Ugh, they can't even get the important details right:
As the millenium dawns in North America....
What, this movie covers a whole year? Doubt it. NBC, here's a clue: The 3rd Millennium and the 21st Century both start on January 1st, 2001. Why don't people get it?
Free doesn't necessarily mean anything to these guys. This is PHB land. When I hired in, it was mentioned to me that it is often harder to get a $1000 PC than it is to get a $1,000,000 tester. They also aren't thrilled with putting Linux on PCs instead of paying for WinNT or Win95. etc... etc... etc...
Hmm, we have couple Domino servers here at work. I wonder if a huge corporate monolith like my present employer would ever consider the Linux flavor of Domino.
It'd be interesting to hear about who is actually deploying Domino servers on Linux.
Emulating it, not quite shure if that is as impressive. Wouldn't it take away from the power of a direct port?
I think you may have the wrong idea on what they mean by "emulation" here. This is not the same as, say, videogame emulators, where you're emulating one machine on another, and as a result you spend alot of machine cycles emulating the other machine.
Rather, this is probably closer to how FreeBSD "emulates" Linux: It provides a compatible Application Binary Interface, and it just shuttles all the Linux system calls over to FreeBSD equivalents. A Linux emulator for OS/390 would probably consist of a special loader, and a POSIX to OS/390 translation layer. Depending on how it's done, it could be a reasonable environment, and not nearly as computationally costly as, say, emulating a full computer running Linux as a process on a 390 mainframe.
Someone whack me with a clue stick if I got any of this wrong.:-)
AOL claimed that because Microsoft was using *their* servers for MS' services with authorization, they had basically hacked into AOL's networks and proceeded to (apparently) use a buffer overflow exploit to detect AIM clients.
The grammar of this sentence is confusing. Microsoft was using AOLs servers for Microsoft's instant-messanger product because it uses AOL's protocol to talk to other AIM users. AOL has tweaked their protocol a dozen times to prevent this, and each time, Microsoft tweaks their client to match. Finally, AOL decided to exploit a buffer overflow in their own client in order to prevent MS from being able to further tweak to be compatible.
I'm sorry, but I'd have to agree with MS on this one: AOL should open up their protocol and secure your clients. I'm not holding my breath though. It's pretty clear that AOL is only interested in security to the extent it affects their bottom line. Unless people just decide to give up on AIM and AOL and take their dollars elsewhere, this isn't going to hit their pocketbook, which is why AOL still hasn't fixed it. After all, consider the average AOL user. (Yes, there are a few intelligent people who use AOL. It's a little like saying "Yeah, there are a few intelligent people on Earth." Most people are idiots.)
Since the S@H client takes over so much of the machine, I believe most people only run it when they aren't using their computers.
That may be true on the Windows and the Mac clients. The Linux client runs happily backgrounded at a nice-level of 19 on my fiancee's dual PII-450 box. Even when she's busy running LaTeX or Nethack or WordPerfect. Running an RC5 client alongside would not be out of the picture at all, even though she has two SETI clients running (one for each CPU).
Greater optimization means that the SETI folks can do more analysis on the data they have and build more redundancy (and therefore more security) into the system. I don't understand why you would argue against faster clients when the problem is already infinite in scope.
as long as you're operating in a completely different domain,
By that, I meant, "as long as your product or service is not in the same field covered by the other Trademark holder's field, then you don't infringe." For instance, you're fine as long as you're not doing chicken or being an educational institution.
An operating system is supposed to provide a protected interface to the hardware. MS-DOS does not do that. This notion of carefully controlled, mediated access to the computer's underlying raw resources is hardly a new concept today, nor was it back when Multics was doing rings of protection -- which, you will note, antedates Unix significantly.
The concepts of protection and security are relatively new concepts in the personal computer world. Microsoft has never really embraced these concepts either, it would seem, and I imagine it's because most of their customers don't care. (Or, at least didn't care.) Rather, they seem to be more interested in the opposite -- integrating everything with everything else and separating nobody from anything.
Part of the reason for this, I imagine, is that the original user base for PCs and related equipment really didn't want anything in the way between themselves and the machine. The OS was a glorified boot loader that additionally provided some useful routines. Look at the Apple ][, Commodore 64, IBM PC, etc. at their inception. The only machine that truly insulated you from the hardware (TI-99/4 and TI-99/4A) died earliest.
I remember someone musing around this time (early/mid 80s) that the hardest thing you could try with your computer was to hook it up to another computer. This remained largely truly until the last decade, and for the bulk of non-business computers, the last few years. Is it any wonder that the notions of security and paranoia just aren't built in?
The issue with GPL is that you need to distribute or make available the source to your changes whenever yout distribute your software to someone else. If the PRC's government makes proprietary modifications to Linux and only deploys it within the governement, they have absolutely no reason to share it with the world. None.
Nowhere in the GPL does it say you need to share all of your development work with the world if you happen to use a GPL'd work as a starting point. The GPL kicks in the moment you try to redistribute your work, since you can't redistribute a modified GPL'd work unless you make the source available for the modified work.
The GPL really is about freedom, including the freedom to take the source and do whatever the hell you want with it behind closed doors. The only requirement the GPL makes is that you must extend the same freedoms to others that were extended to you if you choose to distribute your work. I doubt we're going to see a PRC Linux Distribution anytime soon.
Take a look at this page on the USPTO's own web site. It allows you to order customized set of patents on their site, and they can either be downlaoded onto your machine or sent to you in the mail.
CD-Now's patent seems to cover the general concept of customizing a product via a website, and automating its manufacture and delivery. The key difference they site between their patent and the (dubious) prior art is that it involves a network such as the Internet. The only thing that ties their patent to burning CDs is that that is the "preferred embodiment."
This whole idea of patenting business models is absurd, and needs to stop.
Of course an ethernet card is "not the gateway to the communication infrastructure," at least if you define that as the POTS network or the Internet backbone. Neither is a cable modem. Cable modem networks actually behave alot like Ethernet in that it's a broadcast medium within each segment. Where it differs is in the fact that segments are arranged hierarchially, and all nodes on the network are not peers as they are in ethernet. (Apparently, nodes that are nearer to the top of the hierarchy cannot report collisions to nodes beneath them on a cable network, if I recall correctly. On Ethernet, everyone's a peer -- there is no above/below.)
In both cases, you need a separate gateway device which hooks your network segment up to the Internet or whatever "communication infrastructure" you're conceiving. They really aren't very different AT ALL.
I dare say that the PHY layer behaves more like an analog layer than a digital one. Hence the term Carrier Sense, Multiple Access / Collision Detection. Pure binary digital signals are mutually exclusive with the concept of a "collision".
Also, Manchester encoding is a very light form of modulation, in that it ensures a transition during every bit period (a bit like XORing the data sequence with the clock -- looks alot like AM modulation).
BTW, your final result (19MBytes) looks about right from your original equation, since my adjusted result is the same as yours divided by 0.7. It looks like you just typed in the substitution incorrectly when you said "299792458/300000".
Incidentally, this phenomenon is why Linux allows you to allow large TCP windows. Look at the Configure.help entry for CONFIG_SKB_LARGE. It points out that "[a]t 45 MBit/second there are alot of bits between New York and London...".
By that argument, any interface which converts digital signals into an analog transport and back with some form of modulation should be called a modem. Like your Ethernet card, for instance. We all call them Ethernet Modems don't we? Didn't think so. That said, the term "cable modem" is a reasonably accurate name for what it is/does, and since it replaces a traditional POTS modem, its name matches its intended purpose pretty well. --Joe --
I'm curious, can the speed of light be measured in Gb/s?
*sigh* People keep getting rate of propogation confused with rate of transfer. This is latency vs. bandwidth folks.
For instance, consider the ancient communication method consisting of two people atop hills signalling with lanterns and shades. The latency is really low because the light propogates at near 3e8 m/s in air. The bandwidth sucks. Now consider a freight-train loaded to the gills with DVD-ROMs. The bandwidth is enormous, but the latency sucks.
The speed of light governs how quickly a packet of data gets from point A to point B. Bandwidth measures the total number of packets of data that you can send from point A to point B at a given time. The two are different, unless you somehow treat each photon as its own packet of data, and we're not there yet.
Bzzzt.... try again. The UART's speed setting has nothing to do with the actual data rate on the wire. And don't say "but I have compression turned on." Doesn't help where you need it most -- downloading huge files. Also, 115kbps is far below even ISA's bandwidth.
Next time, do some research before you post.
And to make this post marginally on-topic: It's doubtful that these optical developments will speed up POTS modems at all, except in the cases that they cause old, crufty analog lines to be bulk-replaced with new digital lines. These optical developments are more important for backbones and long-haul connections, not connections between you and your CO.
You mention: "If you stay within the internal 4k data and 4k code the '211 is fairly self contained . .." Don't forget that there's the 64K L2 on there, which can be configured partly as SRAM and partly as cache. Cache misses to L2 aren't all that expensive.
Overall, though, looks like you've got a good impression of the part. The reason I ask is that I work with the folks who put the C6211 together.:-) I'm always interested in hearing "reports from the field." Perhaps send me an email sometime?
Bourne Again Shell (aka bash) and Korn Shell (aka ksh) both are very similar at a scripting level. What scripting features are available in ksh that aren't as nice in bash?
Actually, from what I understand, the normal ettiquette of a spoken conversation starts breaking down once the one-way travel time gets above around 100ms. (eg. a Round-Trip Time of 200ms.) Something like the gamer's modem might be nice here.:-)
DSPs are in modems, cell-phones, hard-drives, sound cards, and tons of other places. I wonder which market Intel is going to aim for. The article seems to state they're interested in wireless, which is currently a big business for DSP. I can't imagine that's all they're interested in though.... Hrmm......
People can't even decide how to pronounce vi and GIF, when both of these have unambiguous official pronounciations. (Quickly: vee eye, and jiff, NOT vye and giff (as in 'gift')). Next, you'll be arguing over whether bin and lib should be pronounced with a long or short I. (I vote 'short'.)
Give it up, people.
Besides, how many of you could correctly pronounce my last name, ZBICIAK, even if I told you what the correct pronounciation was? (I don't know the truly correct pronounciation myself, a testament to the fact that it doesn't freakin' matter.)
I actually got in an argument with a teacher over how "DOS" was pronounced. She insisted it was pronounced with a long O, as in "dose" (eg. like the Spanish number 2). I of course insisted on a short O, "doss", if you will...
Slashdot Mirror
Ugh, they can't even get the important details right:
What, this movie covers a whole year? Doubt it. NBC, here's a clue: The 3rd Millennium and the 21st Century both start on January 1st, 2001. Why don't people get it?
--Joe--
Free doesn't necessarily mean anything to these guys. This is PHB land. When I hired in, it was mentioned to me that it is often harder to get a $1000 PC than it is to get a $1,000,000 tester. They also aren't thrilled with putting Linux on PCs instead of paying for WinNT or Win95. etc... etc... etc...
--Joe--
Hmm, we have couple Domino servers here at work. I wonder if a huge corporate monolith like my present employer would ever consider the Linux flavor of Domino.
It'd be interesting to hear about who is actually deploying Domino servers on Linux.
--Joe--
I think you may have the wrong idea on what they mean by "emulation" here. This is not the same as, say, videogame emulators, where you're emulating one machine on another, and as a result you spend alot of machine cycles emulating the other machine.
Rather, this is probably closer to how FreeBSD "emulates" Linux: It provides a compatible Application Binary Interface, and it just shuttles all the Linux system calls over to FreeBSD equivalents. A Linux emulator for OS/390 would probably consist of a special loader, and a POSIX to OS/390 translation layer. Depending on how it's done, it could be a reasonable environment, and not nearly as computationally costly as, say, emulating a full computer running Linux as a process on a 390 mainframe.
Someone whack me with a clue stick if I got any of this wrong. :-)
--Joe--
The grammar of this sentence is confusing. Microsoft was using AOLs servers for Microsoft's instant-messanger product because it uses AOL's protocol to talk to other AIM users. AOL has tweaked their protocol a dozen times to prevent this, and each time, Microsoft tweaks their client to match. Finally, AOL decided to exploit a buffer overflow in their own client in order to prevent MS from being able to further tweak to be compatible.
I'm sorry, but I'd have to agree with MS on this one: AOL should open up their protocol and secure your clients. I'm not holding my breath though. It's pretty clear that AOL is only interested in security to the extent it affects their bottom line. Unless people just decide to give up on AIM and AOL and take their dollars elsewhere, this isn't going to hit their pocketbook, which is why AOL still hasn't fixed it. After all, consider the average AOL user. (Yes, there are a few intelligent people who use AOL. It's a little like saying "Yeah, there are a few intelligent people on Earth." Most people are idiots.)
--Joe--
That may be true on the Windows and the Mac clients. The Linux client runs happily backgrounded at a nice-level of 19 on my fiancee's dual PII-450 box. Even when she's busy running LaTeX or Nethack or WordPerfect. Running an RC5 client alongside would not be out of the picture at all, even though she has two SETI clients running (one for each CPU).
Greater optimization means that the SETI folks can do more analysis on the data they have and build more redundancy (and therefore more security) into the system. I don't understand why you would argue against faster clients when the problem is already infinite in scope.
--Joe--
By that, I meant, "as long as your product or service is not in the same field covered by the other Trademark holder's field, then you don't infringe." For instance, you're fine as long as you're not doing chicken or being an educational institution.
--Joe--
Trademarks are granted over a particular domain. For instance Microsoft has trademarked Windows as it applies to their software product, and so has Randall Products International for their sunscreen product.
So, whether you spell it Purdue or Perdue, as long as you're operating in a completely different domain, you don't infringe on their trademark.
--JoePS. The links above work for me right now, but they did result from a search for "Windows" on the USPTO site. I don't know if search links expire.
--
The concepts of protection and security are relatively new concepts in the personal computer world. Microsoft has never really embraced these concepts either, it would seem, and I imagine it's because most of their customers don't care. (Or, at least didn't care.) Rather, they seem to be more interested in the opposite -- integrating everything with everything else and separating nobody from anything.
Part of the reason for this, I imagine, is that the original user base for PCs and related equipment really didn't want anything in the way between themselves and the machine. The OS was a glorified boot loader that additionally provided some useful routines. Look at the Apple ][, Commodore 64, IBM PC, etc. at their inception. The only machine that truly insulated you from the hardware (TI-99/4 and TI-99/4A) died earliest.
I remember someone musing around this time (early/mid 80s) that the hardest thing you could try with your computer was to hook it up to another computer. This remained largely truly until the last decade, and for the bulk of non-business computers, the last few years. Is it any wonder that the notions of security and paranoia just aren't built in?
--Joe--
The issue with GPL is that you need to distribute or make available the source to your changes whenever yout distribute your software to someone else. If the PRC's government makes proprietary modifications to Linux and only deploys it within the governement, they have absolutely no reason to share it with the world. None.
Nowhere in the GPL does it say you need to share all of your development work with the world if you happen to use a GPL'd work as a starting point. The GPL kicks in the moment you try to redistribute your work, since you can't redistribute a modified GPL'd work unless you make the source available for the modified work.
The GPL really is about freedom, including the freedom to take the source and do whatever the hell you want with it behind closed doors. The only requirement the GPL makes is that you must extend the same freedoms to others that were extended to you if you choose to distribute your work. I doubt we're going to see a PRC Linux Distribution anytime soon.
--Joe--
Take a look at this page on the USPTO's own web site. It allows you to order customized set of patents on their site, and they can either be downlaoded onto your machine or sent to you in the mail.
CD-Now's patent seems to cover the general concept of customizing a product via a website, and automating its manufacture and delivery. The key difference they site between their patent and the (dubious) prior art is that it involves a network such as the Internet. The only thing that ties their patent to burning CDs is that that is the "preferred embodiment."
This whole idea of patenting business models is absurd, and needs to stop.
--Joe--
Grab the actual RM file here: L inus' keynote RM file, 46862773 bytes
--Joe--
It's not ludicrous. Check your facts.
Of course an ethernet card is "not the gateway to the communication infrastructure," at least if you define that as the POTS network or the Internet backbone. Neither is a cable modem. Cable modem networks actually behave alot like Ethernet in that it's a broadcast medium within each segment. Where it differs is in the fact that segments are arranged hierarchially, and all nodes on the network are not peers as they are in ethernet. (Apparently, nodes that are nearer to the top of the hierarchy cannot report collisions to nodes beneath them on a cable network, if I recall correctly. On Ethernet, everyone's a peer -- there is no above/below.)
In both cases, you need a separate gateway device which hooks your network segment up to the Internet or whatever "communication infrastructure" you're conceiving. They really aren't very different AT ALL.
--Joe--
I dare say that the PHY layer behaves more like an analog layer than a digital one. Hence the term Carrier Sense, Multiple Access / Collision Detection. Pure binary digital signals are mutually exclusive with the concept of a "collision".
Also, Manchester encoding is a very light form of modulation, in that it ensures a transition during every bit period (a bit like XORing the data sequence with the clock -- looks alot like AM modulation).
See this online guide for more details.
--Joe--
There's more than that in-flight in the wire, since, as someone else pointed out, the speed of light in fiber is about 0.7C.
BTW, you accidentally inverted your fraction in there. The correct equation is:
speed = 0.7 c = 0.7 * (299792458 m/s) = 209854721 m/sdist = 300000m
bps*nbsp; = 160000000000 bits/sec
bits_in_wire = (dist / speed) * bps
bits_in_wire = (300000m / 209854721 m/s) * 160000000000 bits/sec
bits_in_wire = 228729664 bits = approx 27 MegaBytes
BTW, your final result (19MBytes) looks about right from your original equation, since my adjusted result is the same as yours divided by 0.7. It looks like you just typed in the substitution incorrectly when you said "299792458/300000".
Incidentally, this phenomenon is why Linux allows you to allow large TCP windows. Look at the Configure.help entry for CONFIG_SKB_LARGE. It points out that "[a]t 45 MBit/second there are alot of bits between New York and London ...".
--Joe--
By that argument, any interface which converts digital signals into an analog transport and back with some form of modulation should be called a modem. Like your Ethernet card, for instance. We all call them Ethernet Modems don't we? Didn't think so. That said, the term "cable modem" is a reasonably accurate name for what it is/does, and since it replaces a traditional POTS modem, its name matches its intended purpose pretty well. --Joe
--
*sigh* People keep getting rate of propogation confused with rate of transfer. This is latency vs. bandwidth folks.
For instance, consider the ancient communication method consisting of two people atop hills signalling with lanterns and shades. The latency is really low because the light propogates at near 3e8 m/s in air. The bandwidth sucks. Now consider a freight-train loaded to the gills with DVD-ROMs. The bandwidth is enormous, but the latency sucks.
The speed of light governs how quickly a packet of data gets from point A to point B. Bandwidth measures the total number of packets of data that you can send from point A to point B at a given time. The two are different, unless you somehow treat each photon as its own packet of data, and we're not there yet.
--Joe--
Bzzzt.... try again. The UART's speed setting has nothing to do with the actual data rate on the wire. And don't say "but I have compression turned on." Doesn't help where you need it most -- downloading huge files. Also, 115kbps is far below even ISA's bandwidth.
Next time, do some research before you post.
And to make this post marginally on-topic: It's doubtful that these optical developments will speed up POTS modems at all, except in the cases that they cause old, crufty analog lines to be bulk-replaced with new digital lines. These optical developments are more important for backbones and long-haul connections, not connections between you and your CO.
--Joe--
You mention: "If you stay within the internal 4k data and 4k code the '211 is fairly self contained . . ." Don't forget that there's the 64K L2 on there, which can be configured partly as SRAM and partly as cache. Cache misses to L2 aren't all that expensive.
Overall, though, looks like you've got a good impression of the part. The reason I ask is that I work with the folks who put the C6211 together. :-) I'm always interested in hearing "reports from the field." Perhaps send me an email sometime?
--Joe--
Bourne Again Shell (aka bash) and Korn Shell (aka ksh) both are very similar at a scripting level. What scripting features are available in ksh that aren't as nice in bash?
--Joe--
BTW, what do you think of the C6211? :-)
--Joe--
Actually, from what I understand, the normal ettiquette of a spoken conversation starts breaking down once the one-way travel time gets above around 100ms. (eg. a Round-Trip Time of 200ms.) Something like the gamer's modem might be nice here. :-)
--Joe--
DSPs are in modems, cell-phones, hard-drives, sound cards, and tons of other places. I wonder which market Intel is going to aim for. The article seems to state they're interested in wireless, which is currently a big business for DSP. I can't imagine that's all they're interested in though.... Hrmm......
--Joe--
People can't even decide how to pronounce vi and GIF , when both of these have unambiguous official pronounciations. (Quickly: vee eye, and jiff, NOT vye and giff (as in 'gift')). Next, you'll be arguing over whether bin and lib should be pronounced with a long or short I. (I vote 'short'.)
Give it up, people.
Besides, how many of you could correctly pronounce my last name, ZBICIAK, even if I told you what the correct pronounciation was? (I don't know the truly correct pronounciation myself, a testament to the fact that it doesn't freakin' matter.)
--Joe--
I actually got in an argument with a teacher over how "DOS" was pronounced. She insisted it was pronounced with a long O, as in "dose" (eg. like the Spanish number 2). I of course insisted on a short O, "doss", if you will...
*sigh*
--Joe--