AP - The artificial lifeform known to the computing world as ALICE came to a violent and tragic end this evening. The good-natured AI was interacting with several online users when an unprovoced attack was leveled against her by a geek gang known as Slashdot. The miscreants pelted ALICE with connection requests until she finally expired. FBI and local authorities are investigating and promise to bring these geeks to justise.
ALICE is survived by a grandfather TRS-80, her mother C-64, her sister IBM RS6000, and lifelong companion Athlon. In lieu of flowers, the family asks the donations be made to the Free Software Foundation.
I will second that. I have bought about $60k worth of stuff from mwave over the past few years. Good prices. Good service. Only complaint is the delivery time from CA to the east coast, but this is not their fault.
I was thinking the same thing, in the opposite direction, I guess. The "gimmick" here is a lower price due and the different OS. But, there is no "Sims" or "Deer Hunter" for Linux, so who is going knowingly buy this. Linux-types have a high likelihood of rolling their own. Business-types (the real potential market for Linux, IMO) will not be buying from a retailer like Walmart.
The *best* this could do for the average joe is provide capabilities similar to the iOpener or other web appliances. Email, web access, write a book report. But, when it doesn't work with the new digital camera or $40 ink-jet-of-the-month printer, there will be big problems. There is a significant chance that these "penguin" systems will turn into albatrosses.
I would personally prefer to see Linux inflitrate the home-user market last after a significant portion of the non-geek population is exposed to it at work. Honestly, the home market is the area where Linux has the least to offer.
Re:Piss-on-the-perr review system
on
Wolframania
·
· Score: 1
I believe that he is both a crack pot and a genius.
I think that I agree with you on this point. I too am a computational scientist (CFD). Of course, I was most interested in what Wolfram touted as a major success of his work, a description of fluid turbulence. So, I slogged my way through introductory chapters and wondered at some of the complex patterns which could be coaxed from simple programs and initial conditions. I was hopeful.
When I read his take on CFD and fluid turbulence, I started to see that he didn't really have anything to offer me. Yes, he has shown that simple programs can capture much of the "structure" of turbulence and that it wonderful. So what? This was known 15 years ago. And, CA applications to computational fluid dynamics problems have been largely a curiousity, and that assessment may be kind.
This approach of using CA to model a physical system might be instructive to others who are basking in the light of this "new" science. Lattice gas models were tried. These were pure CAs, little particles that rattled around on discrete grids. The simulations were fast and simple to program. The problem was that the results were too noisy. A large fraction of simulation time was spend filtering the "results" to deduce physical pressures, velocities, etc. Evenually, the lattice gas models gave way to lattice Boltzmann models which wrote TRANSPORT equations for these particles to implicitly remove the noise. So the "new science" description gave way to the "old science" version.
So, as a scientist and engineer presented with a new model, I am always forced to ask, "So what?" What new capabilities does this offer me? What problems can I solve now that I couldn't before? How can I solve old problems faster or more accurately then before? Unfortunately, for me, this goes on the pile of interesting but currently useless theories. I will keep an eye on the field, but I would encourage its proponents to reach a bit further and make solid contact with tangible results. Make the theory applicable to real problems.
You wouldn't need to do a 9-hour exposure either. Just take a dozen or so digital photos from the same vantage point. Then correlate the resulting images. Retain the pixels that are the same in two or more of the frames. That should effectively delete any moving objects. Aside from expediting the process, it would also avoid the photographic problems of longtime exposure, like the lighting changing, etc.
We will shift to other resources as economic and/or political factors demand. The government can do a great deal to *encourage* the development of alternative technologies. That is a useful mechanism to help emerging technologies (i.e., fuel cells) reach a critical mass as which point they would become economically self-sustaining.
They can also encourage conservation, recycling efforts, and alternative technologies with tax credits, research grants, etc. I am not the least bit opposed to this.
The critical point is that it must be economically self sustaining. Otherwise, we will get the same "bust" that occurred with solar power in the early 80's. The technologies were propped up with tax incentives but inevitably were unable to compete in the free market. Encourage new technologies with SBIR funding or tax incentives, but let's not get stupid and *mandate* this sort of thing.
Non-biodegradeable this and that. We need to recognize something. We as a species will almost always do what is easiest, cheapest, and most directly to the purpose. One hundred years ago, most things were made out of wood, because it was readily available and it was fairly cheap to process. Fifty years ago, everything was made out of steel because it was readily available and cheap to process. Now, everything is made out of plastic because...you get the idea.
In the future, oil reserves will diminish. We will need to adjust our modes of production and our supply lines. This goes for consumer goods, for food, and for energy. It is all part of the economic optimization process. *That* is the fundamental principle that has been at work in the rise of nature and human civilization. As a working engineer, I know that it takes effort and know-how to build a reasonably efficient design process that minimize costs and maximizes operational efficiency and still conforms to the current emissions requirements. Adding further constraints to that engineering process will come at some cost in quality or availability. In an economic reality, this stuff will just cost more.
The most telling point is the institution of the "fourth R." Regulation. Yep, more regulation is the key. That will guarantee that the "environmentalist" teet doesn't dry up for they can retire. We need better engineered systems that do their utmost to optimize system efficiency and minimize costs.
Commercial. As you suggest, I usually avoid these types of limitations by dropping sshd onto 2222 instead of 22. But the current problem arises because the firewall now drops outgoing connections going to ANY ports other than 20, 21, 23, 80, 443. That's right...telnet and ftp are allowed, but ssh isn't.
I also agree that a server makes the most sense. I would amplify these recommended transport mechanisms to include a few others that will allow remote connectivity.
First is a secure IMAP server for centralized email. This will allows any SSL-enabled IMAP client to access your mailbox. Also, Squirrelmail running on an SSL web server can give your access to your centralize mail repository from any web browser.
SMB and NFS are the obvious choices for LAN-based access, but WAN access needs more care. I think that a VPN setup using
CIPE is a good approach. One the CIPE links are build, you can use most services as if you were located on your wired LAN.
The other need might be for file access from "arbitrary" locations. In addition to the normal scp and sftp apps in OpenSSH, there is a nice SCP client for windows, WinSCP. Lastly, if you have a SSL web server there already, Web-FTP will give you access to your files via https.
This sounds like a lot. In the end, you would need to expose SSH, SSL IMAP, SSL Apache, and CIPE servers. I am midway through this deployment myself, but it has stalled a bit because one of primary Internet access points started disallowing outgoing SSH.
The MS ergonomic keyboard *taught* me to type. This is no lie. I had been using computers since I was 6 or 7 years old...(Timex Sinclair, C64, etc)...and had never really learned to type. Instead, I had developed a semi-hunt-n-peck process that was not very efficient but it got me by.
Then about five years ago, I built a new computer and decided to pick up an MS Natural keyboard because I was starting to notice a little pain in my wrists after hours and hours of coding. So, I got the MS keyboard and immediately hated it. With it, your hands lay across the keyboard and obsure about 50 to 75% of the keys on the keyboard. Looking at the keyboard was now difficult, which is probably why many people dislike them. But, it did force me to stop using that visual crutch and start relying on muscle memory to find the keys.
Within a month or so, I had completely adapted and I could type 40 or 50 words per minute as a result. Now, I am reticent to use any other keyboard. I have MS Natural keyboards (and now MS optical mice) on almost every computer that I use. Ironically, almost all of those systems run Linux.
You have hit upon the crux of the argument. Old analog tech transmitters and receivers did not possess sufficiently narrow frequency response, so the whole system is a little "sloppy." Off-frequency rejection can now be quite high and so much tighter channels are possible. Although, in principle channels could be separated by less than 1 kHz as far as the transmitter and receiver are concerned, there is potential atmospheric effects and other interference which would probably dictate larger channel width. Nevertheless, the channels should be much smaller than current allocations.
An additional factor in the "scarcity" of spectrum use is the encoding technology. With high-quality encoding schemes (think *DSL), the potential throughput of the FM band is enormous. Note the fast ethernet runs at one frequency, 100 MHz, with a trivial encoding scheme and moves 100 Mbps. 100 MHz is right in the middle of the FM band.
No. You can only represent a finite number of states with a finite capacity computer. For integers, you can start counting at zero, one, two and keep going until you fill your computer memory. You might have some ridiculous precision (2^38-bits in 1 GB of RAM). This could then represent one unsigned integer between 0 and 2^(2^38), and that is still finite.
Indeed. I shelled out $120 plus shipping from Hong Kong to get a Japanese-version Broadband Adapter. Prices on Ebay are hovering north of $100 for used US versions. It works well with the Linux distribution, but it is certainly an expensive add-on to a system I got for $49 used at the local EB.
That's a pretty dumb thing to say. You don't know what people who post to this site have accomplised. Period. There are some pretty clever people here.
Indeed. John Carmack posts here occassionally, and I believe that most people would respect him as competent programmer.
Re:Your frame bends, but bends back.
on
The Sexiest Metal
·
· Score: 1
In addition to the modulus of elasticity and plastic limit, we also have to consider the heat treatment and formation details. There is a reason that forged parts are stronger than cast equivalents. Likewise, you can have a piece of spring steel that is grossly flexible, but if you heat it above the utectic point, and cool it quickly, it will be extremely brittle.
Metallurgy is still about 60% art/40% science. There are so many different manufacturing processes and microscopic additives that can result in significant changes in overall material behavior. Issues here are residual state of stress, grain size, defect mobility, crystalline structure, intersticial availability, etc. So, IMO, it is impossible to make gross generalizations like "stronger than" or "more flexible than" based solely on the material choice.
...then licensing shouldn't be that much of a percentage-wise increase.
That assumes that the software is not licensed per-cpu. In fact, most engineering software is licensed per-cpu, so it costs a lot more to run a fifty processor job compared to a single processor job.
If you are using commodity parts in your computing resources, annual software licenses could easily exceed your initial hardware costs, sometimes even by a large factor.
He said "live" through a nuclear war. Cher has been fully animatronic since she showed up to perform on that Navy vessel in the fishnet-thong thing back in the early 90s.
but the $2-300 "price point" is becoming farcial with this generation of consoles.
Excellent point. 'til I was done with my Xbox purchase last November, I dropped about $700! That was an XBox, an extra controller, the S-Video kit, and five games (Madden, Gotham, Halo, DOA3, Tony Hawk). I don't regret it. All of the games were beautiful and got a lot of play.
Of course, last week, the local EB was selling off their Dreamcast stock. I dropped about $400 and got a unit, a pile of controllers (including the all-important fishing controller!), a pile of memory cards, a keyboard/mouse, and about a dozen and a half pretty good games. Running Linux on it is pretty cool too.
Anyway, it is right to remember that the "buy-in" is at least $500 on the Xbox, GameCube, and PS2. I might buy a PS2 if they drop below $200. FF10 just looks too nice!
It's been shown before that electromagnetic interference from processors can show up in a radio if you listen on the same frequency of the processor.
That is very true. Several years ago, I was working on an antenna design project at a university. We had a spectrum analyzer and a small antenna test rig. Even if I connected a low gain antenna to the unit, I could see spikes at all of the "computer" frequencies...20, 25, 33, 50, 60, 66, 75, 90, 100, 133 MHz. Those were the heady days of the fast 486 and the first- and second-generation Pentium I.
Just to check that it was coming from the neighboring engineering building, I put a directional antenna and could "detect" which computers were in which floors. The undergrad lab had all of the crap 33 MHz boxes. The grad lab on a different floor had the 100s and 133s.
I agree with your first point. There is every reason to believe that Microsoft will be able to adapt to new market realities and price their OS accordingly. They will be forced to accept lower profit margins, which could be equally damaging to the company. Don't forget, Microsoft is already scrambling to find new market structures because the PC market is approaching saturation.
I do take some issue with your second point. I believe that for the vast majority of customers, cheaper will win out. The discriminating customer may well choose the more capable system priced higher, but the typical customer will buy whatever Dell or HP is selling cheap. This becomes more and more true as the currently available budget computer systems already eclipse typical user requirements for memory, speed, and storage by at least a factor of four.
And your argument about the C64 vs. Apple II vs. PC is a bit misapplied. That was the era of the "platform" wars. Technologies were wholly incompatible. Software existed for one but not the other. Now, the hardware is homogenized. (ATA hard drives, SDRAM, CD/DVD drives in PCs/Macs/Alpha alike). Most software (or functional equivalents) is available on all major platforms. The market as a whole was "wised up" since those heady days in the 80s.
And the PCs "win" was not about market penetration. The C64 was *the* gaming platform for over 5 years and sold millions of units. The eventual demise of those other systems was the fact that the companies, specifically Commodore, failed to innovate quickly enough, failed to maintain backwards compatibility, and failed to market to business. The clone companies built the PC market by following IBM's lead and stealing their market. In many ways, the open-source market is doing the same thing--following Microsoft's lead and stealing their market.
The hardware is capable "enough" and growing ever cheaper so as to dig deeper into a nearly saturated market ("Yes, honey, I think we *do* need a third computer.") Either Microsoft will have to concede hugh reductions in licensing costs and hence lower profits or they will have to find and exploit a very different revenue stream (i.e.,.NET) to maintain their financial growth.
Combining two smaller displays into one effective display system is a smart idea. I wish that "stitchable" displays where available. I remember a discussion about this once. Essentially, the probabability of failure of a display is proportional to the number of pixels in the display. The ratio of "wasted" pixels, is of course, proportional to the number of pixels in the display. So as displays contain larger numbers of pixels, both the probability of failure and the "wasted" pixel count grow quickly.
But if one could manufacture smaller patches of pixels (say 128x128 blocks) and stitch them together seemlessly, building larger displays is as easy as stacking more of these blocks together. Yield problems become less significant, because per-block testing can be done to ferit out units containing bad pixels. The amount of wasted display area is also small and independent of the final display size.
Is anyone working in this area? I thought that other technologies (Plasma or LED?) were heading this way. Also, I think that this addressed some of the signaling issues too. The 128x128 pixel blocks could be accessible in a "networked" fashion. Data would stream from a central controller to each block using a high-speed protocol. The block would contain its own patch of RAM and drive the display locally.
Anyway, I hope someone get's this working soon. We need ubiquitous 21" flat monitors and 65" HD displays.
Slashdot Mirror
AP - The artificial lifeform known to the computing world as ALICE came to a violent and tragic end this evening. The good-natured AI was interacting with several online users when an unprovoced attack was leveled against her by a geek gang known as Slashdot. The miscreants pelted ALICE with connection requests until she finally expired. FBI and local authorities are investigating and promise to bring these geeks to justise.
ALICE is survived by a grandfather TRS-80, her mother C-64, her sister IBM RS6000, and lifelong companion Athlon. In lieu of flowers, the family asks the donations be made to the Free Software Foundation.
I will second that. I have bought about $60k worth of stuff from mwave over the past few years. Good prices. Good service. Only complaint is the delivery time from CA to the east coast, but this is not their fault.
I was thinking the same thing, in the opposite direction, I guess. The "gimmick" here is a lower price due and the different OS. But, there is no "Sims" or "Deer Hunter" for Linux, so who is going knowingly buy this. Linux-types have a high likelihood of rolling their own. Business-types (the real potential market for Linux, IMO) will not be buying from a retailer like Walmart.
The *best* this could do for the average joe is provide capabilities similar to the iOpener or other web appliances. Email, web access, write a book report. But, when it doesn't work with the new digital camera or $40 ink-jet-of-the-month printer, there will be big problems. There is a significant chance that these "penguin" systems will turn into albatrosses.
I would personally prefer to see Linux inflitrate the home-user market last after a significant portion of the non-geek population is exposed to it at work. Honestly, the home market is the area where Linux has the least to offer.
I believe that he is both a crack pot and a genius.
I think that I agree with you on this point. I too am a computational scientist (CFD). Of course, I was most interested in what Wolfram touted as a major success of his work, a description of fluid turbulence. So, I slogged my way through introductory chapters and wondered at some of the complex patterns which could be coaxed from simple programs and initial conditions. I was hopeful.
When I read his take on CFD and fluid turbulence, I started to see that he didn't really have anything to offer me. Yes, he has shown that simple programs can capture much of the "structure" of turbulence and that it wonderful. So what? This was known 15 years ago. And, CA applications to computational fluid dynamics problems have been largely a curiousity, and that assessment may be kind.
This approach of using CA to model a physical system might be instructive to others who are basking in the light of this "new" science. Lattice gas models were tried. These were pure CAs, little particles that rattled around on discrete grids. The simulations were fast and simple to program. The problem was that the results were too noisy. A large fraction of simulation time was spend filtering the "results" to deduce physical pressures, velocities, etc. Evenually, the lattice gas models gave way to lattice Boltzmann models which wrote TRANSPORT equations for these particles to implicitly remove the noise. So the "new science" description gave way to the "old science" version.
So, as a scientist and engineer presented with a new model, I am always forced to ask, "So what?" What new capabilities does this offer me? What problems can I solve now that I couldn't before? How can I solve old problems faster or more accurately then before? Unfortunately, for me, this goes on the pile of interesting but currently useless theories. I will keep an eye on the field, but I would encourage its proponents to reach a bit further and make solid contact with tangible results. Make the theory applicable to real problems.
You wouldn't need to do a 9-hour exposure either. Just take a dozen or so digital photos from the same vantage point. Then correlate the resulting images. Retain the pixels that are the same in two or more of the frames. That should effectively delete any moving objects. Aside from expediting the process, it would also avoid the photographic problems of longtime exposure, like the lighting changing, etc.
Neat idea though.
We will shift to other resources as economic and/or political factors demand. The government can do a great deal to *encourage* the development of alternative technologies. That is a useful mechanism to help emerging technologies (i.e., fuel cells) reach a critical mass as which point they would become economically self-sustaining.
They can also encourage conservation, recycling efforts, and alternative technologies with tax credits, research grants, etc. I am not the least bit opposed to this.
The critical point is that it must be economically self sustaining. Otherwise, we will get the same "bust" that occurred with solar power in the early 80's. The technologies were propped up with tax incentives but inevitably were unable to compete in the free market. Encourage new technologies with SBIR funding or tax incentives, but let's not get stupid and *mandate* this sort of thing.
Non-biodegradeable this and that. We need to recognize something. We as a species will almost always do what is easiest, cheapest, and most directly to the purpose. One hundred years ago, most things were made out of wood, because it was readily available and it was fairly cheap to process. Fifty years ago, everything was made out of steel because it was readily available and cheap to process. Now, everything is made out of plastic because...you get the idea.
In the future, oil reserves will diminish. We will need to adjust our modes of production and our supply lines. This goes for consumer goods, for food, and for energy. It is all part of the economic optimization process. *That* is the fundamental principle that has been at work in the rise of nature and human civilization. As a working engineer, I know that it takes effort and know-how to build a reasonably efficient design process that minimize costs and maximizes operational efficiency and still conforms to the current emissions requirements. Adding further constraints to that engineering process will come at some cost in quality or availability. In an economic reality, this stuff will just cost more.
The most telling point is the institution of the "fourth R." Regulation. Yep, more regulation is the key. That will guarantee that the "environmentalist" teet doesn't dry up for they can retire. We need better engineered systems that do their utmost to optimize system efficiency and minimize costs.
</rant>
Commercial. As you suggest, I usually avoid these types of limitations by dropping sshd onto 2222 instead of 22. But the current problem arises because the firewall now drops outgoing connections going to ANY ports other than 20, 21, 23, 80, 443. That's right...telnet and ftp are allowed, but ssh isn't.
Also, note the absense of the word "corporatism." Most disturbing.
I also agree that a server makes the most sense. I would amplify these recommended transport mechanisms to include a few others that will allow remote connectivity.
First is a secure IMAP server for centralized email. This will allows any SSL-enabled IMAP client to access your mailbox. Also, Squirrelmail running on an SSL web server can give your access to your centralize mail repository from any web browser.
SMB and NFS are the obvious choices for LAN-based access, but WAN access needs more care. I think that a VPN setup using CIPE is a good approach. One the CIPE links are build, you can use most services as if you were located on your wired LAN.
The other need might be for file access from "arbitrary" locations. In addition to the normal scp and sftp apps in OpenSSH, there is a nice SCP client for windows, WinSCP. Lastly, if you have a SSL web server there already, Web-FTP will give you access to your files via https.
This sounds like a lot. In the end, you would need to expose SSH, SSL IMAP, SSL Apache, and CIPE servers. I am midway through this deployment myself, but it has stalled a bit because one of primary Internet access points started disallowing outgoing SSH.
The MS ergonomic keyboard *taught* me to type. This is no lie. I had been using computers since I was 6 or 7 years old...(Timex Sinclair, C64, etc)...and had never really learned to type. Instead, I had developed a semi-hunt-n-peck process that was not very efficient but it got me by.
Then about five years ago, I built a new computer and decided to pick up an MS Natural keyboard because I was starting to notice a little pain in my wrists after hours and hours of coding. So, I got the MS keyboard and immediately hated it. With it, your hands lay across the keyboard and obsure about 50 to 75% of the keys on the keyboard. Looking at the keyboard was now difficult, which is probably why many people dislike them. But, it did force me to stop using that visual crutch and start relying on muscle memory to find the keys.
Within a month or so, I had completely adapted and I could type 40 or 50 words per minute as a result. Now, I am reticent to use any other keyboard. I have MS Natural keyboards (and now MS optical mice) on almost every computer that I use. Ironically, almost all of those systems run Linux.
You have hit upon the crux of the argument. Old analog tech transmitters and receivers did not possess sufficiently narrow frequency response, so the whole system is a little "sloppy." Off-frequency rejection can now be quite high and so much tighter channels are possible. Although, in principle channels could be separated by less than 1 kHz as far as the transmitter and receiver are concerned, there is potential atmospheric effects and other interference which would probably dictate larger channel width. Nevertheless, the channels should be much smaller than current allocations.
An additional factor in the "scarcity" of spectrum use is the encoding technology. With high-quality encoding schemes (think *DSL), the potential throughput of the FM band is enormous. Note the fast ethernet runs at one frequency, 100 MHz, with a trivial encoding scheme and moves 100 Mbps. 100 MHz is right in the middle of the FM band.
No. You can only represent a finite number of states with a finite capacity computer. For integers, you can start counting at zero, one, two and keep going until you fill your computer memory. You might have some ridiculous precision (2^38-bits in 1 GB of RAM). This could then represent one unsigned integer between 0 and 2^(2^38), and that is still finite.
Or Mars Attacks....substitute corny 40's country music for 'computer virus.' The storyline has both been redone and parodied.
Move along. Nothing more to see.
Just a shame about the broadband adapter drought.
Indeed. I shelled out $120 plus shipping from Hong Kong to get a Japanese-version Broadband Adapter. Prices on Ebay are hovering north of $100 for used US versions. It works well with the Linux distribution, but it is certainly an expensive add-on to a system I got for $49 used at the local EB.
That's a pretty dumb thing to say. You don't know what people who post to this site have accomplised. Period. There are some pretty clever people here.
Indeed. John Carmack posts here occassionally, and I believe that most people would respect him as competent programmer.
In addition to the modulus of elasticity and plastic limit, we also have to consider the heat treatment and formation details. There is a reason that forged parts are stronger than cast equivalents. Likewise, you can have a piece of spring steel that is grossly flexible, but if you heat it above the utectic point, and cool it quickly, it will be extremely brittle.
Metallurgy is still about 60% art/40% science. There are so many different manufacturing processes and microscopic additives that can result in significant changes in overall material behavior. Issues here are residual state of stress, grain size, defect mobility, crystalline structure, intersticial availability, etc. So, IMO, it is impossible to make gross generalizations like "stronger than" or "more flexible than" based solely on the material choice.
...then licensing shouldn't be that much of a percentage-wise increase.
That assumes that the software is not licensed per-cpu. In fact, most engineering software is licensed per-cpu, so it costs a lot more to run a fifty processor job compared to a single processor job.
If you are using commodity parts in your computing resources, annual software licenses could easily exceed your initial hardware costs, sometimes even by a large factor.
He said "live" through a nuclear war. Cher has been fully animatronic since she showed up to perform on that Navy vessel in the fishnet-thong thing back in the early 90s.
but the $2-300 "price point" is becoming farcial with this generation of consoles.
Excellent point. 'til I was done with my Xbox purchase last November, I dropped about $700! That was an XBox, an extra controller, the S-Video kit, and five games (Madden, Gotham, Halo, DOA3, Tony Hawk). I don't regret it. All of the games were beautiful and got a lot of play.
Of course, last week, the local EB was selling off their Dreamcast stock. I dropped about $400 and got a unit, a pile of controllers (including the all-important fishing controller!), a pile of memory cards, a keyboard/mouse, and about a dozen and a half pretty good games. Running Linux on it is pretty cool too.
Anyway, it is right to remember that the "buy-in" is at least $500 on the Xbox, GameCube, and PS2. I might buy a PS2 if they drop below $200. FF10 just looks too nice!
It's been shown before that electromagnetic interference from processors can show up in a radio if you listen on the same frequency of the processor.
That is very true. Several years ago, I was working on an antenna design project at a university. We had a spectrum analyzer and a small antenna test rig. Even if I connected a low gain antenna to the unit, I could see spikes at all of the "computer" frequencies...20, 25, 33, 50, 60, 66, 75, 90, 100, 133 MHz. Those were the heady days of the fast 486 and the first- and second-generation Pentium I.
Just to check that it was coming from the neighboring engineering building, I put a directional antenna and could "detect" which computers were in which floors. The undergrad lab had all of the crap 33 MHz boxes. The grad lab on a different floor had the 100s and 133s.
I agree with your first point. There is every reason to believe that Microsoft will be able to adapt to new market realities and price their OS accordingly. They will be forced to accept lower profit margins, which could be equally damaging to the company. Don't forget, Microsoft is already scrambling to find new market structures because the PC market is approaching saturation.
.NET) to maintain their financial growth.
I do take some issue with your second point. I believe that for the vast majority of customers, cheaper will win out. The discriminating customer may well choose the more capable system priced higher, but the typical customer will buy whatever Dell or HP is selling cheap. This becomes more and more true as the currently available budget computer systems already eclipse typical user requirements for memory, speed, and storage by at least a factor of four.
And your argument about the C64 vs. Apple II vs. PC is a bit misapplied. That was the era of the "platform" wars. Technologies were wholly incompatible. Software existed for one but not the other. Now, the hardware is homogenized. (ATA hard drives, SDRAM, CD/DVD drives in PCs/Macs/Alpha alike). Most software (or functional equivalents) is available on all major platforms. The market as a whole was "wised up" since those heady days in the 80s.
And the PCs "win" was not about market penetration. The C64 was *the* gaming platform for over 5 years and sold millions of units. The eventual demise of those other systems was the fact that the companies, specifically Commodore, failed to innovate quickly enough, failed to maintain backwards compatibility, and failed to market to business. The clone companies built the PC market by following IBM's lead and stealing their market. In many ways, the open-source market is doing the same thing--following Microsoft's lead and stealing their market.
The hardware is capable "enough" and growing ever cheaper so as to dig deeper into a nearly saturated market ("Yes, honey, I think we *do* need a third computer.") Either Microsoft will have to concede hugh reductions in licensing costs and hence lower profits or they will have to find and exploit a very different revenue stream (i.e.,
Off topic...but that is a seriously cool sig. I sure wasn't expecting a Sierpinski triangle.
And wouldn't it be cool if these guys get together with the guys that figured out how Guiness makes some bubbles sink.
Too late. That has already been solved using Computational Fluid Dynamics.
Combining two smaller displays into one effective display system is a smart idea. I wish that "stitchable" displays where available. I remember a discussion about this once. Essentially, the probabability of failure of a display is proportional to the number of pixels in the display. The ratio of "wasted" pixels, is of course, proportional to the number of pixels in the display. So as displays contain larger numbers of pixels, both the probability of failure and the "wasted" pixel count grow quickly.
But if one could manufacture smaller patches of pixels (say 128x128 blocks) and stitch them together seemlessly, building larger displays is as easy as stacking more of these blocks together. Yield problems become less significant, because per-block testing can be done to ferit out units containing bad pixels. The amount of wasted display area is also small and independent of the final display size.
Is anyone working in this area? I thought that other technologies (Plasma or LED?) were heading this way. Also, I think that this addressed some of the signaling issues too. The 128x128 pixel blocks could be accessible in a "networked" fashion. Data would stream from a central controller to each block using a high-speed protocol. The block would contain its own patch of RAM and drive the display locally.
Anyway, I hope someone get's this working soon. We need ubiquitous 21" flat monitors and 65" HD displays.