The pace of development has been very rapid and people are predicting the replacement of Intel.
Sorry, you lost all credibility there. The Core is a single core with a bunch of DSPs tacked on. It's a great replacement for a general purpose PowerPC in many embedded applications, but won't touch Intel's target market any time soon. In the year and a half since that article was written we've learned how much Intel and AMD can do to keep ahead of the game and how applicable to general-purpose computing the Cell isn't.
This is a great summary of the standard Microsoft line. The reality is:
1. The cost to make the disc is completely irrelevant to the format war. Everyone's going to do both formats, and everyone's going to price them the same to make the consumers happy. In fact, it's likely that any extra profits from HD DVD being cheaper to manufacture will be used to subsidize Blu-Ray ramp-up.
2. The video codecs are irrelevant to the format war. There's nothing that's fundamentally easier or harder about decoding VC-1 or MPEG4 (which both formats require support for) whether it's coming from either hardware source. Neither has an advantage here.
3. Java vs. iHD is a point-- it means Blu-Ray is more flexible and HD DVD is easier for beginners to develop for. It will probably remain relevant for 6-12 months until the production studio apps abstract the differences away. In the meantime, you may get some special features in one format but not the other.
4. PS3 means Blu-Ray wins. Mass production brings down costs faster than anything, and despite HD DVD being released a few months earlier, Blu-Ray will probably hit mass production a year earlier, while HD DVD is still in the early-adopter phase, because of the Playstation 3. Thus, Blu-Ray is likelier to be much cheaper, much faster.
5. The thinner substrate... I'm not sure how much this matters. The whole "in the lab" argument is pretty facetious, since both formats are still 100% "in the lab" until you can actually buy them. It will be interesting to see what happens with the first-generation writable discs and how they hold up under real-world conditions.
Their supply problems are definitely one problem, but not the biggest one. The supply problems may have been understandable if the XBox 360 was selling like hotcakes, but it's not. Having mediocre sales and STILL not being able to meet demand is bad news. Of course, their supply situation is still 100% better than the PS3.
Nice summary. Actually, SWT, Swing, and AWT aren't your only choices in Java, though. There are a lot of small toolkits out there that each have some fascinating features, such as using the 3D graphics accelerator, handling zooming or animation better, etc.
The company I work for years ago branched one called "Jazz" (which has since been renamed Piccolo and mostly moved to C#: http://www.cs.umd.edu/hcil/jazz/ ) and basically developed our own Java graphics toolkit. This let us make the things that are important to use very, very fast (Yes, GUIs in Java can be extremely responsive) and give us control over our own bugs and glitches. I don't think it's unreasonable for people writing Java applications for a living to have an engineer or two reserved for just maintaining graphics issues, at which point having your own toolkit that tracks one of the major or minor ones on the market becomes feasible.
One thing that starts to become a necessity for this sort of thing, though, is the ability to embed a component of a different toolkit inside your own for compatibility with everyone else in the world. In our toolkit, embedded Swing components are impossible to get 100% right (because Sun made some of the event system "private"-- even the getters) but not hard to get 98% right. SWT is much harder.
Come on... Slashdot folks worried about a non-official kernel compile? I've recompiled the Darwin kernel on my MacOS X install before and run it off the custom compile. Yay open source, and stuff.
I think the parent is exactly right. All the people talking about "use the best tool for the job" have probably not worked very long in the industry. Being able to build on each other's successes, re-use code across projects that gets more vetted with each project, and build expertise with all the "gotchas" in your language of choice will make your company's product better and the company more profitable. These days, you can do almost anything with any of the popular languages, so there's no point in using more than one or two of them.
I'd say, standardize on Java, C#, or C++ (depending on your needs) as your primary language, add your scripting language of choice, then fire anyone who can't handle that. You'll be better off five years down the road than if you'd all built your fiefdoms around the tower of babel and every project becomes a throwaway codebase as the next "best tool" comes along 4 months later.
Any company that can't standardize on a language doesn't really have a coherent vision anyway, and probably is either a bunch of folks pretending to be a consulting firm or will disappear before too long.
(I'm sure I'm going to lose mod points for this reply, seeing what other people have written, but I don't see how any other approach is practical in the long term.)
The article discusses the Power6, and the power consumption is probably compared to the Power5, not the PowerPC. These are probably not desktop chips.
Remember, the Power5 chips had dual-core before it was "cool" to have dual-core. Everyone thought the Mac was going to be the first desktop with a dual-core processor years ago. Instead, they finally released one last year that barely keeps up with the state-of-the-art technology. And things like copper, SOI, and similar techniques were supposed to do the same for IBM's chips.
IBM chips compete very well in press releases. Running desktop code, not so well.
Yeah, the software industry has sure suffered in the 20 years since software patents were instituted. A lot of people talk about what could be without them, but one might argue that they've worked really well if the success of the software industry is any indication.
You make good points. Let me expand on my suggestion: 1. Developers desktops are NOT included in the "sandbox" area. They are controlled by IT with the same rigid standards as everyone else. Although they may get Administrator access to their Windows machines, allowed software is monitored and "personal servers" are restricted. 2. When a developer VPNs into the "sandbox" area, they can do what they want with the machines on the other side. They can try anything, upgrade the hardware, set up 10 different test servers to see what works best-- whatever. Windows RDC is great here, although X and VNC are also usable. 3. When an application goes from development to staging (or preferrably shortly before that happens), it is thoroughly vetted, the hardware set up by IT, and the software replaced, updated, patched, or otherwise brought into line with IT policies. Then QA works from staging to qualify it for release. After that it is pushed from staging to release by IT, of course, since developers aren't allowed to affect the live infrastructure.
Yes, this was exactly my point-- provide an area that the developers can use that the developers are 100% responsible for maintaining. IT provides no support. Sorry if I didn't make that clear enough. The firewall separates the areas that IT supports from the area that developers support to make sure there is no contamination of the official network.
Progress and stability are often conflicting goals. IT departments generally prefer stability, and that's why your deployments have probably been so stable and passed so many audits. Developers, of course, are charged with driving progress.
The real answer if you need flexibility with regards to "non-production stuff" is to not let IT have anything to do with it at all. Create a separate sub-net if you have to to keep the non-production machines off the IT network, and a firewall between your network and theirs to prevent any viruses, or other effects, from leaking from your net to theirs (this may require having to VPN through it just to work with these machines, c'est la vie). Keep the machines in a different room than the official server room. Maintain them all 100% yourself. Then do what you need to. Anything less and you're asking IT to aid in your development, a task they're probably not equipped to do while maintaining stability.
It's not uncommon for companies to have a "developer", "staging", and "live" system setup that are all completely independent, with some established mechanism and metrics to push products from one level to the next.
That wikipedia entry is just plain wrong. Columbus' own records described how he effectively enslaved the very first native population he met, including rape, torture, and limited food. He was so harsh to the natives that the entire population of one of the islands was completely wiped out-- one of the first well-recorded successful acts of complete genocide. When the native population dwindled, the Africans were imported to take their place.
The first african slaves on the north american content were probably also the first permanent settlers, as they were left behind when the earliest colonies failed and likely merged with some of the native populations. The first African slaves in New England were traded with the West Indies for Northeast American Indian slaves, on the thoughts that mixing them up would make it harder for them to escape.
The problem with wikipedia is that it's written by popular consent, and thus these things tend to get a little whitewashed.
On the one hand, it means the only people allowed to use the research for commercial purposes are the owners of the original patent, so universities effectively become research arms of the patent owners. On the other hand, the researched extensions to the patent are themselves unpatentable, and potentially gets the technology into the public domain sooner and cleaner. (Remember patents don't go on "forever", like copyrights effectively do under Disney's America.) Otherwise, most patents get subsequent filings to add minor features in order to extend their lifetimes.
So I see this as a step in the right direction. There's no question in my mind that patents are a necessary and good thing for any innovation-based society, but obviously there are some area of it that need refinement.
"(in not trusting the FSF to "do the right thing" with future versions of the GPL)"
Have you *read* GPLv3? I think Linus is currently the only thing standing between Stallman and a complete meltdown of the viability of open source in the economy.
No theory that invokes an all-powerful intelligent creator can be falsifiable, since the counter-argument to any experimental result that doesn't match could be that "the Creator decided to do that."
And in order to be taught in schools, especially elementary and middle schools where the students' educational foundation is insufficient to explain things to much depth, there has to be more of a test than "it's a testable theory." Darwinian evolution, and all sorts of recent augmentations to it, is a pretty well-accepted theory that explains everything from disease gaining resistance to drugs, to why new physical characteristics don't suddenly jump between species in the archeological timeline.
(Personally, I think ID is an affront to both science *and* religion, but I care more about the purity of my sciences.)
If you're one of the small handful of people who actually used a FireWire 800 drive, you might want a card for the MacBook with a FW800 port. Or, if you're more forward-looking, you might want a card that implements the 4x faster eSATA interface. The new slot is extremely fast, and is virtually the equivalent of putting another chip on the motherboard bus anyway.
The bounty, currently being over $4,000, is now twice the price of a near-fully loaded Intel iMac itself. It's almost tempting to buy one and go for the win to get the machine free. (No pun intended.)
We live in a free country. Don't be ridiculous... These things should *not* be illegal. If I build my own machine and write the OS, I shouldn't be required to make sure it boots Windows. If I build a machine that lacks support for booting anything but linux, I doubt Slashdotters would be saying it should be illegal.
IMHO, what it really depends on boils down to two things: 1. Is it worth the time to develop it for release? (Return on investment, factoring in goodwill and brand loyalty, etc.) 2. Would it be a support nightmare after release? (If you can't reproduce problems, you can't fix/mitigate them very well, and the customers may end up being more frustrated than if you'd just told them "Sorry, use Firefox".)
The summary claims the tested software was specifically designed for the new machines. I'd put a lot of money that no software outside the MacOS X kernel was "designed" for these machines. "Compiled" might be more accurate. I suspect as the tools improve and the designs really do incorporate changes that showcase the highlights of the new hardware that the benchmarks may be closer to Apple's claim. I mean, give them a break-- we're ten days into the new architecture's lifetime.
I think Mitnick made the point that he was accused of causing many millions of dollars in damages, but these (public) companies did not list such a charge on their quarterly reports. In fact, I have yet to see hacker damage appear on any quarterly report, including the more recent ones under the stricter Sarbanes-Oxley rules. So what's happening? Is this being overblown, or are companies mis-representing the damage to shareholders?
If you're arguing that Mythbusters isn't educational, you haven't watched enough episodes. Yes, they make mistakes. So do over half of all peer-reviewed scientists' papers, last I read. But it's still a very educational show, and more importantly, one that gets the watcher thinking instead of passively being entertained.
Even if the show contains a greater proportion of entertainment to education than some might like, I think it educates more than some of the old dry shows, because more people watch them. Just to use some silly math, if a show is 90% educational and is watched by 100K people, let's say it has provided 90K education-people worth of education to the world. If a show is 60% educational and watched by 1M people, it's provided 600K education-people worth of education! How's that for a Mythbusters-style estimate?
He was also one of the first vocal proponents of the separation of church and state. It's because of him that "We hold these truths to be self-evident" instead of the original text, which read "We hold these truths sacred."
When it comes to patents, the draft is actually not very aggressive about them. There is no general patent retaliation clause as in some other licenses, because the FSF believes that disallowing an offender to use any free software would not be too much of a deterrent for some.
ANY patent retaliation clause is EXTREMELY aggressive about patents. If I'm using a GPL suite for my web presence, and someone else adds code that infringes on one of my patents, in order to sue them I have to stop using the GPL'ed suite? No thanks.
This revision of the GPL looks like a shot in the foot just when open source was starting to catch on. Hopefully it will be dramatically revised and made more friendly to industry before going live. Either that, or I see a fork in every commercially viable GPL'ed project in existence.
Slashdot Mirror
Sorry, you lost all credibility there. The Core is a single core with a bunch of DSPs tacked on. It's a great replacement for a general purpose PowerPC in many embedded applications, but won't touch Intel's target market any time soon. In the year and a half since that article was written we've learned how much Intel and AMD can do to keep ahead of the game and how applicable to general-purpose computing the Cell isn't.
This is a great summary of the standard Microsoft line. The reality is:
1. The cost to make the disc is completely irrelevant to the format war. Everyone's going to do both formats, and everyone's going to price them the same to make the consumers happy. In fact, it's likely that any extra profits from HD DVD being cheaper to manufacture will be used to subsidize Blu-Ray ramp-up.
2. The video codecs are irrelevant to the format war. There's nothing that's fundamentally easier or harder about decoding VC-1 or MPEG4 (which both formats require support for) whether it's coming from either hardware source. Neither has an advantage here.
3. Java vs. iHD is a point-- it means Blu-Ray is more flexible and HD DVD is easier for beginners to develop for. It will probably remain relevant for 6-12 months until the production studio apps abstract the differences away. In the meantime, you may get some special features in one format but not the other.
4. PS3 means Blu-Ray wins. Mass production brings down costs faster than anything, and despite HD DVD being released a few months earlier, Blu-Ray will probably hit mass production a year earlier, while HD DVD is still in the early-adopter phase, because of the Playstation 3. Thus, Blu-Ray is likelier to be much cheaper, much faster.
5. The thinner substrate... I'm not sure how much this matters. The whole "in the lab" argument is pretty facetious, since both formats are still 100% "in the lab" until you can actually buy them. It will be interesting to see what happens with the first-generation writable discs and how they hold up under real-world conditions.
Their supply problems are definitely one problem, but not the biggest one. The supply problems may have been understandable if the XBox 360 was selling like hotcakes, but it's not. Having mediocre sales and STILL not being able to meet demand is bad news. Of course, their supply situation is still 100% better than the PS3.
Nice summary. Actually, SWT, Swing, and AWT aren't your only choices in Java, though. There are a lot of small toolkits out there that each have some fascinating features, such as using the 3D graphics accelerator, handling zooming or animation better, etc.
The company I work for years ago branched one called "Jazz" (which has since been renamed Piccolo and mostly moved to C#: http://www.cs.umd.edu/hcil/jazz/ ) and basically developed our own Java graphics toolkit. This let us make the things that are important to use very, very fast (Yes, GUIs in Java can be extremely responsive) and give us control over our own bugs and glitches. I don't think it's unreasonable for people writing Java applications for a living to have an engineer or two reserved for just maintaining graphics issues, at which point having your own toolkit that tracks one of the major or minor ones on the market becomes feasible.
One thing that starts to become a necessity for this sort of thing, though, is the ability to embed a component of a different toolkit inside your own for compatibility with everyone else in the world. In our toolkit, embedded Swing components are impossible to get 100% right (because Sun made some of the event system "private"-- even the getters) but not hard to get 98% right. SWT is much harder.
Come on... Slashdot folks worried about a non-official kernel compile? I've recompiled the Darwin kernel on my MacOS X install before and run it off the custom compile. Yay open source, and stuff.
I think the parent is exactly right. All the people talking about "use the best tool for the job" have probably not worked very long in the industry. Being able to build on each other's successes, re-use code across projects that gets more vetted with each project, and build expertise with all the "gotchas" in your language of choice will make your company's product better and the company more profitable. These days, you can do almost anything with any of the popular languages, so there's no point in using more than one or two of them.
I'd say, standardize on Java, C#, or C++ (depending on your needs) as your primary language, add your scripting language of choice, then fire anyone who can't handle that. You'll be better off five years down the road than if you'd all built your fiefdoms around the tower of babel and every project becomes a throwaway codebase as the next "best tool" comes along 4 months later.
Any company that can't standardize on a language doesn't really have a coherent vision anyway, and probably is either a bunch of folks pretending to be a consulting firm or will disappear before too long.
(I'm sure I'm going to lose mod points for this reply, seeing what other people have written, but I don't see how any other approach is practical in the long term.)
The article discusses the Power6, and the power consumption is probably compared to the Power5, not the PowerPC. These are probably not desktop chips.
Remember, the Power5 chips had dual-core before it was "cool" to have dual-core. Everyone thought the Mac was going to be the first desktop with a dual-core processor years ago. Instead, they finally released one last year that barely keeps up with the state-of-the-art technology. And things like copper, SOI, and similar techniques were supposed to do the same for IBM's chips.
IBM chips compete very well in press releases. Running desktop code, not so well.
Yeah, the software industry has sure suffered in the 20 years since software patents were instituted. A lot of people talk about what could be without them, but one might argue that they've worked really well if the success of the software industry is any indication.
You make good points. Let me expand on my suggestion:
1. Developers desktops are NOT included in the "sandbox" area. They are controlled by IT with the same rigid standards as everyone else. Although they may get Administrator access to their Windows machines, allowed software is monitored and "personal servers" are restricted.
2. When a developer VPNs into the "sandbox" area, they can do what they want with the machines on the other side. They can try anything, upgrade the hardware, set up 10 different test servers to see what works best-- whatever. Windows RDC is great here, although X and VNC are also usable.
3. When an application goes from development to staging (or preferrably shortly before that happens), it is thoroughly vetted, the hardware set up by IT, and the software replaced, updated, patched, or otherwise brought into line with IT policies. Then QA works from staging to qualify it for release. After that it is pushed from staging to release by IT, of course, since developers aren't allowed to affect the live infrastructure.
Yes, this was exactly my point-- provide an area that the developers can use that the developers are 100% responsible for maintaining. IT provides no support. Sorry if I didn't make that clear enough. The firewall separates the areas that IT supports from the area that developers support to make sure there is no contamination of the official network.
Progress and stability are often conflicting goals. IT departments generally prefer stability, and that's why your deployments have probably been so stable and passed so many audits. Developers, of course, are charged with driving progress.
The real answer if you need flexibility with regards to "non-production stuff" is to not let IT have anything to do with it at all. Create a separate sub-net if you have to to keep the non-production machines off the IT network, and a firewall between your network and theirs to prevent any viruses, or other effects, from leaking from your net to theirs (this may require having to VPN through it just to work with these machines, c'est la vie). Keep the machines in a different room than the official server room. Maintain them all 100% yourself. Then do what you need to. Anything less and you're asking IT to aid in your development, a task they're probably not equipped to do while maintaining stability.
It's not uncommon for companies to have a "developer", "staging", and "live" system setup that are all completely independent, with some established mechanism and metrics to push products from one level to the next.
That wikipedia entry is just plain wrong. Columbus' own records described how he effectively enslaved the very first native population he met, including rape, torture, and limited food. He was so harsh to the natives that the entire population of one of the islands was completely wiped out-- one of the first well-recorded successful acts of complete genocide. When the native population dwindled, the Africans were imported to take their place.
The first african slaves on the north american content were probably also the first permanent settlers, as they were left behind when the earliest colonies failed and likely merged with some of the native populations. The first African slaves in New England were traded with the West Indies for Northeast American Indian slaves, on the thoughts that mixing them up would make it harder for them to escape.
The problem with wikipedia is that it's written by popular consent, and thus these things tend to get a little whitewashed.
On the one hand, it means the only people allowed to use the research for commercial purposes are the owners of the original patent, so universities effectively become research arms of the patent owners. On the other hand, the researched extensions to the patent are themselves unpatentable, and potentially gets the technology into the public domain sooner and cleaner. (Remember patents don't go on "forever", like copyrights effectively do under Disney's America.) Otherwise, most patents get subsequent filings to add minor features in order to extend their lifetimes.
So I see this as a step in the right direction. There's no question in my mind that patents are a necessary and good thing for any innovation-based society, but obviously there are some area of it that need refinement.
"(in not trusting the FSF to "do the right thing" with future versions of the GPL)"
Have you *read* GPLv3? I think Linus is currently the only thing standing between Stallman and a complete meltdown of the viability of open source in the economy.
No theory that invokes an all-powerful intelligent creator can be falsifiable, since the counter-argument to any experimental result that doesn't match could be that "the Creator decided to do that."
And in order to be taught in schools, especially elementary and middle schools where the students' educational foundation is insufficient to explain things to much depth, there has to be more of a test than "it's a testable theory." Darwinian evolution, and all sorts of recent augmentations to it, is a pretty well-accepted theory that explains everything from disease gaining resistance to drugs, to why new physical characteristics don't suddenly jump between species in the archeological timeline.
(Personally, I think ID is an affront to both science *and* religion, but I care more about the purity of my sciences.)
If you're one of the small handful of people who actually used a FireWire 800 drive, you might want a card for the MacBook with a FW800 port. Or, if you're more forward-looking, you might want a card that implements the 4x faster eSATA interface. The new slot is extremely fast, and is virtually the equivalent of putting another chip on the motherboard bus anyway.
The bounty, currently being over $4,000, is now twice the price of a near-fully loaded Intel iMac itself. It's almost tempting to buy one and go for the win to get the machine free. (No pun intended.)
We live in a free country. Don't be ridiculous... These things should *not* be illegal. If I build my own machine and write the OS, I shouldn't be required to make sure it boots Windows. If I build a machine that lacks support for booting anything but linux, I doubt Slashdotters would be saying it should be illegal.
IMHO, what it really depends on boils down to two things:
1. Is it worth the time to develop it for release? (Return on investment, factoring in goodwill and brand loyalty, etc.)
2. Would it be a support nightmare after release? (If you can't reproduce problems, you can't fix/mitigate them very well, and the customers may end up being more frustrated than if you'd just told them "Sorry, use Firefox".)
Offtopic? How can a post about determining the cost of hacking be offtopic in a discussion on the cost of hacking?
The summary claims the tested software was specifically designed for the new machines. I'd put a lot of money that no software outside the MacOS X kernel was "designed" for these machines. "Compiled" might be more accurate. I suspect as the tools improve and the designs really do incorporate changes that showcase the highlights of the new hardware that the benchmarks may be closer to Apple's claim. I mean, give them a break-- we're ten days into the new architecture's lifetime.
I think Mitnick made the point that he was accused of causing many millions of dollars in damages, but these (public) companies did not list such a charge on their quarterly reports. In fact, I have yet to see hacker damage appear on any quarterly report, including the more recent ones under the stricter Sarbanes-Oxley rules. So what's happening? Is this being overblown, or are companies mis-representing the damage to shareholders?
If you're arguing that Mythbusters isn't educational, you haven't watched enough episodes. Yes, they make mistakes. So do over half of all peer-reviewed scientists' papers, last I read. But it's still a very educational show, and more importantly, one that gets the watcher thinking instead of passively being entertained.
Even if the show contains a greater proportion of entertainment to education than some might like, I think it educates more than some of the old dry shows, because more people watch them. Just to use some silly math, if a show is 90% educational and is watched by 100K people, let's say it has provided 90K education-people worth of education to the world. If a show is 60% educational and watched by 1M people, it's provided 600K education-people worth of education! How's that for a Mythbusters-style estimate?
He was also one of the first vocal proponents of the separation of church and state. It's because of him that "We hold these truths to be self-evident" instead of the original text, which read "We hold these truths sacred."
ANY patent retaliation clause is EXTREMELY aggressive about patents. If I'm using a GPL suite for my web presence, and someone else adds code that infringes on one of my patents, in order to sue them I have to stop using the GPL'ed suite? No thanks.
This revision of the GPL looks like a shot in the foot just when open source was starting to catch on. Hopefully it will be dramatically revised and made more friendly to industry before going live. Either that, or I see a fork in every commercially viable GPL'ed project in existence.