Well, it can tell you several things. Those things are not very useful in themselves, but they can aid you in determining the system configuration:
1. Server or desktop/workstation. Desktop systems are rebooted quite regularly. Servers tend to stay up for weeks to months.
2. From the uptime you may make some inferences about the OS on the machine. A machine with a year of uptime is likely a SUN, HP, IBM, etc "big iron" machine. Basially you know it's not windows so you can skip all those attacks.
3. If a machine isn't rebooted often, there probably isn't a monitor attached to it so no-one is looking directly at it very often. Sure there may be load monitors and such, but unless you do something harsh the extra CPU load won't get noticed. Whereas a desktop system intrusion may be noticed by the user via slow downs or "hiccups" in GUI response.
4. If an OS constantly uses the same timestamp for TCP packets, that's a dead giveaway as to the system/OS that's running.
Much of this comes down to "security by obscurity", which while not a viable mechinism in it's own right, is quite valuable when combined with other techniques. You want your TCP stack to give away as little information to a potential attacker as possible. The less they know, the more they have to search and the more likely you will find them before they get in the system.
What "propaganda"? Bush lied. When he didn't outright lie, he spun the truth or omitted exculpitory information to create support for his agenda. Other than his statements that "Saddam Hussein is a ruthless dictator", I am hard pressed to find any solid truth in his statements about Iraq.
But XposrFacto allows you to install OS X on unsupported machines. The move to EOL the biege G3 seems to be more of a political thing than a requirement for the OS. Or perhaps they've fianlly pulled support for ADB? I already use a USB mouse and getting a KB would be trivial. I just need something that will force the installer to put Panther on my drive.
It would be interesting to use a supported machine to install Panther, then try booting that drive in an older un-supported system. Anyone have such a configuration they could try?
For the time being, Apple will ship the Xserve with Jaguar Server and Panther Server for those organizations that need a bit more time before they upgrade.
Apple isn't asking them to make a choice, they're saying "If you use Jaguar server now, keep using it, but here's a copy of Panther server so yo can upgrade to it when you're ready."
Could you ever imagine Microsoft allowing Dell, HP or Compaq to ship 2000 and XP on the same order just in case the customer was using 2000 and didn't want to upgrade to XP yet?
The switch to a different default shell in the terminal application has nothing to do with any shell commands that are included for administration of the computer.
Any shell program/script that is written by anyone who knows anything about writing shells programs/scripts will contain a shebang as the first line. The syntax look like this: #!/bin/sh This specially formatted line tells the OS what program (in that case the "sh" shell) will be able to interpret this file properly. No matter what the "default" shell is, no matter what shell you are actually running when you type the command name, the program will be run with the shell specified in that line. Ex: most perl programs have #!/usr/bin/perl or #!/usr/local/bin/perl
For the Mac savvy this is really equivilant to the "creator" field in a file's meta information.
Granted, but I don't think people build super computers to get benchmark rankings, they get benchmark ratings because they have supercomputers. While LinPack may not allow for AltiVec to be used, many of the simulations and algorithms they will run on that system WILL take advantage of vector processing.
I'm guessing that if you just wanted a high ranking, you could build a custom CPU that did nothing but the types of operations in LinPack, and put a bunch of them on circuit boards, but again... I don't really know much about what LinPack demands or tests.
Most responses in here are about how the G5 should be performing better, or should have better numbers than the Xenon or Sparc, or whatever. What seems to be missing from most of the conversation is that it's not the Mac's that are loosing efficiency per se, it's the network (the interconnects) that is slowing the machine as a whole down. I know little about the LinPac test, but I would assume that it's written to test/stress the entire machine: CPU, disk, memory and interconnects. If the Macs can finish parts of a problem really fast, but can't get new data in to the nodes fast enough, that will casue a tremendous loss in effieciency. Perhaps they need a mechanism for buffering new data on the nodes so that incoming and outgoing data can stream as the network is available and keep the CPUs working all the time.
It's not a measure of heat transfer, it's a measure of energy. You could measure the output of an automobile engine in calories if you like. Convert calories to watts to HP to torque(more or less) to thrust, it's all a different scale of the same thing.
Because the Power4 is hotter and uses more current than the G5. To use 2200 Power 4 CPUs they would have to about triple the cooling capacity of the room. For all the heat and power, the Power4 lacks the AltiVec units that allow the G4/G5 to process vector operations so quickly.
The G5 is also significantly lower cost than the Power4
Surely you left of a few zeros from that number? There's probably more than 5 million people on-line in California alone, never mind the rest of the country and world. I MIGHT buy 500 million, but even that seems low.
Unless the people that are allowed to survive "artificially" by technology or social help will in the end weaken the species. Humans (and all other species) are as resilient and strong today as they are because all the weaker, less intelligent, defective samples were killed off or died as a result of their lesser condition.
There's several inherited diseases that are starting to have major increaseses in occourance. The reason? Before all this "everyone must live" crap, they died before they could breed. Now they are given all the medical care, drugs and support they need to survive against the rules of nature. The result is that these people breed and pass the genes/disease on the their offspring. The children in turn may or may not exibit the disease, but they carry it and pass it on. Instead of fighting disiease by allowing it to die out naturally we are allowing (forcing) these people to live and instead spending billions to sove a problem that would solve itself if we just stopped trying to eliminate death.
But we've gotten way off the topic of VoIP here, so lets just end all this now and agree to disagree.
Mapquest has this feature called "trip planning". You can set "waypoints" on your trip.
I recently used the feature to tell the thing to give bme directions from Missouri to Phoenix, AZ via Flagstaff, AZ. If I recall it can be used down to the street address level.
But now your getting away from "free" services. Someone has to pay for the hardware and the local circuits to the sponsoring telcos that the gatways use. Without local dialtone from a telco (for a fee of course) you can't call from VoIP to a POTS customer. If you don't want to do that then someone has to pay to run new wires to each and every home/business.
I'm really getting tired of this incorrect focus on the fringe.
Fact: the RIAA did not subpoena a grandmother or a 12 year old girl. They sent subpoenas to ISPs in order to determine the identities of people who where using certain IP addresses. When RIAA filed these documents they had absoloutly NO idea of who the people where, their age, sex, gender, race, income, location or anything else but an IP address.
Out of 268 cases (??) so far we've had two "fringe" cases. One was a young girl. Would you be bitching about that if she'd been selling crack on the street and got busted? Even if she claimed to have thought it was okay because she was paying a policeman for the privlidge? The other case was apparently a clerical error on the part of the ISP in providing the information to answer the sunpoena. This had nothing to do with the RIAA.
Peerhaps we should use the extreems to end other practices also: In many military actions around the world, the US military has killed non-combatants including 12 year old girls and grandmas. Would you argue that everything the US Military does is then invalidated and it should stop all such action now and in the future?
No? Then take a moment and re-think your position on RIAA.
If your child falls in the pool then you did something stupid by not watching your child, or not enclosing the pool to prevent such a thing. You're genes should not be propagated. If someone breaks in to my house I am perfectly capable of handling the situation myself. In fact I have done so. Calling 911 does not prevent any damage, or injury to my person or family.
The police are not crime preventers or stoppers, they simply clean up afterward.
Natural selection works everywhere that technology doesn't interfere with it. Stupid people should be allowed to die when they do something really stupid.
This type of VoIP does not replace your standard telephone service. You can't call a telephone number with it, you can only "call" other people that use the same software that you are using to originate the call.
For instance... you can't use either service to call up the local take away shop to place an order, nor can you call you non-computer using grandmother.
You still need the landline to call other landline (or mobile) phones. You still need the landline for reliability.
To get a little off topic, 911 should be paid for by those who use it, not through my telephone bills.
On a deeper level, we as a society need to get off this "everyone needs to be rescuable at any moment at any cost" bandwagon. If we don't let the law of natural selection have any bearing on our evolution as a species, I fear we are doomed.
The problem I have with stories like this is that the calls aren't really free. You do have to have a rather high-speed internet connection to make these calls with any reasonable quality and reliability, and you have to pay that fee on top of your existing phone charges.
A major limitation is that you can only call your fiends who use the same "service". And they are for the most part defining "service" rather loosely, they're more like applications in software and hardware than a service. I know it's only on Mac now, but I'm curious why iChatAV from Apple is excluded from these types of comparisons. It does the same things, plus video and uses the AOL screen name and buddy list infrastructure.
There's a reliability issue with VoIP, I for one will not cut my dial tone off until I have nearly 100% uptime on my net connection. In all my life I think there was one time (after a hurricane) that I picked up my telephone and did not hear a dial tone. I can't count how many minutes per month/year my net connection is down for one reason or another.
I also take issue with the statement "...They do illustrate, however, just how far VOIP has come - it's actually good enough to offer a viable alternative to existing phones.". I don't think it's the VoIP technology that's improved, I think it's the Internet's infrastructure that's improved. There's finally enough bandwidth that you don't need a lot of buffering to ensure packet delivery order to the audio decoders. It's still possible and routine to get out-of-order delivery, but no-where as severe as it was even just two years ago.
I for one had a terribly hard time reading the graphs. Three problems in particular:
1) I (like about 20% of the male population) am partially color deficient. The graphs' colors where very hard for me to distinguish in the ledgend; what with the 1 pixel wide font and sample line/symbols. The colors chosen were all of about the same intensity/saturation also, so if you are color blind the entire graph would look all about the same level of gray. This problem is further compounded by all the graphs having different color assigments (Linux 2.6
2) crowding. On a few of the graphs, there were differences in order of magnitude between groupings. Since this dataset was supposed to show the fastest, that would naturally be the most interesting data. You don't take photo finishes of the 7th and 8th finishers in an 8 horse race. Using a logarithmic scale for those charts would make the data easier to read, showing detail in the lower(faster) end and allowing the less significant higher(slower) data to "blend" in the curve.
3) Point size. The symbols are too large for that many data points. This may be personal preference, but single points should be used for scatter graphs like these, not symbols. Symbols may mask other more subtle changes in the results because they overflow other points nearby. Similarly, make your legend samples substantial, a block of color is much more apreciated than a single pixel line, and bold your identifier text.
If you make graphs for the web, please take them in to Photoshop, GIMP, or something similar and reduce or remove the color data (convert to grey scale). The chart should still be at least fairly reable in greyscale. Choose a wide range of Hue, Saturation and Lightness If you are going to use light/desturated colors, use a black background. If you will use dark/heavily saturated colors, use a light background. If you can't/won't do that, then please try to make the raw data available in CSV or some other "universal" format so we can create our own graphs.
Yup, I'm glad at least a few others know of those channels and the web sites with the video clips. DishNetwork carries both, along with I think one or two other university based channels. The Research Channel I think is Univerity of California's, UWTV is University of Washington. There may also be a Rutgers based channel.
On top of that I also regularly watch NASA TV which frequently does educational/lecture type shows, and they seem to show almost any testimony (that's not classified secret) by the administrator to Congress.
Discovery Science channel isn't too bad, and the National Geographic channel is great for Earth/nature based science and education. "Daily Planet" and "National Geographic Today" are two great news shows for finding out what's really going on the world besides just the local house fire, bank robbery, who's running for governor in a state where you don't live, etc.
And of course, the Weather Channel is great for learning about meteorology with some of the shows they have.
Perhaps the author of the article needs to spend a little more money and get a good satellite DBS, or a higher tier of service from his cable provider. I can't imagine all the "science" that I get on these channels being crammed in to just one channel called like "ultimate science".
I do understand, I've been baffled myself at the lack of 80s music on iTMS. Mind you, there is quite a bit, but not as much as you might think. The "80s crowd" would seem to be a prime audience for the iTMS, we're old enough to have enough disposable income, and sentimental enough about the music at this point to spend a butt load to DL it. Of course, I've read post from about every genre about lack of selection: jazz, classical, soundtracks, metal, punk, indie... etc.
My guess is that the 80s will be getting filled in fairly soon and fairly completely now that iTMS has hit "critical mass".
You are 100% correct. Since launch the iTMS has been biased toward the music that the majority of those who are on-line purchase and listen to. You can apply the 80/20 rule to this in all probability; 80% of the people will tend to listen to about 20% of the music available. Hence, when you launch a service, you make the most profit by first including the 20% of content that will encompass most of your sales. You later fill in the remainder to satisfy the others.
In the five or so months since iTMS has been on-line, they have grown from 200,000 songs to 400,000 songs, and this isn't the BuyMusic method of accounting, you can purchase each and every song Apple counts. They've just started getting the indie labels on board and set up to submit tracks, 200 according to Steve's presentation. The indies were tripping over themselves to sign up when the meeting was held a few months back. Why wouldn't they? Apple provdides the hardware, submission is free, Apple handles all billing. Since each music company is apparently responsible for encoding and submitting the tracks, the rate of increase should be greater in the future. The next 200,000 tracks may well be added within the next two months.
There's plenty of "obscure stuff" on the store, or do you perhaps consider Andy Griffith's 'Fishin' Hole' to be main stream music?
Slashdot Mirror
Well, it can tell you several things. Those things are not very useful in themselves, but they can aid you in determining the system configuration:
1. Server or desktop/workstation. Desktop systems are rebooted quite regularly. Servers tend to stay up for weeks to months.
2. From the uptime you may make some inferences about the OS on the machine. A machine with a year of uptime is likely a SUN, HP, IBM, etc "big iron" machine. Basially you know it's not windows so you can skip all those attacks.
3. If a machine isn't rebooted often, there probably isn't a monitor attached to it so no-one is looking directly at it very often. Sure there may be load monitors and such, but unless you do something harsh the extra CPU load won't get noticed. Whereas a desktop system intrusion may be noticed by the user via slow downs or "hiccups" in GUI response.
4. If an OS constantly uses the same timestamp for TCP packets, that's a dead giveaway as to the system/OS that's running.
Much of this comes down to "security by obscurity", which while not a viable mechinism in it's own right, is quite valuable when combined with other techniques. You want your TCP stack to give away as little information to a potential attacker as possible. The less they know, the more they have to search and the more likely you will find them before they get in the system.
What "propaganda"? Bush lied. When he didn't outright lie, he spun the truth or omitted exculpitory information to create support for his agenda.
Other than his statements that "Saddam Hussein is a ruthless dictator", I am hard pressed to find any solid truth in his statements about Iraq.
But XposrFacto allows you to install OS X on unsupported machines. The move to EOL the biege G3 seems to be more of a political thing than a requirement for the OS. Or perhaps they've fianlly pulled support for ADB?
I already use a USB mouse and getting a KB would be trivial. I just need something that will force the installer to put Panther on my drive.
It would be interesting to use a supported machine to install Panther, then try booting that drive in an older un-supported system. Anyone have such a configuration they could try?
For the time being, Apple will ship the Xserve with Jaguar Server and Panther Server for those organizations that need a bit more time before they upgrade.
Apple isn't asking them to make a choice, they're saying "If you use Jaguar server now, keep using it, but here's a copy of Panther server so yo can upgrade to it when you're ready."
Could you ever imagine Microsoft allowing Dell, HP or Compaq to ship 2000 and XP on the same order just in case the customer was using 2000 and didn't want to upgrade to XP yet?
The switch to a different default shell in the terminal application has nothing to do with any shell commands that are included for administration of the computer.
Any shell program/script that is written by anyone who knows anything about writing shells programs/scripts will contain a shebang as the first line. The syntax look like this:
#!/bin/sh
This specially formatted line tells the OS what program (in that case the "sh" shell) will be able to interpret this file properly. No matter what the "default" shell is, no matter what shell you are actually running when you type the command name, the program will be run with the shell specified in that line.
Ex: most perl programs have #!/usr/bin/perl or #!/usr/local/bin/perl
For the Mac savvy this is really equivilant to the "creator" field in a file's meta information.
Granted, but I don't think people build super computers to get benchmark rankings, they get benchmark ratings because they have supercomputers.
While LinPack may not allow for AltiVec to be used, many of the simulations and algorithms they will run on that system WILL take advantage of vector processing.
I'm guessing that if you just wanted a high ranking, you could build a custom CPU that did nothing but the types of operations in LinPack, and put a bunch of them on circuit boards, but again... I don't really know much about what LinPack demands or tests.
Most responses in here are about how the G5 should be performing better, or should have better numbers than the Xenon or Sparc, or whatever.
What seems to be missing from most of the conversation is that it's not the Mac's that are loosing efficiency per se, it's the network (the interconnects) that is slowing the machine as a whole down. I know little about the LinPac test, but I would assume that it's written to test/stress the entire machine: CPU, disk, memory and interconnects. If the Macs can finish parts of a problem really fast, but can't get new data in to the nodes fast enough, that will casue a tremendous loss in effieciency.
Perhaps they need a mechanism for buffering new data on the nodes so that incoming and outgoing data can stream as the network is available and keep the CPUs working all the time.
It's not a measure of heat transfer, it's a measure of energy. You could measure the output of an automobile engine in calories if you like. Convert calories to watts to HP to torque(more or less) to thrust, it's all a different scale of the same thing.
Because the Power4 is hotter and uses more current than the G5. To use 2200 Power 4 CPUs they would have to about triple the cooling capacity of the room. For all the heat and power, the Power4 lacks the AltiVec units that allow the G4/G5 to process vector operations so quickly.
The G5 is also significantly lower cost than the Power4
...There's only about 5 million people online...
Surely you left of a few zeros from that number? There's probably more than 5 million people on-line in California alone, never mind the rest of the country and world. I MIGHT buy 500 million, but even that seems low.
and describe, in broad technical terms, how such a migration could be carried out. T
1. Format C:
2. Install freeware OS
3. Fart in the general direction of Redmond , WA
Unless the people that are allowed to survive "artificially" by technology or social help will in the end weaken the species.
Humans (and all other species) are as resilient and strong today as they are because all the weaker, less intelligent, defective samples were killed off or died as a result of their lesser condition.
There's several inherited diseases that are starting to have major increaseses in occourance. The reason? Before all this "everyone must live" crap, they died before they could breed. Now they are given all the medical care, drugs and support they need to survive against the rules of nature. The result is that these people breed and pass the genes/disease on the their offspring.
The children in turn may or may not exibit the disease, but they carry it and pass it on. Instead of fighting disiease by allowing it to die out naturally we are allowing (forcing) these people to live and instead spending billions to sove a problem that would solve itself if we just stopped trying to eliminate death.
But we've gotten way off the topic of VoIP here, so lets just end all this now and agree to disagree.
I'm by no means young.
Fact is that humans have survived quite nicely for quite a few thousands of years without a 911 system.
Mapquest has this feature called "trip planning". You can set "waypoints" on your trip.
I recently used the feature to tell the thing to give bme directions from Missouri to Phoenix, AZ via Flagstaff, AZ.
If I recall it can be used down to the street address level.
But now your getting away from "free" services. Someone has to pay for the hardware and the local circuits to the sponsoring telcos that the gatways use. Without local dialtone from a telco (for a fee of course) you can't call from VoIP to a POTS customer.
If you don't want to do that then someone has to pay to run new wires to each and every home/business.
I'm really getting tired of this incorrect focus on the fringe.
Fact: the RIAA did not subpoena a grandmother or a 12 year old girl. They sent subpoenas to ISPs in order to determine the identities of people who where using certain IP addresses. When RIAA filed these documents they had absoloutly NO idea of who the people where, their age, sex, gender, race, income, location or anything else but an IP address.
Out of 268 cases (??) so far we've had two "fringe" cases. One was a young girl. Would you be bitching about that if she'd been selling crack on the street and got busted? Even if she claimed to have thought it was okay because she was paying a policeman for the privlidge?
The other case was apparently a clerical error on the part of the ISP in providing the information to answer the sunpoena. This had nothing to do with the RIAA.
Peerhaps we should use the extreems to end other practices also: In many military actions around the world, the US military has killed non-combatants including 12 year old girls and grandmas. Would you argue that everything the US Military does is then invalidated and it should stop all such action now and in the future?
No? Then take a moment and re-think your position on RIAA.
If your child falls in the pool then you did something stupid by not watching your child, or not enclosing the pool to prevent such a thing. You're genes should not be propagated.
If someone breaks in to my house I am perfectly capable of handling the situation myself. In fact I have done so. Calling 911 does not prevent any damage, or injury to my person or family.
The police are not crime preventers or stoppers, they simply clean up afterward.
Natural selection works everywhere that technology doesn't interfere with it. Stupid people should be allowed to die when they do something really stupid.
This type of VoIP does not replace your standard telephone service. You can't call a telephone number with it, you can only "call" other people that use the same software that you are using to originate the call.
For instance... you can't use either service to call up the local take away shop to place an order, nor can you call you non-computer using grandmother.
You still need the landline to call other landline (or mobile) phones. You still need the landline for reliability.
To get a little off topic, 911 should be paid for by those who use it, not through my telephone bills.
On a deeper level, we as a society need to get off this "everyone needs to be rescuable at any moment at any cost" bandwagon. If we don't let the law of natural selection have any bearing on our evolution as a species, I fear we are doomed.
The problem I have with stories like this is that the calls aren't really free. You do have to have a rather high-speed internet connection to make these calls with any reasonable quality and reliability, and you have to pay that fee on top of your existing phone charges.
A major limitation is that you can only call your fiends who use the same "service". And they are for the most part defining "service" rather loosely, they're more like applications in software and hardware than a service. I know it's only on Mac now, but I'm curious why iChatAV from Apple is excluded from these types of comparisons. It does the same things, plus video and uses the AOL screen name
and buddy list infrastructure.
There's a reliability issue with VoIP, I for one will not cut my dial tone off until I have nearly 100% uptime on my net connection. In all my life I think there was one time (after a hurricane) that I picked up my telephone and did not hear a dial tone. I can't count how many minutes per month/year my net connection is down for one reason or another.
I also take issue with the statement "...They do illustrate, however, just how far VOIP has come - it's actually good enough to offer a viable alternative to existing phones.". I don't think it's the VoIP technology that's improved, I think it's the Internet's infrastructure that's improved. There's finally enough bandwidth that you don't need a lot of buffering to ensure packet delivery order to the audio decoders.
It's still possible and routine to get out-of-order delivery, but no-where as severe as it was even just two years ago.
I for one had a terribly hard time reading the graphs. Three problems in particular:
1)
I (like about 20% of the male population) am partially color deficient.
The graphs' colors where very hard for me to distinguish in the ledgend; what with the 1 pixel wide font and sample line/symbols. The colors chosen were all of about the same intensity/saturation also, so if you are color blind the entire graph would look all about the same level of gray. This problem is further compounded by all the graphs having different color assigments (Linux 2.6
2)
crowding. On a few of the graphs, there were differences in order of magnitude between groupings. Since this dataset was supposed to show the fastest, that would naturally be the most interesting data. You don't take photo finishes of the 7th and 8th finishers in an 8 horse race. Using a logarithmic scale for those charts would make the data easier to read, showing detail in the lower(faster) end and allowing the less significant higher(slower) data to "blend" in the curve.
3)
Point size. The symbols are too large for that many data points. This may be personal preference, but single points should be used for scatter graphs like these, not symbols. Symbols may mask other more subtle changes in the results because they overflow other points nearby. Similarly, make your legend samples substantial, a block of color is much more apreciated than a single pixel line, and bold your identifier text.
If you make graphs for the web, please take them in to Photoshop, GIMP, or something similar and reduce or remove the color data (convert to grey scale). The chart should still be at least fairly reable in greyscale.
Choose a wide range of Hue, Saturation and Lightness
If you are going to use light/desturated colors, use a black background. If you will use dark/heavily saturated colors, use a light background.
If you can't/won't do that, then please try to make the raw data available in CSV or some other "universal" format so we can create our own graphs.
Ulch - that meat was tainted! You feel deathly sick.
Bedlam?? If I recall, you put the pill in the meat and gave it to the dogs, who then fell asleep and you could get out of the room.
Yup, I'm glad at least a few others know of those channels and the web sites with the video clips.
DishNetwork carries both, along with I think one or two other university based channels. The Research Channel I think is Univerity of California's, UWTV is University of Washington. There may also be a Rutgers based channel.
On top of that I also regularly watch NASA TV which frequently does educational/lecture type shows, and they seem to show almost any testimony (that's not classified secret) by the administrator to Congress.
Discovery Science channel isn't too bad, and the National Geographic channel is great for Earth/nature based science and education. "Daily Planet" and "National Geographic Today" are two great news shows for finding out what's really going on the world besides just the local house fire, bank robbery, who's running for governor in a state where you don't live, etc.
And of course, the Weather Channel is great for learning about meteorology with some of the shows they have.
Perhaps the author of the article needs to spend a little more money and get a good satellite DBS, or a higher tier of service from his cable provider. I can't imagine all the "science" that I get on these channels being crammed in to just one channel called like "ultimate science".
I do understand, I've been baffled myself at the lack of 80s music on iTMS. Mind you, there is quite a bit, but not as much as you might think. The "80s crowd" would seem to be a prime audience for the iTMS, we're old enough to have enough disposable income, and sentimental enough about the music at this point to spend a butt load to DL it.
Of course, I've read post from about every genre about lack of selection: jazz, classical, soundtracks, metal, punk, indie... etc.
My guess is that the 80s will be getting filled in fairly soon and fairly completely now that iTMS has hit "critical mass".
You are 100% correct. Since launch the iTMS has been biased toward the music that the majority of those who are on-line purchase and listen to. You can apply the 80/20 rule to this in all probability; 80% of the people will tend to listen to about 20% of the music available.
Hence, when you launch a service, you make the most profit by first including the 20% of content that will encompass most of your sales. You later fill in the remainder to satisfy the others.
In the five or so months since iTMS has been on-line, they have grown from 200,000 songs to 400,000 songs, and this isn't the BuyMusic method of accounting, you can purchase each and every song Apple counts.
They've just started getting the indie labels on board and set up to submit tracks, 200 according to Steve's presentation. The indies were tripping over themselves to sign up when the meeting was held a few months back. Why wouldn't they? Apple provdides the hardware, submission is free, Apple handles all billing. Since each music company is apparently responsible for encoding and submitting the tracks, the rate of increase should be greater in the future. The next 200,000 tracks may well be added within the next two months.
There's plenty of "obscure stuff" on the store, or do you perhaps consider Andy Griffith's 'Fishin' Hole' to be main stream music?