As someone who troubbleshoots Macs for a living on a very large campus, I doubt that MacOS X would solve any problme with frequent crashes you may be expericening with 9. If you are having that many crashes there, it is probably because your Admins have no idea what they are doing... A new OS will not solve incompetence...
I am not sure where you are getting the "tends to trash people's systems" from, but I have been using 9 since early beta's (legitimate seeing), and have found that it works very well, with one exception: If you have ObjectSupportLib there. It is a file that was needed by pre sys 8 computers, and is detrimental to 8.0 and above (if you find it in the extensions folder, trash it). My guess would be that your lab there is full of ObjectSupportLib, and the admins have never learned that this is a problem... (this would eb like putting win3.11 dll's in a NT box...). The amazing thing is that the computers will still continue to work, with only intermitant problems even with this going on....
On the cost comment... you do get what you pay for.. And you can upgrade most any mac out there.. in every case memory and HD's.. and in most cases the processor and add any sort of PCI cards you want to (on PCI macs, of course.. NuBus or PDS on very-much older systems).. On some systems that shiped with PPC 601 processors (shiped against 486's and Pentium66's), you can go all the way to a G4 processor (of courece this would be silly, as the motherboard woudl starve the processor for bandwidth... somthing that most people forget about..). How exactly is this less upgradeable than the PC world? Just because people do it less does not mean it cannot be done....
It is "nice", but do you really want to go into the command line, and manually find each process and adjust the priority of each? As it stands in MacOS 9 the process that you have selected gets the most time, and "background" apps get less. I woudl love it if MacOS X does something like this automaticly, and then also has a GUI controll widget to allow power users to muck arround with the settings to tweak on this. And there is a good chance that Apple will do something like this... They have in the past...
The whole argument of Pre-Emptive vs. Cooperative multitasking boils down to this: Cooperative is (or can be...) more efficent, if (and only if!) every process is well written, this is why it was used in the orriginal MacOs, and why Apple has to use the whole Classic/Carbon/Cocca system to achive this (and other things too).
Pre-Emptive systems still work when things go to hell with poorly written apps, but does have quite a bit of overhead. With Moors law still very much in evidence, this is not as big an issue as it was ten years ago...
Neither system is inheriently better for everything, and anyone who claims otherwise is full of themselves! In a desktop operating system, I would prefer pre-emtive, but neither that aspect, or any other buzz-word complientcy will make or break a operating system.
In MacOS X Apple is going with their own imaging system (to call it a Window manager is simply not adequate), the 2D portion of which is called Quartz. Quartz uses a PDF-sytle system, and this creates a lot of great benifits:
Every object drawn to the screen through this method can have transparency values (as in a percentage), everything can be anti-aliased (PDF is resolution/dpi independant, so an inch is an inch whether you are at 72 dpi or 1600dpi.. wonderfull for printing things..). Rather than aproching the screen from the perspctive of pixels, now everything is a location in a floating point sence.. a revolutionary idea if you ask me...
The simple answers that you want to know are: This is Apples propritory work, the famimily jewls, there is NO good reason for them to open source this (and give people like MS the gun to kill them with)! Someone will probably write a way to get X11 emulated into Quartz, but because this system add so much new to the whole idea of imaging that I don't see someone writing a open way of getting this onto X11...
Just a minor corection, that was 17% converts from Windows, and 11% new computer buyers. Adding them up it is 28% were buyers new to the Macintosh Platform...
Version numbers are not decimals, but instead are tuple numbers. Ignore that it is a period that seperates the numbers, it could be a slash, a dash, or anything else. File Paths are another form of tuples, as are IP numbers...
There is actually a system that is supposed to be in place with the tuppol numbering system. The first number is supposed to be incremented only when there is a change to the file format that prevents the older versions from reading the newer's files. The second number is supposed to be for added features (that do not render older versions incompatible), the third is for bugfixes.
I remember all this from an old C book (as in before C++ was really arround), that I picked up when I was first starting to get into languages beond Apple-Basic, and TurtleGraphics (now those were happy days... *sigh*)
I on the other hand think that version numbers with a date in them are sometimes a nice thing. I don't have to think much about when something came out. I can easily tell that Office 98 can read Office 97's file format, but does not have a chance at Office2000's...
I am not saying that I want sendmail, or Apache, or anything else along those lines to go with this sort of a system. But for customer clarity (in the mass market), this is sometimes a nice thing.
I can assure you all that this is a nice feature for most laptop users. I use a PowerBook G3 Series (Macintosh laptop), that has this feature. I can chooose my power settings to give me the mix of performance and battery life that I want. As most of my work on the road is glorified text editing (scripting/developing), I let my processor cyle down, but keep the screen at full energy (punctuation is hard to read on a dimmed screen in high glare environs..).
I get to keep working for an hour longer on a single battery. And this does bring up one more note on this, whenevery you think about the battery life figures now, remember that they are baseing this on the most power-conserving settings...
OK.. if you had done your homework here, and looked throught the posts, it seems that everyone who is having this sort of problem is behind thir own firewall (or NAT system), and is therefore totally imune to this form of attack, and, if they had actually read the read-me, were not even told to install this.
Apple is putting out this patch, and recomending it only to people with dedicated internet connections (large-LAN or CableModemm custiomers), not to people with AirPort, or other Modems... I do agree that it should not affet these customers, but what do you expect for shoot-from-the-hip solution. If you are not affected, take the extension out of your system folder, and forget about it... when 9.1 comes out, this will be completely taken care of...
In the desktop side of things (meaning not enterprise-class-servers), graphics and video apps are the apps that will benifit most from access to multiple processors. In fact they don't usually care if those processors are SMP or AMP (asymetric), they just want as many processor cycles as they can get!
Think about applying a complex filter to a large image (a poster sixed image at 1600 DPI for example.. an especialy good example as almost all movie posters are done on Macs...), what the computer is doing in this render is applying a algorithm repeatedly to each point in the picture. Usually the algorithm only involves a small number of pixels for each pass, but it has to pass over every single pixel (sometimes more than once). If you have one processor, then that one has to do the whole job, add more hands to the work, and each one can be working on its own piece of the picture, not interfering with the others.
Since the people you pay for this sort of creative work tend to be paid a lot per hour, any speed-up you can provie in their tools saves you money (and generally makes the creative types happier people). SMP specificly is only mildly important to them (maybe they can work on email on one processor while the other grinds away..), but the MP part is absolutely critical!
...and just to respond ot the "Macophiles" part of the post, for Apple's graphics arts consumers this is essential, and woth any price. For the rest of us, the fact that they are working on SMP rather than AMP (finally!) means that the beinifits of MP can be put in at the system level, so that any app will instantly benifit from having multiple processors in the system, even if that app can only run on one of those processors!
There are a couple of problems here, ones of scale. The Human Genome Project simply has too much data to ever be put on a Distributed network. The data involved at each HGR center is in the Terabyte range, there is no way that even one center's data could be put onto CDs and distributed that way. The second problem is that, unlike SETI@Home's data, you cannot break this data up into relativly small packets for processing. The work that is currently being done at the computing centers is one of trying to fit the scraps of data together like jigsaw puzzel. Imagine the futility of randomly mailing 100 of your friends pieces of a puzzel. There is a chance that a ew of them could put together fragments, and help you solve the puzzel that way, but a much more effective way of doing this woudl be to bring all the participants together, and all work on the same huge bin of pieces. This is what they are doing rigth now. The project is not one that takes very complicated and prossesor intensive calulations, but that the data needs to me massaged en-masse, Terabyte by Terabyte! And then, once the baseline Human Genome is pieced together, then you start to figure out what all these genes do (the real work in this project). Then you start to pair up real people with the genes that they have, and to put individuals names or even refernce nummbers woudl be an serios invasion of privacy! Summary: the Human genome Project will never make a good canidate for Distributed Prossesing for both practical and privacy reasons!
High level languages are great for many things. I program in Java and Perl almost exclusively. But I would never write a real datatbase program, or a arcade style gaem, in either of those languages. High level languages in general, and interpreted langagues in particular are just not cut out for that sort of thing. Never mind when we start talking about writing OS's.
It is more than just the speed issue (although that is a big one!). There are things like compactness of code, etc.. In a low level language you have to work a lot harder to get anything done, but (if you are good, and have the right ideas), the code you come up with is far better than what you could produce in anythign else. Really good compilers/interpreters can help to narrow the difference, but..
Lesson to be learned here: Just like every other issue that splits the computer world, there are always tradeoffs. You just have to pick the tool that best fits the job!
Spin is mearly a graphical image, used to help you visualize a force that has no real analouge in the macroscopic universe. "Spin" is a fiction, but so far as modern science goes, the force/ide behind it is real.
If you are spending 20K on software.. I think you can do better than comsumer space prossesors (Intel, AMD, PPC G4, etc).. this is a definate case for the big boys like Alpha, PA-RISC, or IBM's high end versions of PPC, etc... This is just simply where those processors are far better, and more economical, than anything else out there.
And I do not want to hear anything about a Beowolf cluster, you simply cannot scale multiple boxes as well as you can a single box with a dedicated backplane like the big boys use...
"because they are not *true* Firewire drives. They are aftermarket modifications to IDE or SCSI hard drives."
This last shows a profound mis-understanding of the difference between IDE and SCSI drives. The Mechanisms are THE EXACT SAME! The only differnces are that the best of any run of a mechinism, and the newest technologies in drive production, go into the SCSI channel, where they are paired up with a SCSI controller card (SCSI bus talks to card, card talks to mechanism..).
All you have to do is look at a SCSI drive, and a IDE drive from the same vendor and look at them. The green board on the SCSI drive is more complicated (assuming that you are looking at older drives.. newer ones have more concetrated in the ASICS). In principal you caould take a SCSI drive rip off the controller card, and put on a IDE card. The same is genericly true for FireWire.
Now there is one differnce will all the FireWire drives that I have had to play with, they are all designed to be very rugged, and portable. That means that they are based on the same mechanisms used for laptop drives. This means they have sacraficed some speed, and a lot of cost, in order to be more rugged, and to be smaller (lower power too...). You can shake that VST drives all you want, while they are reading data, and they still keep rigth no going.. try that with a desktop drive!
Here at the University of Wisonsin the school of journalism has been streeming a live news show as a part of one of the journalism classes. I have been in close communication with the professor who has been doing the technical side of things.
After quite a bit of research into possible solutions that he could brodcast, without worrying about legal issues... he came to the conclusion that an Apple solution was the cheapest, most usefull solution.
They are using a 2 Mac solution, one to so the sorenson compression (this one is an older iMac running MacOS 8.6), and a second G3 (beige tower running MacOS X Server, and the free Quicktime Streaming Server). This solution is workign VERY well. As they have the show only for one hour durring the day, the setup is scritped so that the server starts up 5 minutes before the broadcast, straeams, and records, the broadcast at the right time, and then posts the recorded stream to a website that it modifies to suit, in case anyone missed the live broadcat.
I tend to agree that you don't need a G4 to do the job. A pair of iMacs woudl do you just fine. Buy a copy of Sorenson brodcaster and pop it on one of them, and either go with LinuxPPC and the Darwin Streaming Server (a little bit of work to get it set up), or go with MacOS X Server (my recomendation), and you have an instant setup. Very reliale, and you can also stream video if you choose to do so.
The only probem I would see is that there is currently no version of QuickTime 4 for Unix/Linux. Well.. there is the Java vrsion.. but.. For Windows amd MacOS users there will be no problems, and the streaming server can even be setup so that it will use MBONE, so that you don't trash the network if everyone happens ot be listning....
This would be an important note on a PC, but since Mac Users have gotten used to being able to boot off a almost-completely-funcitonal copy of the OS on a CD-ROM (ever since CD-ROM came out for the Mac...), the most important function of a floppy in the PC world, boot disks, is moot on the Mac side. Tom's Boot disk is a PC-Only nessecity! The function of sneaker-net is the other reason to have a floppy drive, mostly for e-mail purposes, but even those days are over with IMAP, and network sharing, any computer with email can get to network drives.... The days of sub-Gigabyte removable storage are gone, let them rest in pease!
There has been a lot of discussion about this whole deal between MS and AOL on the messanging issue, but to me the biggest point here has been missed! AOL is running the servers that allow the connection between users. MS's client connects to these servers, using them to connect the user to other users, while implicitly giving Microsoft the credit for doing so! MS cannot "innovate" (by this I mean change "standards" to keep out competition) because they do not provide ANY servers to change the standards on. If the servers do not recognize new "inovations" then the clients cannot use them! Right now AOL is the only one out there who is providing ANY servers (both the AIM and ICQ servers), and they are doing so for FREE!! No advertisements, no banners, no NOTHING! They are doing this because they thing that having their name out there on a product that you use daily will entice you to sign up with them as an ISP (I also think that eventually they will start advertising..). Now Microstoft is comming in and pirating their servers, using the same idea to advertise themselves, and giving AOL no credit, while still making AOL foot the bill for running the servers that make this all possible. In my mind this is dirty pool, and Microsoft is just using the standards buzzword as a smokescreen so that it can use its monopoly to crush its way into another sector of the internet. How long do you think it will be untill we see instant messaging as a "intigrated part of the operating system"?
Slashdot Mirror
As someone who troubbleshoots Macs for a living on a very large campus, I doubt that MacOS X would solve any problme with frequent crashes you may be expericening with 9. If you are having that many crashes there, it is probably because your Admins have no idea what they are doing... A new OS will not solve incompetence...
I am not sure where you are getting the "tends to trash people's systems" from, but I have been using 9 since early beta's (legitimate seeing), and have found that it works very well, with one exception: If you have ObjectSupportLib there. It is a file that was needed by pre sys 8 computers, and is detrimental to 8.0 and above (if you find it in the extensions folder, trash it). My guess would be that your lab there is full of ObjectSupportLib, and the admins have never learned that this is a problem... (this would eb like putting win3.11 dll's in a NT box...). The amazing thing is that the computers will still continue to work, with only intermitant problems even with this going on....
On the cost comment... you do get what you pay for.. And you can upgrade most any mac out there.. in every case memory and HD's.. and in most cases the processor and add any sort of PCI cards you want to (on PCI macs, of course.. NuBus or PDS on very-much older systems).. On some systems that shiped with PPC 601 processors (shiped against 486's and Pentium66's), you can go all the way to a G4 processor (of courece this would be silly, as the motherboard woudl starve the processor for bandwidth... somthing that most people forget about..). How exactly is this less upgradeable than the PC world? Just because people do it less does not mean it cannot be done....
It is "nice", but do you really want to go into the command line, and manually find each process and adjust the priority of each? As it stands in MacOS 9 the process that you have selected gets the most time, and "background" apps get less. I woudl love it if MacOS X does something like this automaticly, and then also has a GUI controll widget to allow power users to muck arround with the settings to tweak on this. And there is a good chance that Apple will do something like this... They have in the past...
The whole argument of Pre-Emptive vs. Cooperative multitasking boils down to this: Cooperative is (or can be...) more efficent, if (and only if!) every process is well written, this is why it was used in the orriginal MacOs, and why Apple has to use the whole Classic/Carbon/Cocca system to achive this (and other things too).
Pre-Emptive systems still work when things go to hell with poorly written apps, but does have quite a bit of overhead. With Moors law still very much in evidence, this is not as big an issue as it was ten years ago...
Neither system is inheriently better for everything, and anyone who claims otherwise is full of themselves! In a desktop operating system, I would prefer pre-emtive, but neither that aspect, or any other buzz-word complientcy will make or break a operating system.
In MacOS X Apple is going with their own imaging system (to call it a Window manager is simply not adequate), the 2D portion of which is called Quartz. Quartz uses a PDF-sytle system, and this creates a lot of great benifits:
Every object drawn to the screen through this method can have transparency values (as in a percentage), everything can be anti-aliased (PDF is resolution/dpi independant, so an inch is an inch whether you are at 72 dpi or 1600dpi.. wonderfull for printing things..). Rather than aproching the screen from the perspctive of pixels, now everything is a location in a floating point sence.. a revolutionary idea if you ask me...
The simple answers that you want to know are: This is Apples propritory work, the famimily jewls, there is NO good reason for them to open source this (and give people like MS the gun to kill them with)! Someone will probably write a way to get X11 emulated into Quartz, but because this system add so much new to the whole idea of imaging that I don't see someone writing a open way of getting this onto X11...
Just a minor corection, that was 17% converts from Windows, and 11% new computer buyers. Adding them up it is 28% were buyers new to the Macintosh Platform...
Version numbers are not decimals, but instead are tuple numbers. Ignore that it is a period that seperates the numbers, it could be a slash, a dash, or anything else. File Paths are another form of tuples, as are IP numbers...
There is actually a system that is supposed to be in place with the tuppol numbering system. The first number is supposed to be incremented only when there is a change to the file format that prevents the older versions from reading the newer's files. The second number is supposed to be for added features (that do not render older versions incompatible), the third is for bugfixes.
I remember all this from an old C book (as in before C++ was really arround), that I picked up when I was first starting to get into languages beond Apple-Basic, and TurtleGraphics (now those were happy days... *sigh*)
I on the other hand think that version numbers with a date in them are sometimes a nice thing. I don't have to think much about when something came out. I can easily tell that Office 98 can read Office 97's file format, but does not have a chance at Office2000's...
I am not saying that I want sendmail, or Apache, or anything else along those lines to go with this sort of a system. But for customer clarity (in the mass market), this is sometimes a nice thing.
I can assure you all that this is a nice feature for most laptop users. I use a PowerBook G3 Series (Macintosh laptop), that has this feature. I can chooose my power settings to give me the mix of performance and battery life that I want. As most of my work on the road is glorified text editing (scripting/developing), I let my processor cyle down, but keep the screen at full energy (punctuation is hard to read on a dimmed screen in high glare environs..).
I get to keep working for an hour longer on a single battery. And this does bring up one more note on this, whenevery you think about the battery life figures now, remember that they are baseing this on the most power-conserving settings...
OK.. if you had done your homework here, and looked throught the posts, it seems that everyone who is having this sort of problem is behind thir own firewall (or NAT system), and is therefore totally imune to this form of attack, and, if they had actually read the read-me, were not even told to install this.
Apple is putting out this patch, and recomending it only to people with dedicated internet connections (large-LAN or CableModemm custiomers), not to people with AirPort, or other Modems... I do agree that it should not affet these customers, but what do you expect for shoot-from-the-hip solution. If you are not affected, take the extension out of your system folder, and forget about it... when 9.1 comes out, this will be completely taken care of...
In the desktop side of things (meaning not enterprise-class-servers), graphics and video apps are the apps that will benifit most from access to multiple processors. In fact they don't usually care if those processors are SMP or AMP (asymetric), they just want as many processor cycles as they can get!
Think about applying a complex filter to a large image (a poster sixed image at 1600 DPI for example.. an especialy good example as almost all movie posters are done on Macs...), what the computer is doing in this render is applying a algorithm repeatedly to each point in the picture. Usually the algorithm only involves a small number of pixels for each pass, but it has to pass over every single pixel (sometimes more than once). If you have one processor, then that one has to do the whole job, add more hands to the work, and each one can be working on its own piece of the picture, not interfering with the others.
Since the people you pay for this sort of creative work tend to be paid a lot per hour, any speed-up you can provie in their tools saves you money (and generally makes the creative types happier people). SMP specificly is only mildly important to them (maybe they can work on email on one processor while the other grinds away..), but the MP part is absolutely critical!
...and just to respond ot the "Macophiles" part of the post, for Apple's graphics arts consumers this is essential, and woth any price. For the rest of us, the fact that they are working on SMP rather than AMP (finally!) means that the beinifits of MP can be put in at the system level, so that any app will instantly benifit from having multiple processors in the system, even if that app can only run on one of those processors!
There are a couple of problems here, ones of scale. The Human Genome Project simply has too much data to ever be put on a Distributed network. The data involved at each HGR center is in the Terabyte range, there is no way that even one center's data could be put onto CDs and distributed that way.
The second problem is that, unlike SETI@Home's data, you cannot break this data up into relativly small packets for processing. The work that is currently being done at the computing centers is one of trying to fit the scraps of data together like jigsaw puzzel. Imagine the futility of randomly mailing 100 of your friends pieces of a puzzel. There is a chance that a ew of them could put together fragments, and help you solve the puzzel that way, but a much more effective way of doing this woudl be to bring all the participants together, and all work on the same huge bin of pieces. This is what they are doing rigth now. The project is not one that takes very complicated and prossesor intensive calulations, but that the data needs to me massaged en-masse, Terabyte by Terabyte!
And then, once the baseline Human Genome is pieced together, then you start to figure out what all these genes do (the real work in this project). Then you start to pair up real people with the genes that they have, and to put individuals names or even refernce nummbers woudl be an serios invasion of privacy!
Summary: the Human genome Project will never make a good canidate for Distributed Prossesing for both practical and privacy reasons!
High level languages are great for many things. I program in Java and Perl almost exclusively. But I would never write a real datatbase program, or a arcade style gaem, in either of those languages. High level languages in general, and interpreted langagues in particular are just not cut out for that sort of thing. Never mind when we start talking about writing OS's.
It is more than just the speed issue (although that is a big one!). There are things like compactness of code, etc.. In a low level language you have to work a lot harder to get anything done, but (if you are good, and have the right ideas), the code you come up with is far better than what you could produce in anythign else. Really good compilers/interpreters can help to narrow the difference, but..
Lesson to be learned here: Just like every other issue that splits the computer world, there are always tradeoffs. You just have to pick the tool that best fits the job!
Spin is mearly a graphical image, used to help you visualize a force that has no real analouge in the macroscopic universe. "Spin" is a fiction, but so far as modern science goes, the force/ide behind it is real.
If you are spending 20K on software.. I think you can do better than comsumer space prossesors (Intel, AMD, PPC G4, etc).. this is a definate case for the big boys like Alpha, PA-RISC, or IBM's high end versions of PPC, etc... This is just simply where those processors are far better, and more economical, than anything else out there.
And I do not want to hear anything about a Beowolf cluster, you simply cannot scale multiple boxes as well as you can a single box with a dedicated backplane like the big boys use...
"because they are not *true* Firewire drives. They are aftermarket modifications to IDE or SCSI hard drives."
This last shows a profound mis-understanding of the difference between IDE and SCSI drives. The Mechanisms are THE EXACT SAME! The only differnces are that the best of any run of a mechinism, and the newest technologies in drive production, go into the SCSI channel, where they are paired up with a SCSI controller card (SCSI bus talks to card, card talks to mechanism..).
All you have to do is look at a SCSI drive, and a IDE drive from the same vendor and look at them. The green board on the SCSI drive is more complicated (assuming that you are looking at older drives.. newer ones have more concetrated in the ASICS). In principal you caould take a SCSI drive rip off the controller card, and put on a IDE card. The same is genericly true for FireWire.
Now there is one differnce will all the FireWire drives that I have had to play with, they are all designed to be very rugged, and portable. That means that they are based on the same mechanisms used for laptop drives. This means they have sacraficed some speed, and a lot of cost, in order to be more rugged, and to be smaller (lower power too...). You can shake that VST drives all you want, while they are reading data, and they still keep rigth no going.. try that with a desktop drive!
Here at the University of Wisonsin the school of journalism has been streeming a live news show as a part of one of the journalism classes. I have been in close communication with the professor who has been doing the technical side of things.
After quite a bit of research into possible solutions that he could brodcast, without worrying about legal issues... he came to the conclusion that an Apple solution was the cheapest, most usefull solution.
They are using a 2 Mac solution, one to so the sorenson compression (this one is an older iMac running MacOS 8.6), and a second G3 (beige tower running MacOS X Server, and the free Quicktime Streaming Server). This solution is workign VERY well. As they have the show only for one hour durring the day, the setup is scritped so that the server starts up 5 minutes before the broadcast, straeams, and records, the broadcast at the right time, and then posts the recorded stream to a website that it modifies to suit, in case anyone missed the live broadcat.
I tend to agree that you don't need a G4 to do the job. A pair of iMacs woudl do you just fine. Buy a copy of Sorenson brodcaster and pop it on one of them, and either go with LinuxPPC and the Darwin Streaming Server (a little bit of work to get it set up), or go with MacOS X Server (my recomendation), and you have an instant setup. Very reliale, and you can also stream video if you choose to do so.
The only probem I would see is that there is currently no version of QuickTime 4 for Unix/Linux. Well.. there is the Java vrsion.. but.. For Windows amd MacOS users there will be no problems, and the streaming server can even be setup so that it will use MBONE, so that you don't trash the network if everyone happens ot be listning....
That is my 2 cents worth.. good luck!
This would be an important note on a PC, but since Mac Users have gotten used to being able to boot off a almost-completely-funcitonal copy of the OS on a CD-ROM (ever since CD-ROM came out for the Mac...), the most important function of a floppy in the PC world, boot disks, is moot on the Mac side.
Tom's Boot disk is a PC-Only nessecity! The function of sneaker-net is the other reason to have a floppy drive, mostly for e-mail purposes, but even those days are over with IMAP, and network sharing, any computer with email can get to network drives....
The days of sub-Gigabyte removable storage are gone, let them rest in pease!
There has been a lot of discussion about this whole deal between MS and AOL on the messanging issue, but to me the biggest point here has been missed! AOL is running the servers that allow the connection between users. MS's client connects to these servers, using them to connect the user to other users, while implicitly giving Microsoft the credit for doing so!
MS cannot "innovate" (by this I mean change "standards" to keep out competition) because they do not provide ANY servers to change the standards on. If the servers do not recognize new "inovations" then the clients cannot use them! Right now AOL is the only one out there who is providing ANY servers (both the AIM and ICQ servers), and they are doing so for FREE!! No advertisements, no banners, no NOTHING! They are doing this because they thing that having their name out there on a product that you use daily will entice you to sign up with them as an ISP (I also think that eventually they will start advertising..).
Now Microstoft is comming in and pirating their servers, using the same idea to advertise themselves, and giving AOL no credit, while still making AOL foot the bill for running the servers that make this all possible. In my mind this is dirty pool, and Microsoft is just using the standards buzzword as a smokescreen so that it can use its monopoly to crush its way into another sector of the internet. How long do you think it will be untill we see instant messaging as a "intigrated part of the operating system"?