Mind you, the word 'free' here is not meant 'free' as in free beer, but 'free' as in free speech.
What you're paying for at the doctor is not the disease information, because you have free access to all the information the doctors has in the freely accessible university libraries. When paying the doctor, you're paying for an *expert* to handle the information.
"it creates a situation in which an IT staffer may make changes that no one else knows about"
Translation: "Somebody may do something undocumented". What a moot point. As if only Linux is vulnerable to something like that.
Even then, if you don't want to differ from the crowd (read: if you prefer mass-produce jeans to tailor-made clothes), tell your staffers to use only original media found in shrink-wrap software, which is definitely abundantly available for Linux... (and watch your staffers quit their jobs to work for the wiser competitor).
"Having somebody who can screw around with my operating system would make me very, very nervous"
Can you say ILOVEYOU or Back Orifice?, Mr Firstbrook.
This guy is not an 'analists', he's nervous. He's not nervous about any perceived vulerability of Linux, but he's nervous about Linux period. Most of us here understand why...
If any OS allows people to screw around, it's the other OS, definitely not Linux.
-- HOWTO Change your Linux PC into a Windows PC: "chmod -R a+rwxs / && chown -R root.root / && echo guest::0:0::/tmp:/bin/bash >>/tmp/passwd"
"but you can only read the odd or even whatevers at a time."
I don't see why that would have to cause any more than a tiny slowdown in any half-decent RAMBUS interface implementation. It can be dealt with in the chipset with a simple permutation of the address lines (make the address line that changes on the odd/even boundaries the MSB), which does not cost any hardware. Then the only place where the 'bank switch' occurs is at 50% capacity (MSB bit flip).
I can imagine that any decent RAMBUS interfacing ASIC does (would do) such a thing.
Not true, heat is the dissipation of power. On Ohms loads (load with linear resistance), this is current multiplied by voltage (P=VI). The voltage I'm talking about here is the voltage drop on the path that the current is following, which is the supply voltage of the CPU.
That's why heat (power) goes up squared when voltage is increased with constant resistance (P=VI & V=IR, P=(I*I*R))
In addition to that, I can make some educated guesses: part of the current is the CMOS gate current, and part of the current is charging and discarging of capacitance.
The CMOS gate current is quite linear with respect to the clock frequency and the voltage applied between source and drain, but the capacitance charging and discharging current increases with higher clock frequencies.
Hence when overclocking, the heat dissipation goes up more than squared the increase of supply voltage, because the load is not Ohms, and you're increasing the charging current with your higher clock frequency.
I think AMD is currently making most of it's money on the 700-800Mhz athlons, not on the 1Gz version, which at it's price premium is mostly for careless spenders anyway (at least in my application area, the extra money is better spent on SDRAMS, bless the Abit KA7).
... depending on the market conditions and current yield, they might even stock up on a couple more 1Ghz versions, which they will then sell later when the improved ASIC stepping and production lines peak at 1.2Ghz.
And that's why some chips overclock better than others, and manufacturing date and location is interesting data on overclockers.com
Philips _is_ a pretty powerful company. By selling Polygram, they basically said "this RIAA/MPAA stuff isn't working for us", so I guess this is one of their early steps to go head-on with the old establishment. Of course, We'll have to see it first to believe it, but I think that what Philips is trying to here is good: It's time to get rid of the ancient analog VHS, and if that's gonna step on some toes, too bad for them. No reason to halt progress because some people have long toes.
How very true, especially it's backlight. I guess the contrast on LCDs isn't good enough for indirect light sources (mirror instead of backlight?).
About cooling the CPU, time for the naked die on the inside of the casing? The drive could use a good thermal coupling with the case too. Then, in the water, the laptop would have water cooling to take care of the heat.
Well, somebody wanted to use that free advertising for his website!
whois blowthedotoutyourass.com@whois.register.com shows that the domain is registered to somebody in Chicago, and there even is a web server at www.blowthedotoutyourass.com.
If ever, but when such cards will be available at near '100mbit' prices, then who's to say that a low-cost setup will not be made with a PC with four of those cards in it, acting as a router, switch, or hub.
Sure, the theoretical max of 133MBps of the PCI bus is low for a switch backplane for 4 1gig cards, and the latency will not be a winner, but it will beat a 100mbit switch in quite a number of circumstances.
Either being able to return an opened shrink-wrap box of software, or being able to review the license before buying it would however be necessary to protect the consumer. If the consumer can only find out about unwanted license restrictions after the sale is final, then the consumer is screwed anyway.
In the mean time, maybe we need a www.softwarelicenseregistry.org to keep track of software licenses?
> If you want powerful shell scripting, grab the > Cygwin tools which include bash, make, gcc, etc.
That's really funny, so you fix a shortcoming of MS W2K by downloading a RedHat product off the Web and using it?
(or maybe you didn't know that Cygnus is part of RedHat software).
Btw: can perl for win32 do a decent fork() yet? Last time I checked, it did not fork at all (If so, please point me towards the url to download it, I could really use it).
I think the basic problem here is what you mentioned yourself, that system administrators forget to remove (unnecessary) default accounts, or forget to patch for security bugs.
What always has been part in the equation used as for why the MS solution would be best (beating Unix), was the ease of use, and the resulting lower cost of ownership because you could hire cheaper people for administering your systems, and that those cheaper people would require less time per server to administer, because the OS was to userfriendly.
That part of the equation has now, repeatedly, been proven to be faulty.
I always thought that hiring (cheap) half-knowledgable system (security) administrators and compensating by using an expensive point-and-click server operating system was a guarantee to result in secure Internet servers. That that basic rule would be the reason for NT to replace Unix, wasn't it? Low cost of ownership resulting from low cost of the people administering and securing the system?
Visit these sites, and try those progams out (at least view some screenshots). You'll probably be surprized. These programs may not all be as complete as the windows/dos 'original', but they are all free and under development, which means: better to come soon, and all remaining free.
www.soundtracker.org SoundTracker is a music tracking tool for Unix / X11 with a design similar to the DOS program FastTracker and the Amiga legend ProTracker.
Brahms/KooBase Brahms intends to be for Linux, what CuBase is for MacOS/Windows.
WaveForge WaveForge is a free Sound Editor. It is aimed to be a free Sound Forge Clone for Linux. All the capabilitied of the Sound forge will be (hopefully) implemented in this version.
What keyboard shortcuts does windows have to lower or raise a window besided the painfully slow alt-tab? Both Windowmaker and Enlightenment have Alt-arrowdown and Alt-arrowup.
What keyboard shortcuts does windows have to move around on virtual desktops? Oh wait. windows doesn't even have that most basic GUI feature
Do you really have to click on the titlebar to move a window or aim for the four-pixel wide window borders to resize? Both WindowMaker and Enlightenment have alt-mousedrag and alt-mousedrag-leftkey to do that without needing the surgeon's aim.
"And until Linux [blah bla blah], it will never make any inroads into the desktop world."
Sigh. And this is just like what many people said about Linux as a server.
Sure, Linux will never be anything unless X or Y happens...
When I look at what's happening in colleges and business workplaces around the world, I can only conclude Linux is already making inroads in all aspects of the computing world, that is server, desktop, embedded, and wireless.
So I say, Linux will get as far as it needs to go, no matter what happens.
The most difficult problem would be to personally keeping it up-to-date. In comes an automatic update system (bandwidth, esp in the 256kbit range will be damn cheap and common too (how big does a broadcast channel need to be to be able to send out _all_ new music?)), but then why download music you're not going to listen to. So my guess is there will be repositories easy access, much like video on demand, but then audio (or video clips thrown in for free?), and then not on-demand, but automagically tailored to your own personal taste by smart selection, 'buddy'-systems, deejays, and whatnot. A whole new industry I'd say. 1000 million people in the richest countries, each of them their own personal radio channel.
And still we'd have to listen to those ads that pay for it all...
Everything will change, but then again it will all stay the same.
The bandwidth can be taken care of however. All that is needed is the ability to optically multiplex and demultiplex the bitstream before interfacing to electronics. In most cases, the signal has originated in the electrical domain anyway (how 'bout inside a PC somewhere), so that's the place to compress/decompress, before it's mixed with the other electical signals.
So does this mean somebody should stand up and challenge the standard, and maybe make something based on The GNU Privacy Guard (GPG) to get encryption into something called GBIND?
Is the Open Source community strong enough yet to overturn a bad standard in such way?
We could conclude that they (MS) are telling the truth and we are too suspicious. But then again, maybe not. It has been said that "being paranoid doesn't mean that you're not being followed"...
Possibly, Microsoft can not admit to having installed a backdoor simply because they are required so by law, and/or by a non disclosure agreement.
I know one thing, this smells fishy and just inforces my personal preference for Netscape or even better, open source Mozilla (btw, when will Mozilla finally give us the final gecko)?
Slashdot Mirror
"Information wants to be free"
Mind you, the word 'free' here is not meant 'free' as in free beer, but 'free' as in free speech.
What you're paying for at the doctor is not the disease information, because you have free access to all the information the doctors has in the freely accessible university libraries. When paying the doctor, you're paying for an *expert* to handle the information.
Of course: We can't expect a WallStreetJournal reporter to join linux-kernel@vger to get an interview, can we?
"it creates a situation in which an IT staffer may make changes that no one else knows about"
Translation: "Somebody may do something undocumented". What a moot point. As if only Linux is vulnerable to something like that.
Even then, if you don't want to differ from the crowd (read: if you prefer mass-produce jeans to tailor-made clothes), tell your staffers to use only original media found in shrink-wrap software, which is definitely abundantly available for Linux... (and watch your staffers quit their jobs to work for the wiser competitor).
"Having somebody who can screw around with my operating system would make me very, very nervous"
Can you say ILOVEYOU or Back Orifice?, Mr Firstbrook.
This guy is not an 'analists', he's nervous. He's not nervous about any perceived vulerability of Linux, but he's nervous about Linux period. Most of us here understand why...
If any OS allows people to screw around, it's the other OS, definitely not Linux.
-- HOWTO Change your Linux PC into a Windows PC: "chmod -R a+rwxs / && chown -R root.root / && echo guest::0:0::/tmp:/bin/bash >>/tmp/passwd"
"but you can only read the odd or even whatevers at a time."
I don't see why that would have to cause any more than a tiny slowdown in any half-decent RAMBUS interface implementation. It can be dealt with in the chipset with a simple permutation of the address lines (make the address line that changes on the odd/even boundaries the MSB), which does not cost any hardware. Then the only place where the 'bank switch' occurs is at 50% capacity (MSB bit flip).
I can imagine that any decent RAMBUS interfacing ASIC does (would do) such a thing.
"Current is what causes heat, NOT voltage.."
Not true, heat is the dissipation of power. On Ohms loads (load with linear resistance), this is current multiplied by voltage (P=VI). The voltage I'm talking about here is the voltage drop on the path that the current is following, which is the supply voltage of the CPU.
That's why heat (power) goes up squared when voltage is increased with constant resistance (P=VI & V=IR, P=(I*I*R))
In addition to that, I can make some educated guesses: part of the current is the CMOS gate current, and part of the current is charging and discarging of capacitance.
The CMOS gate current is quite linear with respect to the clock frequency and the voltage applied between source and drain, but the capacitance charging and discharging current increases with higher clock frequencies.
Hence when overclocking, the heat dissipation goes up more than squared the increase of supply voltage, because the load is not Ohms, and you're increasing the charging current with your higher clock frequency.
I think AMD is currently making most of it's money on the 700-800Mhz athlons, not on the 1Gz version, which at it's price premium is mostly for careless spenders anyway (at least in my application area, the extra money is better spent on SDRAMS, bless the Abit KA7).
... depending on the market conditions and current yield, they might even stock up on a couple more 1Ghz versions, which they will then sell later when the improved ASIC stepping and production lines peak at 1.2Ghz.
And that's why some chips overclock better than others, and manufacturing date and location is interesting data on overclockers.com
...
Philips _is_ a pretty powerful company. By selling Polygram, they basically said "this RIAA/MPAA stuff isn't working for us", so I guess this is one of their early steps to go head-on with the old establishment. Of course, We'll have to see it first to believe it, but I think that what Philips is trying to here is good: It's time to get rid of the ancient analog VHS, and if that's gonna step on some toes, too bad for them. No reason to halt progress because some people have long toes.
Hey, that's a nice one for my private jesred/squid redirector rules. ;-)
Thanks for the tip!
How very true, especially it's backlight. I guess the contrast on LCDs isn't good enough for indirect light sources (mirror instead of backlight?).
About cooling the CPU, time for the naked die on the inside of the casing? The drive could use a good thermal coupling with the case too. Then, in the water, the laptop would have water cooling to take care of the heat.
AFAIK, Mosix can run over any TCP/IP, so ethernet included.
Oops, it was already in the header. Sorry, 'bout that.
;-))
There goes karma
whois blowthedotoutyourass.com@whois.register.com
shows that the domain is registered to somebody in Chicago, and there even is a web server at www.blowthedotoutyourass.com.
Just thought I would mention that a cluster of Linux computers is by definition a Beowulf Cluster.
Not completely, there is also mosix.
If ever, but when such cards will be available at near '100mbit' prices, then who's to say that a low-cost setup will not be made with a PC with four of those cards in it, acting as a router, switch, or hub.
Sure, the theoretical max of 133MBps of the PCI bus is low for a switch backplane for 4 1gig cards, and the latency will not be a winner, but it will beat a 100mbit switch in quite a number of circumstances.
Good point.
Either being able to return an opened shrink-wrap box of software, or being able to review the license before buying it would however be necessary to protect the consumer. If the consumer can only find out about unwanted license restrictions after the sale is final, then the consumer is screwed anyway.
In the mean time, maybe we need a www.softwarelicenseregistry.org to keep track of software licenses?
> If you want powerful shell scripting, grab the
> Cygwin tools which include bash, make, gcc, etc.
That's really funny, so you fix a shortcoming of MS W2K by downloading a RedHat product off the Web and using it?
(or maybe you didn't know that Cygnus is part of RedHat software).
Btw: can perl for win32 do a decent fork() yet? Last time I checked, it did not fork at all (If so, please point me towards the url to download it, I could really use it).
I don't think it's about quality of the software.
I think the basic problem here is what you mentioned yourself, that system administrators forget to remove (unnecessary) default accounts, or forget to patch for security bugs.
What always has been part in the equation used as for why the MS solution would be best (beating Unix), was the ease of use, and the resulting lower cost of ownership because you could hire cheaper people for administering your systems, and that those cheaper people would require less time per server to administer, because the OS was to userfriendly.
That part of the equation has now, repeatedly, been proven to be faulty.
Geesh, I wonder what went wrong...
I always thought that hiring (cheap) half-knowledgable system (security) administrators and compensating by using an expensive point-and-click server operating system was a guarantee to result in secure Internet servers. That that basic rule would be the reason for NT to replace Unix, wasn't it? Low cost of ownership resulting from low cost of the people administering and securing the system?
Suppose something is wrong with that?
(irony present)
Isnt the ::$DATA bug one that was found over a year ago and was supposed to be fixed by MS ages ago as well?
Visit these sites, and try those progams out (at least view some screenshots). You'll probably be surprized. These programs may not all be as complete as the windows/dos 'original', but they are all free and under development, which means: better to come soon, and all remaining free.
www.soundtracker.org SoundTracker is a music tracking tool for Unix / X11 with a design similar to the DOS program FastTracker and the Amiga legend ProTracker.
Brahms/KooBase Brahms intends to be for Linux, what CuBase is for MacOS/Windows.
WaveForge WaveForge is a free Sound Editor. It is aimed to be a free Sound Forge Clone for Linux. All the capabilitied of the Sound forge will be (hopefully) implemented in this version.
(how did I find them? Freshmeat)
What keyboard shortcuts does windows have to lower or raise a window besided the painfully slow alt-tab? Both Windowmaker and Enlightenment have Alt-arrowdown and Alt-arrowup.
What keyboard shortcuts does windows have to move around on virtual desktops? Oh wait. windows doesn't even have that most basic GUI feature
Do you really have to click on the titlebar to move a window or aim for the four-pixel wide window borders to resize? Both WindowMaker and Enlightenment have alt-mousedrag and alt-mousedrag-leftkey to do that without needing the surgeon's aim.
Sigh. And this is just like what many people said about Linux as a server.
Sure, Linux will never be anything unless X or Y happens...
When I look at what's happening in colleges and business workplaces around the world, I can only conclude Linux is already making inroads in all aspects of the computing world, that is server, desktop, embedded, and wireless.
So I say, Linux will get as far as it needs to go, no matter what happens.
The most difficult problem would be to personally keeping it up-to-date. In comes an automatic update system (bandwidth, esp in the 256kbit range will be damn cheap and common too (how big does a broadcast channel need to be to be able to send out _all_ new music?)), but then why download music you're not going to listen to. So my guess is there will be repositories easy access, much like video on demand, but then audio (or video clips thrown in for free?), and then not on-demand, but automagically tailored to your own personal taste by smart selection, 'buddy'-systems, deejays, and whatnot. A whole new industry I'd say. 1000 million people in the richest countries, each of them their own personal radio channel.
And still we'd have to listen to those ads that pay for it all...
Everything will change, but then again it will all stay the same.
Parallellism.
Of course, the latency issue remains.
The bandwidth can be taken care of however. All that is needed is the ability to optically multiplex and demultiplex the bitstream before interfacing to electronics. In most cases, the signal has originated in the electrical domain anyway (how 'bout inside a PC somewhere), so that's the place to compress/decompress, before it's mixed with the other electical signals.
Is the Open Source community strong enough yet to overturn a bad standard in such way?
We could conclude that they (MS) are telling the truth and we are too suspicious. But then again, maybe not. It has been said that "being paranoid doesn't mean that you're not being followed"...
Possibly, Microsoft can not admit to having installed a backdoor simply because they are required so by law, and/or by a non disclosure agreement.
I know one thing, this smells fishy and just inforces my personal preference for Netscape or even better, open source Mozilla (btw, when will Mozilla finally give us the final gecko)?