to get some public interest turned me off instantly. "Oh, no, don't use Firewire! We'll have this standard up and running in a few years, and then Firewire will be dead because ours is slightly faster than their current standard. Never mind that speed bump Firewire has planned. Listen to our vaporware instead!"
Firewire is elegant. It doesn't suffer from the star topology bottlenecks of USB 2.0. It's not processor arbitrated and does not require and active computer to be useful. It supports isochronous transfers. It makes a good networking protocol. It is in every way superior, and Intel and, apparently, Microsoft both know it. I hope that this is the knife in USB 2.0 that kills it. Bring on the Firewire devices, I say.
Ah, thanks for the tip and the link.
I remember hearing a rumor back at my last job about a Stratus VOS machine that the same thing happened to in a bank in South America somewhere. You see, Stratus machines are fault-tolerant, and are intened to be ran for years without ever going down. A nifty feature of it is the ability to virtually partition single machines into fully seperated processing areas called modules. A module can also be on another machine, allowing you to combine multiple Status boxen into a single machine.
Well, this bank was remodelling their computer room when they found the Stratus box chugging along. They actually logged a support call with Stratus technical support to find out what the machine was. It turns out that the machine they had discovered had been module 2 of their production system all along. Apparently it had been buried in the wall several years before and had been forgotten about after some changes in staff over the years. Can you imagine the support call? "Hi, we found this extra machine of ours. What is it?" They'd still been paying maintenance on it the whole time!
I always thought the tale was just an urban legend among Stratus employees. I mean, who would be so stupid as to wall up a quarter of a million dollar machine? Apparently, this kind of thing just happens.
I'm not familiar with proper pronounciation of pinyin. How do you pronounce Wong? How is it different from Wang? Does this have something to do with the 4 pitches for pronouncing syllables?