I still rate VMS as the overall best OS I've ever used, including various Unix variants and Mac OS X/OS X Server. From an administrator's perspective, VMS had the same degree of user-facing consistency that Mac OS has, along with a fine-grained protection model that I'm appalled has never been deployed since. This is what Mac OS X Server should evolve to. And there are times when I really miss VMS' file versions.
The single biggest shortfall of VMS was that it was really hard to set up pipe-and-filter kinds of things, although it was doable. A good Unix shell on VMS would have been my idea of Nirvana.
I've never understood how Win NT was so bad when compared to VMS; either Dave Cutler lost it, or he was crippled with Windows compatibility handcuffs.
It's clear from my experience that you need a competent IT staff to run a network of Microsoft machines.
It's also clear from my experience that a reasonably intelligent group of Mac users do NOT need the same level of help. That's not to say they never need "professional experience," rather to point out that a single trained Mac IT support person takes care of a LOT more installations than a trained Windows IT support person. In the company I used to work for, I think that number was about 25-1; there were 2 Mac people supporting an installed Mac user base of several hundred in a department of, I don't know, 25-50 maybe for a Windows installed base of several thousand. Now some of those people did servers, routers, etc, and not just desktops.
Running a server, whether Windows, Mac OS X Server or Linux, requires a deeper level of training, experience and time investment.
It's truly appalling that the great number of discussions are either (a) ignorant of the question of 'host vulnerability', (b) assume that everyone is running Windows; or (c) can't be bothered to determine what hosts are vulnerable. If I were sufficiently paranoid, I'd believe this is part of the continuing conspiracy to make everyone believe that such vulnerabilities are a 'fact of life' for all computers, and not just Microsoft products.
Of course, on a military system I saw something in the software requirements about how the system software had to protect the computer from kinetic penetration. I asked if having the software raise Shields was an adequate software solution (leaving it to the hardware people to actually build those Shields: "Implementing this API in hardware is not a software problem....")
In the same document, by the way, was this tid-bit under environmental considerations: "The software shall recycle bits and shall make use of recycled bits to the maximum extent possible." In This Case, I found who wrote that requirement, and he admitted it was a joke he inserted to see if anyone was actually reading it. No owner for the 'kinetic damage' requirement was ever found (it was probably cut-and-paste from an unrelated system specification.)
If all you want to do is learn to write little individual programs, then join a coding club. Don't claim that's an education, or even effective training. Producing software systems bigger than toy applications requires a much broader knowledge of technology and social factors.
A credible computer science education should include courses in the physical sciences (a lot of what I've done in 30 years in this business is based on physics), probably economics, effective communications both written and oral, and the ability to deal with people with a wide variety of backgrounds. In short, the kinds of things you get from going to a college for an on-site location. Distance learning works OK for advanced degrees, but even then there's an emphasis on interactions with the professor and peers, and the best programs also have residency requirements.
I should pay 50,000 a year to learn how to socialize? Why can't I just go to a park or take up Yoga?
Depends on what you're expecting from your education. If you're going to be a Park Ranger or a Yoga Instructor, maybe that's OK. If you are going to deal with people under any other terms, that's probably not OK. Hell, I learned as much by being "hall tutor" for classmates talking Calc, as I learned in the actual class listening to the professor.
Working with others in classrooms, give-and-take with professors, ability to read body language and non-verbal clues, etc. Having spent a lot of the last 8 years in telecons and VTCs, I know what you lose, particularly when you don't have a lot of face-time with that group.
But then, consider how Microsoft might be different if Gates & Ballmer weren't together at Harvard...
Tom Clancy tells the story about security review of "Hunt for Red October" (published by Naval Institute Press, they routinely send stuff to the Navy just to be sure.) The review came back, "Can't publish, contains classified information." "Well tell me what that is, I'll remove it, and we'll be good to go." "No, sir. You don't have the clearance for that information."
After a couple back-and-forth, apparently Clancy went over his book, line-by-line, justifying everything in there as derived from open source (in the Intel sense, i.e. freely available from the press, unclassified technical reports, etc.). Eventually the Navy had to admit that, if there was something classified in there, it was derived from stuff that anyone could read and deduce on his/her own.
Yossarian is alive and well, it appears... (http://en.wikipedia.org/wiki/Catch-22)
http://magnatune.com/ has already been mentioned, they have some good Early Music recordings. Check out the folk stuff, too, there's some good crossover material.
I'm a big fan of KUSC (http://www.kusc.org), particularly their evening show. Jim Svejda does some pretty obscure stuff at times, and I find his interviews and commentary particularly insightful. Too bad he's on so late on east coast time. Also for really avant-garde stuff, there's John Schaefer's New Sounds on WQXR (turned me onto minimalism back in the early 80s.)
But at home, I've ripped my CD collection and stream via iTunes, Airport Express/AirTunes and an FM transmitter with enough power to reach the entire house and yard. I'm into Early Music and about half my stuff is Medieval or Renaissance. There's nothing like Krummhorns in the Morning:-)
RAID enclosure, OWC Qx4, to be specific. I have 4x1TB drives in a hardware RAID 5 configuration (3TB from the enclosure exposed through a single FW800 port.) This provides file server for my SOHO (4 machine) network. A Mini server costs $1k (less if you go for an Apple refurb) and the Qx4 set me back about $600. The Mini and Qx4 don't make much noise or generate much heat (important in a small home office.)
That enclosure, by the way, gives good performance but is sensitive to the drives. OWC insists you put 4 matched drives in it, and I previously had 1TB Seagate drives, that gave me multiple drive failures. It's been running quite happily with 1TB Toshiba drives. I'll not buy any more Seagate drives, this isn't my first bad experience with that brand.
Frankly, you might want to just decommission this machine. Those G5 CPUs and the associated fans draw a lot of power and throw a lot of heat. Replacing my G5 with a Mini (as a file server) produced a negligible drop in performance as a file server (OS X Leopard Server) by using FW800 drives with hardware RAID instead of SATA drives with MacOS X Software RAID. But the temperature in my home office dropped by a couple of degrees once I turned that G5 off. (It tossed an amazing amount of heat, even when it wasn't running full blast.)
The one thing that G5 was -really good at- was ripping CDs, because PowerPC G5's floating point performance is generally a lot faster than Intel's... I'd suspect you'd get similar performance for video and other floating-point intensive operations.
Maybe they won't have to carry all of the extra batteries, which will help.
Seriously*, you've hit a significant advantage. The normal 'basic load' of batteries is 30 days, from what I remember, and carrying those batteries around, particularly in Ranger/light infantry/Special Ops units is a tremendous drag. Whatever technique gets used, though, has to account for the fact that light infantry soldiers spend a lot of time being still (because what moves can be seen, and what can be seen will be shot...), so either you need some technique to store the power or you need something that can generate some amount of power when the soldier is not moving.
If you want (a) power; (b) biometric identification; (c) biological status/health monitoring, consider the Rectal Thermocouple... This will normally generate substantial additional power in combat as an added benefit.:-)
4 drive bay, USB, FW400/FW800 and eSATA. Will take 2tb drives, RAID 0, 1, 5 and 10. Comes pre-populated or unpopulated, the latter is what I got and added my own drives. http://www.macsales.com/ No financial connection, just a satisfied customer (they have great tech support!)
This is obviously not a build-it-yourself storage array, but is a good option if you want a commercial out of the box solution.
Actually, the total estimate from the repair shop was $1600, $1k for the part. The labor included some sort of reprogramming/adaptation.
And with respect to the R&D costs, why should those be born by the repair part? Isn't that factored into the costs of the vehicle? You know that's the exact same argument used by government contractors for the allocation of R&D and overhead costs over repair parts that yields $1000 hammers and $100 screws.
ABS warning light came on in my 2000 Nissan Frontier. They traced the fault to the ABS control module, and the replacement part is $1000!!! That's an appalling amount of money for a couple bucks worth of silicon!
I'm coming to the conclusion that there needs to be industrial or even government standards for computer security, and there ought to be an investigation on the price of (safety related) repair parts.
1. The languages being considered/charted are ASP, ASPX, CFM, DO, JSP, PHP and PL (I can guess at most of these acronyms).
What's missing, obviously, are 'real' programming languages such as C, Java, FORTRAN, Ada, C++, Eiffel, etc.
2. A lot of these languages share a common (C) heritage, and I'd assert "inherit" a lot of the security weaknesses of C. That's particularly true of weak typing for scalars, including array bounds.
The conclusion I think can be drawn from this is that we need a substantial increase in Web Programming practices, including languages. Any other conclusion is overreach.
"Crippled?" because it doesn't run Flash? By this definition, I've crippled my laptop by installing flash blockers, and you know, I think I like this "crippled" Web A LOT better. Sure, occasionally, I decide I want to see some video on CNN.com, and it is nice to be able to override the Flash blocker. But I don't miss all those dumb-assed Flash-based ads one bit.
And when I go to a website that uses only Flash, I think twice about whether this is a company/place I really want to be. As often as not, if there's no "non-Flash" version, I'll just navigate away. Restaurants, in particular, need to understand that all that glitzy Flash stuff is at best annoying to a lot of people, and at worse just does NOT WORK on mobile devices (not just the iPhone!!). You'd think restaurants in particular would want to encourage mobile customers; the onus is on them to make it easier for me to decide where I want to eat.
I think there are -4- different threads going on here:
1. The 'whose standards/proprietary world do you like better?' debate between Adobe and Apple, Flash & HTML5 (and its own CODEC wars)
2. The 'what kind of rich content is important?' debate - is this really "all about video" as some have suggested, or is it about arbitrary rich content?
3. The 'cross-platform' vs 'optimized for this device' debate (I think this is a really important debate for techies.)
4. The business decisions about how to best reach customers, along with the customer decisions about what technologies are acceptable (i.e. how far would Flash or JavaScript or HTML5 animations go before they become really annoying)?
When you have to resort to lawyers to force someone to use your stuff...
(I've been mostly Flash-free, except for the occasional CNN.com video, for about 18 months now. It's made browsing faster, more reliable particularly in terms of virtual memory usage - Flash leaks storage like a sieve, and I don't miss cheesy animations one bit!)
Slashdot Mirror
A browser that is specifically set up to completely firewall websites from each other?
I still rate VMS as the overall best OS I've ever used, including various Unix variants and Mac OS X/OS X Server. From an administrator's perspective, VMS had the same degree of user-facing consistency that Mac OS has, along with a fine-grained protection model that I'm appalled has never been deployed since. This is what Mac OS X Server should evolve to. And there are times when I really miss VMS' file versions.
The single biggest shortfall of VMS was that it was really hard to set up pipe-and-filter kinds of things, although it was doable. A good Unix shell on VMS would have been my idea of Nirvana.
I've never understood how Win NT was so bad when compared to VMS; either Dave Cutler lost it, or he was crippled with Windows compatibility handcuffs.
It's clear from my experience that you need a competent IT staff to run a network of Microsoft machines.
It's also clear from my experience that a reasonably intelligent group of Mac users do NOT need the same level of help. That's not to say they never need "professional experience," rather to point out that a single trained Mac IT support person takes care of a LOT more installations than a trained Windows IT support person. In the company I used to work for, I think that number was about 25-1; there were 2 Mac people supporting an installed Mac user base of several hundred in a department of, I don't know, 25-50 maybe for a Windows installed base of several thousand. Now some of those people did servers, routers, etc, and not just desktops.
Running a server, whether Windows, Mac OS X Server or Linux, requires a deeper level of training, experience and time investment.
But of course, I had to dig to find that particular piece of information. Most of the write-ups ignore the question of what host OS/systems are vulnerable. http://securityresponse.symantec.com/security_response/writeup.jsp?docid=2010-013112-4647-99
It's truly appalling that the great number of discussions are either (a) ignorant of the question of 'host vulnerability', (b) assume that everyone is running Windows; or (c) can't be bothered to determine what hosts are vulnerable. If I were sufficiently paranoid, I'd believe this is part of the continuing conspiracy to make everyone believe that such vulnerabilities are a 'fact of life' for all computers, and not just Microsoft products.
Actually, I believe it was "killall -45"
Of course, on a military system I saw something in the software requirements about how the system software had to protect the computer from kinetic penetration. I asked if having the software raise Shields was an adequate software solution (leaving it to the hardware people to actually build those Shields: "Implementing this API in hardware is not a software problem....")
In the same document, by the way, was this tid-bit under environmental considerations: "The software shall recycle bits and shall make use of recycled bits to the maximum extent possible." In This Case, I found who wrote that requirement, and he admitted it was a joke he inserted to see if anyone was actually reading it. No owner for the 'kinetic damage' requirement was ever found (it was probably cut-and-paste from an unrelated system specification.)
Too bad we can't mod up the title.
Or maybe Nancy Pelosi caused it...
It all depends on which wing of the bird you sit on.
If all you want to do is learn to write little individual programs, then join a coding club. Don't claim that's an education, or even effective training. Producing software systems bigger than toy applications requires a much broader knowledge of technology and social factors.
A credible computer science education should include courses in the physical sciences (a lot of what I've done in 30 years in this business is based on physics), probably economics, effective communications both written and oral, and the ability to deal with people with a wide variety of backgrounds. In short, the kinds of things you get from going to a college for an on-site location. Distance learning works OK for advanced degrees, but even then there's an emphasis on interactions with the professor and peers, and the best programs also have residency requirements.
I should pay 50,000 a year to learn how to socialize? Why can't I just go to a park or take up Yoga?
Depends on what you're expecting from your education. If you're going to be a Park Ranger or a Yoga Instructor, maybe that's OK. If you are going to deal with people under any other terms, that's probably not OK. Hell, I learned as much by being "hall tutor" for classmates talking Calc, as I learned in the actual class listening to the professor.
Working with others in classrooms, give-and-take with professors, ability to read body language and non-verbal clues, etc. Having spent a lot of the last 8 years in telecons and VTCs, I know what you lose, particularly when you don't have a lot of face-time with that group.
But then, consider how Microsoft might be different if Gates & Ballmer weren't together at Harvard...
Tom Clancy tells the story about security review of "Hunt for Red October" (published by Naval Institute Press, they routinely send stuff to the Navy just to be sure.) The review came back, "Can't publish, contains classified information." "Well tell me what that is, I'll remove it, and we'll be good to go." "No, sir. You don't have the clearance for that information."
After a couple back-and-forth, apparently Clancy went over his book, line-by-line, justifying everything in there as derived from open source (in the Intel sense, i.e. freely available from the press, unclassified technical reports, etc.). Eventually the Navy had to admit that, if there was something classified in there, it was derived from stuff that anyone could read and deduce on his/her own.
Yossarian is alive and well, it appears... (http://en.wikipedia.org/wiki/Catch-22)
http://magnatune.com/ has already been mentioned, they have some good Early Music recordings. Check out the folk stuff, too, there's some good crossover material.
I'm a big fan of KUSC (http://www.kusc.org), particularly their evening show. Jim Svejda does some pretty obscure stuff at times, and I find his interviews and commentary particularly insightful. Too bad he's on so late on east coast time. Also for really avant-garde stuff, there's John Schaefer's New Sounds on WQXR (turned me onto minimalism back in the early 80s.)
But at home, I've ripped my CD collection and stream via iTunes, Airport Express/AirTunes and an FM transmitter with enough power to reach the entire house and yard. I'm into Early Music and about half my stuff is Medieval or Renaissance. There's nothing like Krummhorns in the Morning :-)
RAID enclosure, OWC Qx4, to be specific. I have 4x1TB drives in a hardware RAID 5 configuration (3TB from the enclosure exposed through a single FW800 port.) This provides file server for my SOHO (4 machine) network. A Mini server costs $1k (less if you go for an Apple refurb) and the Qx4 set me back about $600. The Mini and Qx4 don't make much noise or generate much heat (important in a small home office.)
That enclosure, by the way, gives good performance but is sensitive to the drives. OWC insists you put 4 matched drives in it, and I previously had 1TB Seagate drives, that gave me multiple drive failures. It's been running quite happily with 1TB Toshiba drives. I'll not buy any more Seagate drives, this isn't my first bad experience with that brand.
Frankly, you might want to just decommission this machine. Those G5 CPUs and the associated fans draw a lot of power and throw a lot of heat. Replacing my G5 with a Mini (as a file server) produced a negligible drop in performance as a file server (OS X Leopard Server) by using FW800 drives with hardware RAID instead of SATA drives with MacOS X Software RAID. But the temperature in my home office dropped by a couple of degrees once I turned that G5 off. (It tossed an amazing amount of heat, even when it wasn't running full blast.)
The one thing that G5 was -really good at- was ripping CDs, because PowerPC G5's floating point performance is generally a lot faster than Intel's... I'd suspect you'd get similar performance for video and other floating-point intensive operations.
dave
Maybe they won't have to carry all of the extra batteries, which will help.
Seriously*, you've hit a significant advantage. The normal 'basic load' of batteries is 30 days, from what I remember, and carrying those batteries around, particularly in Ranger/light infantry/Special Ops units is a tremendous drag. Whatever technique gets used, though, has to account for the fact that light infantry soldiers spend a lot of time being still (because what moves can be seen, and what can be seen will be shot...), so either you need some technique to store the power or you need something that can generate some amount of power when the soldier is not moving.
* for humor on this topic, see my other post ;-)
If you want (a) power; (b) biometric identification; (c) biological status/health monitoring, consider the Rectal Thermocouple... This will normally generate substantial additional power in combat as an added benefit. :-)
4 drive bay, USB, FW400/FW800 and eSATA. Will take 2tb drives, RAID 0, 1, 5 and 10. Comes pre-populated or unpopulated, the latter is what I got and added my own drives. http://www.macsales.com/ No financial connection, just a satisfied customer (they have great tech support!)
This is obviously not a build-it-yourself storage array, but is a good option if you want a commercial out of the box solution.
Actually, the total estimate from the repair shop was $1600, $1k for the part. The labor included some sort of reprogramming/adaptation.
And with respect to the R&D costs, why should those be born by the repair part? Isn't that factored into the costs of the vehicle? You know that's the exact same argument used by government contractors for the allocation of R&D and overhead costs over repair parts that yields $1000 hammers and $100 screws.
ABS warning light came on in my 2000 Nissan Frontier. They traced the fault to the ABS control module, and the replacement part is $1000!!! That's an appalling amount of money for a couple bucks worth of silicon!
I'm coming to the conclusion that there needs to be industrial or even government standards for computer security, and there ought to be an investigation on the price of (safety related) repair parts.
http://www.defense.gov/speeches/speech.aspx?speechid=1467
I found a lot of the media coverage to be selective, and the headline on this /. posting to be somewhat misleading
1. The languages being considered/charted are ASP, ASPX, CFM, DO, JSP, PHP and PL (I can guess at most of these acronyms).
What's missing, obviously, are 'real' programming languages such as C, Java, FORTRAN, Ada, C++, Eiffel, etc.
2. A lot of these languages share a common (C) heritage, and I'd assert "inherit" a lot of the security weaknesses of C. That's particularly true of weak typing for scalars, including array bounds.
The conclusion I think can be drawn from this is that we need a substantial increase in Web Programming practices, including languages. Any other conclusion is overreach.
"Crippled?" because it doesn't run Flash? By this definition, I've crippled my laptop by installing flash blockers, and you know, I think I like this "crippled" Web A LOT better. Sure, occasionally, I decide I want to see some video on CNN.com, and it is nice to be able to override the Flash blocker. But I don't miss all those dumb-assed Flash-based ads one bit.
And when I go to a website that uses only Flash, I think twice about whether this is a company/place I really want to be. As often as not, if there's no "non-Flash" version, I'll just navigate away. Restaurants, in particular, need to understand that all that glitzy Flash stuff is at best annoying to a lot of people, and at worse just does NOT WORK on mobile devices (not just the iPhone!!). You'd think restaurants in particular would want to encourage mobile customers; the onus is on them to make it easier for me to decide where I want to eat.
I think there are -4- different threads going on here:
1. The 'whose standards/proprietary world do you like better?' debate between Adobe and Apple, Flash & HTML5 (and its own CODEC wars)
2. The 'what kind of rich content is important?' debate - is this really "all about video" as some have suggested, or is it about arbitrary rich content?
3. The 'cross-platform' vs 'optimized for this device' debate (I think this is a really important debate for techies.)
4. The business decisions about how to best reach customers, along with the customer decisions about what technologies are acceptable (i.e. how far would Flash or JavaScript or HTML5 animations go before they become really annoying)?
When you have to resort to lawyers to force someone to use your stuff...
(I've been mostly Flash-free, except for the occasional CNN.com video, for about 18 months now. It's made browsing faster, more reliable particularly in terms of virtual memory usage - Flash leaks storage like a sieve, and I don't miss cheesy animations one bit!)