If they end up being somewhat cheaper or make up for it in other ways, that may not be a problem. After all, replacing laptops and screens every two years isn't such a big deal given how fast the electronics around them evolve anyway.
His main point seems to be that widespread adoption of nuclear power means proliferation of both technology and weapons-grade nuclear materials, and that is a serious problem. He also points out that that is precisely not what you want in a world in which large numbers of nations are threatened by loss of habitable area and natural catastrophes.
To that one might add that there is still no solution in sight for disposing of the kind of nuclear waste even more widespread use of nuclear power would generate.
I see a simple economic solution to all of this: raise oil prices to the point where the market itself will figure out how to save enough energy to make it all work out. Such an increase can happen orderly, steadily, and predictably over, say, the next ten years, so that everybody knows where we are going. Nations can force it to happen through taxes to raise the market prices for oil to a predictable target prices. The extra revenue can be used to help business develop new energy-efficient technologies and to convert.
So, conservation and free market mechanisms rather than nuclear power looks like the real answer.
Yes, they are. That's what constitutes a "niche market".
Use of PHP or ASP is limited to sites where some consultant is paid and doesn't give a crap about maintainability.
Quite right: they just get the job done cheaply right now and rewrite in a couple of years. That's because they know that technology and tastes are changing too fast anyway to invest a lot into a long-term project.
But PHP and ASP can be used for developing long-term maintainable code (and some big sites use it), while it is just as easy in Java as it is in PHP to generate an unmaintainable mess.
I'm not sure if this is within the power of the commission that did it, for whatever benefit. This kind of power creep is exactly the kind of thing citizens should oppose.
I don't see what there is to "oppose". The CPUC purpose is to regulate the industry for the benefit of the consumer, and that's what they are doing. If they didn't already have this power, we should elect represenatives that would give this kind of power to them.
I suppose, it scares some people that parts of government might actually use the powers entrusted in them to do what they are supposed to. They have gotten so used to the idea that voters are irrelevant and power can just be brokered among elected officials. Well, that's just too bad. Maybe we should not re-elect those folks.
And Sun still hopes its Java software will be the programming foundation of choice, spanning Windows, Linux and Solaris.
Yeah, and I suppose Amiga still hopes to recapture the PC market some day. Sun lost their opportunity to make Java a big player when they failed to open up the platform and hand over control to a standards body.
Today, Java is a server-side product now and also hangs on in a few other niche markets. And even on the server, Java is a specialty product, with most people using tools like PHP or ASP.
And even if he had had a solid customer rather than merely a promise, for an independent business, developing a single product for a single paying customer is no way to run a business.
People like him give OSS business models a bad name and give companies like Microsoft ammunition against OSS. Lots of proprietary software companies fail in the same way, of course, but failed OSS businesses often blame lack of community support or lack of sponsorship for their failure, even though they actually just made the same business mistakes any of the failed proprietary businesses made.
There is nothing childish about Squeak or Logo. Squeak is a complex, high-powered Smalltalk development environment. I'd consider it too complex for beginning programmers, although I gather some people are using it for teaching introductory courses. And despite its innocent appearance, Logo is a powerful programming language.
But why not pick some language she might actually use for something? PHP or JavaScript might be a good choice. Or the Macromedia Flash scripting language--that way, she could make animations.
If she really wants to learn it as an intellectual exercise, I'd just stick with Scheme for her--there are good learning environments and tons of materials for learning programming with Scheme.
It's kind of a misnomer to call the AMD64 a fully 64-bit environment -- address space versus word size blah blah blah.
I'm sorry, but in what sense is it a "misnomer"? The Opteron has native 64bit ALUs and arithmetic instructions. Physical and virtual address space limits are not a factor in determining whether something is an n-bit processor. In any case, the Opteron's physical and virtual address space limits are more than enough for the next few years, and they can be raised when the need arises. And in terms of price/performance ratio, it is at least as good as the PPC.
Or you could forget about being a pawn in a marketing war and just buy the hardware that best fits your needs.
"Pawn" implies absence of control. But I do have control over this: buying a Macintosh helps Apple and Apple marketing. Our buying decisions do decide who gets to win or lose in the market, and every dollar we spend counts.
Furthermore, the "marketing war" is one-sided: Linux is not usually marketed against OS X, while Apple heavily markets OS X as an alternative to Linux and tries to win Linux users over to OS X; in fact, our Apple sales rep even told us that we should switch because "OS X is an improved version of Linux".
In any case, I think the best 64bit hardware is Opteron anyway, so it doesn't matter.
Sorry, but that's not how OSS development gets funded; you can't just put up some software on a web site and wait for donations.
Grsecurity looks like something you might be able to fund as part of a security consulting business. Or, if dealing with people is not your thing, you might be able to make a living writing books about security and how to use grsecurity. Or you might be able to do it on the side while working for a large company.
If grsecurity is as useful as you think, if there was a lively community around it, and if the code is usable, there is a good chance someone else will pick it up and actually build a successful business around it. If nobody continues development of grsecurity at this point, then it wasn't really a good, live open source project anyway--it was just some useful code released under the GPL.
Please don't complain about it: while your desire to create open source software is admirable, it is still your problem if you fail because you picked a naive business model.
Incidentally, it seems like iRobot R&D is still largely government funded, so they can probably sell Roombas at cost. Not exactly fair competition with Electrolux...
Also, iRobot apparently stopped making mobile robots for research.
but as for iRobot, the majority of the research robots in our lab (aside from the ones we built) were made by iRobot.
Sure, but how do you think iRobot got started? Decades of DARPA grants to MIT alone, paying for both education and research results, together with immense amounts of technology and research iRobot incorporates from other robotics labs around the world. iRobot's chairman and CTO is presumably still getting research grants and an MIT salary.
Neither iRobot nor Electrolux could exist if it weren't for massive public investment. Those companies are skimming the cream off the top of lots of public funding. There is absolutely nothing wrong with that--it gets the technology out into the real world, where it can be used. But it is still crucial to acknowledge how the research happened in the first place, because companies need to realize that they need to pay taxes so that the next generation of technologies for the next generation of products can be publicly funded.
We are the first linux distribution to offer a 64-bit top-to-bottom solution which is not a toy environment.
Well, either he is saying that 64bit Linux on the Opteron is not a top-to-bottom solution, or that it is a toy environment. I frankly don't care whether it is "top-to-bottom" or "side-to-side", but 64bit Linux on the Opteron seems to work well. I mean, IBM is using 64bit Linux as the basis of their eServer 325, a machine with 2116 Opterons in it.
Also, what PPC hardware am I supposed to run this on? If I buy Apple hardware, Apple is just going to use those sales to claim more OS X installations and to try to use that as a marketing weapon against Linux. And IBM PPC hardware is pretty expensive.
To an extent, any individual robot model needs significant research/engineering into properly getting SLAM (simulataneous localization and mapping) working.
Look, you made a statement justifying the high cost of the device because of the "amount of research that localization and mapping has had over the years". Whether you intended it or not, that implies research by others. I just pointed out that Electrolux probably didn't contribute much to the funding of that research. The fact that Electrolux may have invested significant amounts of R&D specifically in this device may be a valid argument for justifying the price, but it isn't the argument you gave.
It's important to be careful and mindful of where funding comes from because public research funding is under attack. Products like the Electrolux and the Roomba wouldn't exist if it weren't for extensive public funding and for a large amount of public domain knowledge about mobile robots.
No matter how much money Electrolux may have invested in this thing, still almost all the technology comes from elsewhere. Furthermore, is there any evidence that Electrolux has given back anything to the community? Publications? Algorithms? New methods?
Just last week for example I was hard at work running some of our lab bots
Yes, and if you want to continue to be able that kind of research, you should be clear in your statements about how research is funded. Whether or not your specific project is publicly funded (and there is a good chance that it is), it wouldn't be happening if it weren't for large amounts of public funding or if we went over to an all-patent funded research model.
Its kind of high priced, but not too bad considering the amount of research that localization and mapping has had over the years,
I doubt Electrolux sponsored any of that research or is paying any significant licensing fees, so that can't be the reason.
The Roomba gets entangled in wires, strings, cables, and tassles, and it doesn't work well on soft carpeting. I think that makes it useless in most homes. Electrolux claims not to have those problems; if it does, it may be worth the price relative to the Roomba (provided that you need a robotic vacuum cleaner at all).
Slashdot Mirror
If they end up being somewhat cheaper or make up for it in other ways, that may not be a problem. After all, replacing laptops and screens every two years isn't such a big deal given how fast the electronics around them evolve anyway.
His main point seems to be that widespread adoption of nuclear power means proliferation of both technology and weapons-grade nuclear materials, and that is a serious problem. He also points out that that is precisely not what you want in a world in which large numbers of nations are threatened by loss of habitable area and natural catastrophes.
To that one might add that there is still no solution in sight for disposing of the kind of nuclear waste even more widespread use of nuclear power would generate.
I see a simple economic solution to all of this: raise oil prices to the point where the market itself will figure out how to save enough energy to make it all work out. Such an increase can happen orderly, steadily, and predictably over, say, the next ten years, so that everybody knows where we are going. Nations can force it to happen through taxes to raise the market prices for oil to a predictable target prices. The extra revenue can be used to help business develop new energy-efficient technologies and to convert.
So, conservation and free market mechanisms rather than nuclear power looks like the real answer.
The big sites are using Java.
Yes, they are. That's what constitutes a "niche market".
Use of PHP or ASP is limited to sites where some consultant is paid and doesn't give a crap about maintainability.
Quite right: they just get the job done cheaply right now and rewrite in a couple of years. That's because they know that technology and tastes are changing too fast anyway to invest a lot into a long-term project.
But PHP and ASP can be used for developing long-term maintainable code (and some big sites use it), while it is just as easy in Java as it is in PHP to generate an unmaintainable mess.
I'm not sure if this is within the power of the commission that did it, for whatever benefit. This kind of power creep is exactly the kind of thing citizens should oppose.
I don't see what there is to "oppose". The CPUC purpose is to regulate the industry for the benefit of the consumer, and that's what they are doing. If they didn't already have this power, we should elect represenatives that would give this kind of power to them.
I suppose, it scares some people that parts of government might actually use the powers entrusted in them to do what they are supposed to. They have gotten so used to the idea that voters are irrelevant and power can just be brokered among elected officials. Well, that's just too bad. Maybe we should not re-elect those folks.
And Sun still hopes its Java software will be the programming foundation of choice, spanning Windows, Linux and Solaris.
Yeah, and I suppose Amiga still hopes to recapture the PC market some day. Sun lost their opportunity to make Java a big player when they failed to open up the platform and hand over control to a standards body.
Today, Java is a server-side product now and also hangs on in a few other niche markets. And even on the server, Java is a specialty product, with most people using tools like PHP or ASP.
He is doing it now, apparently.
And even if he had had a solid customer rather than merely a promise, for an independent business, developing a single product for a single paying customer is no way to run a business.
People like him give OSS business models a bad name and give companies like Microsoft ammunition against OSS. Lots of proprietary software companies fail in the same way, of course, but failed OSS businesses often blame lack of community support or lack of sponsorship for their failure, even though they actually just made the same business mistakes any of the failed proprietary businesses made.
There is nothing childish about Squeak or Logo. Squeak is a complex, high-powered Smalltalk development environment. I'd consider it too complex for beginning programmers, although I gather some people are using it for teaching introductory courses. And despite its innocent appearance, Logo is a powerful programming language.
But why not pick some language she might actually use for something? PHP or JavaScript might be a good choice. Or the Macromedia Flash scripting language--that way, she could make animations.
If she really wants to learn it as an intellectual exercise, I'd just stick with Scheme for her--there are good learning environments and tons of materials for learning programming with Scheme.
It's kind of a misnomer to call the AMD64 a fully 64-bit environment -- address space versus word size blah blah blah.
I'm sorry, but in what sense is it a "misnomer"? The Opteron has native 64bit ALUs and arithmetic instructions. Physical and virtual address space limits are not a factor in determining whether something is an n-bit processor. In any case, the Opteron's physical and virtual address space limits are more than enough for the next few years, and they can be raised when the need arises. And in terms of price/performance ratio, it is at least as good as the PPC.
Or you could forget about being a pawn in a marketing war and just buy the hardware that best fits your needs.
"Pawn" implies absence of control. But I do have control over this: buying a Macintosh helps Apple and Apple marketing. Our buying decisions do decide who gets to win or lose in the market, and every dollar we spend counts.
Furthermore, the "marketing war" is one-sided: Linux is not usually marketed against OS X, while Apple heavily markets OS X as an alternative to Linux and tries to win Linux users over to OS X; in fact, our Apple sales rep even told us that we should switch because "OS X is an improved version of Linux".
In any case, I think the best 64bit hardware is Opteron anyway, so it doesn't matter.
Sorry, but that's not how OSS development gets funded; you can't just put up some software on a web site and wait for donations.
Grsecurity looks like something you might be able to fund as part of a security consulting business. Or, if dealing with people is not your thing, you might be able to make a living writing books about security and how to use grsecurity. Or you might be able to do it on the side while working for a large company.
If grsecurity is as useful as you think, if there was a lively community around it, and if the code is usable, there is a good chance someone else will pick it up and actually build a successful business around it. If nobody continues development of grsecurity at this point, then it wasn't really a good, live open source project anyway--it was just some useful code released under the GPL.
Please don't complain about it: while your desire to create open source software is admirable, it is still your problem if you fail because you picked a naive business model.
Incidentally, it seems like iRobot R&D is still largely government funded, so they can probably sell Roombas at cost. Not exactly fair competition with Electrolux...
Also, iRobot apparently stopped making mobile robots for research.
but as for iRobot, the majority of the research robots in our lab (aside from the ones we built) were made by iRobot.
Sure, but how do you think iRobot got started? Decades of DARPA grants to MIT alone, paying for both education and research results, together with immense amounts of technology and research iRobot incorporates from other robotics labs around the world. iRobot's chairman and CTO is presumably still getting research grants and an MIT salary.
Neither iRobot nor Electrolux could exist if it weren't for massive public investment. Those companies are skimming the cream off the top of lots of public funding. There is absolutely nothing wrong with that--it gets the technology out into the real world, where it can be used. But it is still crucial to acknowledge how the research happened in the first place, because companies need to realize that they need to pay taxes so that the next generation of technologies for the next generation of products can be publicly funded.
We are the first linux distribution to offer a 64-bit top-to-bottom solution which is not a toy environment.
Well, either he is saying that 64bit Linux on the Opteron is not a top-to-bottom solution, or that it is a toy environment. I frankly don't care whether it is "top-to-bottom" or "side-to-side", but 64bit Linux on the Opteron seems to work well. I mean, IBM is using 64bit Linux as the basis of their eServer 325, a machine with 2116 Opterons in it.
Also, what PPC hardware am I supposed to run this on? If I buy Apple hardware, Apple is just going to use those sales to claim more OS X installations and to try to use that as a marketing weapon against Linux. And IBM PPC hardware is pretty expensive.
To an extent, any individual robot model needs significant research/engineering into properly getting SLAM (simulataneous localization and mapping) working.
Look, you made a statement justifying the high cost of the device because of the "amount of research that localization and mapping has had over the years". Whether you intended it or not, that implies research by others. I just pointed out that Electrolux probably didn't contribute much to the funding of that research. The fact that Electrolux may have invested significant amounts of R&D specifically in this device may be a valid argument for justifying the price, but it isn't the argument you gave.
It's important to be careful and mindful of where funding comes from because public research funding is under attack. Products like the Electrolux and the Roomba wouldn't exist if it weren't for extensive public funding and for a large amount of public domain knowledge about mobile robots.
No matter how much money Electrolux may have invested in this thing, still almost all the technology comes from elsewhere. Furthermore, is there any evidence that Electrolux has given back anything to the community? Publications? Algorithms? New methods?
Just last week for example I was hard at work running some of our lab bots
Yes, and if you want to continue to be able that kind of research, you should be clear in your statements about how research is funded. Whether or not your specific project is publicly funded (and there is a good chance that it is), it wouldn't be happening if it weren't for large amounts of public funding or if we went over to an all-patent funded research model.
Its kind of high priced, but not too bad considering the amount of research that localization and mapping has had over the years, I doubt Electrolux sponsored any of that research or is paying any significant licensing fees, so that can't be the reason.
The Roomba gets entangled in wires, strings, cables, and tassles, and it doesn't work well on soft carpeting. I think that makes it useless in most homes. Electrolux claims not to have those problems; if it does, it may be worth the price relative to the Roomba (provided that you need a robotic vacuum cleaner at all).