Linux has much, much more to offer users than superior security to Windows.
Having switched my desktop machine over to Linux several months ago, here
are the differences I noticed the most:
Speed. Linux does more with less. On my Athlon XP 1500+,
Windows XP lags noticibly on many operations, but there are virtually zero
delays using Linux.
Usability. I'll take open source desktop tools any day over the
Windows equivalent. The GNOME desktop is better than that of Windows,
BeOS, KDE, and NeXT combined. It is designed by people who actually
know what the users need from a desktop, rather than people intent
on writing a desktop that integrates Passport and spyware into every single
applet.
Web browsing. Mozilla 0.9.7 is so compatible, reliable, and
quick that I have uninstalled IE on all of my 80 Windows clients' machines
and replaced it with Mozilla. The users loved the tabbed browsing and have
probably never even looked back.
Accessibility. Linux supports such accessibility features as
sticky modifier keys, text to speech support (even for images, using
OCR), and many other things that make life easier for users with
disabilities. Windows has limited support, at best, for these things.
Standardization. Linux supports all of the latest standards
that Microsoft flouts. It supports open document formats, open web page
formats, and many other encodings that are not patented or non-free. Truly
Linux sets the bar for other OSes to live up to.
Given these many reasons, it is hard to imagine that Windows will be able
to offer more to the desktop user than Windows anytime soon.
Our company is owned by a devout Christian family, who really does have strong feeling about stealing software and other immoral behavior. As the operators of a private network, we have the right and the moral obligation to remove users who are engaged in illegal activities.
(For the record, we do not block porn sites, but we keep a close eye on them to make sure users are not distributing or accessing images that are illegal under current laws.)
From what I understand, it is fairly common for others in our industry to keep their finger on the pulse and maintain good logs on what their users are up to.
I am a network engineer for a small, independent cable franchise in PA, and
we recently were forced to make the difficult decision between open access
and traffic shaping. Some of the things that we needed to weigh were:
Cost. With our traffic levels at the time, we were on the
verge of ordering three new T1 lines for a moderately sized (pop. 80,000)
suburb. Those lines would have cost us about $2000/month for service and
support.
Service quality. Since the rise of KaZaA and Morpheus, our
traffic has doubled from what it was during peak Napster season.
Our upstream was especially swamped.
Maintenance. Many file sharing clients install spyware, "ad
gators," and other software that does a splendid job of screwing up their
network stacks. These customers then required site visits for us to fix
their systems.
Copyright violation. As a small company, we had serious
reservations about knowingly allowing such serious ethical lapses to take
place on our network.
As it turned out, we had no other choice than to start limiting service. I
came up with the following plan, which the managers approved:
Block all file sharing ports at the router level. 1214, 6699,
6346, 40000-42000, and all of their cousins were history.
Block all incoming connections to our users, so that they could not
become servers. We allowed SSH as long as it is OpenSSH >= 2.5.2.
Block all known VPN clients. These were sucking up tremendous amounts
of bandwidth, since we are in a rural area and many people liked to
telecommute using our service.
Cancel three of our T1 circuits.
Institute a "one strike you're out policy" on Nimda, email virii,
spamming, and piracy. So far we have only had three disconnections.
Charge a $209 service fee to users who have crippled their internet
access through a fault of their own.
And, the silver lining on the cloud: Cut rates by 33%.
The result? Profits are up by 7.5%; from the $209 service charge alone, we
have collected several thousand dollars. Most users report much better
latencies to major sites and very good burst bandwidth. We lost a couple
of users from the VPN ban but they were all above-average bandwidth hogs so
we don't miss them. All is right in the world, and I'm very satisfied with
how things worked out.
I haven't heard anything official, but I saw a couple of remarks on linux.redhat on usenet that indicated that Redhat was going to strengthen the way their PGP/GPG checking on rpms worked to prevent unauthorized binaries from being surreptitiously installed.
No word on Debian AFAIK, but I don't really support it so I wouldn't know.
Firewire has some neat features, to be sure. Hostless operation can be extremely useful. But if you're really interested in performance and reliability, I'd recommend USB 2.0 in a heartbeat. I have used both types of peripherals and found that most of the time, USB (even 1.0) hard drives and CD writers outperform their FireWire equilvalents. You don't need a special card and you don't need to pay the Apple tax for USB. And the Linux support for USB is far more refined than the Linux support for FireWire (although I'd imagine this is probably just because USB users outnumber FireWire users by a factor of 10:1).
Although many researchers have insisted that we will run out of petroleum by the year 2050 (or earlier), their calculations are fueled by coarse-grained extrapolation and bad science run amok. What many environmentalists do to advance their anti-oil cause is extrapolate oil use increases on 25-year or 50-year periods. For instance, they will find that oil use has increased (say) 250% from 1950 through 2000. And conclude that use will increase another 250% between 2000 and 2050.
Obviously that is not the case. The usage increases were huge when the nation became industrialized and started needing more energy, but it is ludicrous to think that the rate of change will remain constant. That is like saying that since the number of homes with PCs has increased by 1500% since 1985, it will increase another 1500% by 2017.
-all dead homiez
Re:What they don't tell you
on
80 Gig MP3 Player
·
· Score: 3, Informative
Have you looked into AS31? Not sure on the license but source is available (I suspect it may be GPL). AS31 and several other Linux based tools including an 8051 simulator are available here.
I had the same exact reaction as most everyone else here when I read about the P4 clock throttling - I thought it was shady and underhanded. The ironic part is that about two months later, the fan died overnight on my Athlon 950 and it took the chip with it. (It didn't look disfigured like the tomshardware pics, but it didn't work either.) If it were an Intel chip, I would have been in the clear.
Fortunately, AMD honored the warranty on the chip because I was using it in an "approved" configuration, and replaced it with a 1.2Ghz (!) chip for free. I'm not sure if Intel would do something like that (especially given their financial shape). But it has definitely made me a more loyal AMD customer, despite their debatable thermal issues.
In all likelihood, the Linux ABI will become a standard for all non-Microsoft x86 operating systems. It is simple and legal to implement, and very robust and powerful.
Things definitely are moving in this direction. I just read on the netbsd-discuss mailing list that those folks are considering abandoning the slow BSD-style stack-based kernel calls, in favor of the quick register-based kernel call syntax favored by Linux and Solaris. If they do that, most syscalls will differ only in call number from the ones in Linux.
Actually, most GPL violations are not large enough to show up on the FSF's radar screen at all. I have personally seen several flagrant violations (such as this one), that are not pursued just because nobody has heard of the product or the company behind it.
I'm not sure how widespread it is, but I observed that the way @home blocks connections to port 80 in my area (western Milwaukee suburbs) is by setting extra flags on all SYN packets headed to port 80. IIRC, these packets look like elements of an XMAS scan under tcpdump - many extra flags, such as ACK, FIN, and URG, are set and the packets are discarded as invalid by the stock kernel (and rightly so).
What I did to counter this was to make a very quick and dirty patch to my kernel, which accepted these malformed packets as normal SYN requests. The result? Web services were back to normal and Apache is chugging away as we speak. I've been doing this since the ban and have had no problems at all.
Take a look at the "What are file attributes?" FAQ at www.atheos.cx for one of the reasons why porting KDE or any other UNIX desktops to AtheOS is not something I consider an option.
Actually, Linus has all but endorsed including "streams" or "resource forks" in Linux in this posting. These would make the implementation of AtheOS-like attributes trivial, and certainly KDE/Gnome would provide support for them. So, despite Kurt's other arguments against KDE, there will be few technical reasons that keep it from being ported. (Someday.)
The "disclosure argument," on which this suit is based, will not be a legal problem once this scheme catches on with all of the record companies. Expect all CDs sold in the near future to have a little label on them indicating that they will not work normally (or at all) in PC CD-ROM drives. Why? Because the record companies believe that the reduction in piracy is worth losing a couple of customers who play CDs on their computers. They may be right; they may be wrong; but their goal is content control, not necessarily profit. DeCSS didn't cost movie companies a dime, but they spent millions to squash it.
What can we do about this? We can support companies who make CD-ROM drives that are not affected by the protection. (Several of these have been "discovered" recently.) We can lobby Congress and ask for a bill that gives us our fair use rights again. We can buy a $30 Discman clone and use that to play CDs, like in the good old days. There's no easy answer, but to paraphrase the old cliche, the price of freedom is eternal vigilance.
You're missing an important point: how do you know that a given closed-source email encryption/decryption engine does not "leak" keys? You have no sure way to know that your keys won't wind up:
"Accidentally" sent packed into an IP header and sent to the NSA
Somewhere in your swap space, because some coder doesn't know how to lock memory correctly
Somewhere else on your hard drive, because some coder doesn't care about protecting your keys (or know what he's doing).
Compromised in response to a malicious message that the program is trying to decrypt. Don't forget about buffer overflows.
Trusting a closed source application means that you're trusting every programmer who ever wrote a line of code for the application. When you can't see that code to make sure it's not crap, you've got a security nightmare waiting to happen.
This is an area where OSS really shines. Microsoft NSA key rumors aside, the truth of the matter is that it is almost impossible to audit closed source programs for backdoors and security flaws. As more and more stupid programming mistakes are discovered, more and more people will realize that OSS is the only way to go when security and/or privacy is a concern. Expect many more endorsements of OSS in the near future for this very reason.
There's always a trick to solving these sorts of puzzles. Check this one out - it might help.
-all dead homiez
Potential threat to free speech
on
Rent-a-Game
·
· Score: 1, Flamebait
These "software rental" schemes are really starting to worry me. Why? Because this is what inevitably happens:
Some company (MS, DIVX, BT Openworld, etc.) launches their rental product.
Some 16-year old somewhere in the world, who realizes that if the code can execute on the user's box, the user can save a copy, finds a way to crack the piracy protection and uploads the crack to his favorite FTP site
A bunch of freeloaders all over the U.S. start using the crack and getting the software for free
The company gets pissed off and uses the DMCA to squelch distribution of the crack's source code
The users get pissed off and distribute the code even more
Some hard-ass conservative judge decides that the best way to stop the piracy is to butcher the First Amendment
And now, source code is no longer considered protected speech
It's not the vendors' fault for deploying these schemes, but the damage to our civil liberties that inevitably results is immense.
1. You have wrote the whole OS in a x86 assembly. How much speed you think you gained by using asm-only when compared writting the source in C or C++? Ville Turjanmaa: Parts of Linux was rewritten in assembly and the speed gain was 10-40%. That will give an idea.
I don't mean to flame here, but one of the first things you learn in a computer architecture class is to "make the common case fast." The parts of Linux that were rewritten in asm that improved performance by 10-40% were most likely primitives that were executed hundreds of times a second - like bcopy() and maybe some parts of the VM subsystem. Ville's response draws no distinction between rewriting bcopy() in asm, and rewriting printk() (which is slow, but rarely executed) in asm. Unfortunately, I see no point in rewriting them both if it's not necessary. Sometimes it matters but often it doesn't.
The space advantage to hand-optimized asm is clear, but the cost in portability and time almost certainly outweighs it. I really don't see what this OS offers that Linux doesn't have.
Actually this is a DVD+RW drive. There are at least seven different standards for recording/pressing DVDs and they all have their pros and cons. A summary can be found here.
Indeed, having so many different standards is sure to slow the adoption of a recordable DVD format. But hopefully, someday everyone will use the same format and the media will be cheap. Witness the price drops (over time) that occurred with CD-Rs, and then with CD-RWs - the drives did not become commodities until the media did. Back in 1996, blank CD-Rs were about $20 each, as a point of reference. Be patient; we will have cheap recordable DVDs soon enough.
It's a basic principle of economics. The "rational consumer" does whatever is best for him. It's the same reason why it's more effective to encourage 10 of your friends to vote for Candidate X and never voting, rather than casting one vote for him yourself.
As much as we all love our PVRs and Junkbuster proxies, the problem is that many content producers' revenue models are built with the assumption that most of the users will be viewing advertisements. That means that each time they notice ad viewings or response rates declining, the price of content goes up. In meatspace, this means your cable bill goes up at twice the rate of inflation (sound familiar, folks?). On the Internet, this means that more and more advertising-supported businesses fail.
What is the solution? Well, the ideal mix of users includes a large majority who view the ads, and a small minority (this usually turns out to be the technically-inclined Slashdot crowd) who knows how to avoid them. Keep this mix, and everything is great for both groups. Let the balance get out of hand, and the result will disappoint us all.
The rise of PVRs can improve our enjoyment of TV, or it can destroy the content providers. And at this point, it could go either way.
Obviously there are issues with something like this (especially mailing lists, and worms that do
attachments)
Is there some hidden reason why we would want millions of copies of an email worm's attachment to get through? This could actually be part of the solution to two problems.
Also, do note that a common method of spamming is to connect to an open relay and have the relay take care of sending out thousands of identical messages by simply sending thousands of "RCPT TO:" commands. Checksumming spam would completely break this spamming method and would force the spammer to retransmit the entire message for every recipient in order to vary it, thus making the process more costly.
Though the Congress and the courts may be leery of overturning the DMCA because their pockets are lined with cash from large, moneyedinterests, the Skylarov criminal trial leaves the door wide open for jury nullification. If the EFF provides a sufficiently good defense team, they will have no problem demonstrating the fascist nature of the DMCA to a competent jury. This could be their big chance to stop the oppression and corporate censorship dead in its tracks.
The problem with a "safe for children" area is that this invariably leads to censorship of factual information on unpopular viewpoints. After all, there is no consensus whatsoever (either in America or in the rest of the world) on what materials children should be exposed to - and there is precious little consensus on what children shouldn't be exposed to. Consider:
Should children be exposed to sites that say that maybe it's okay to be gay?
Should children be exposed to sites that disagree with the government's "one hit of pot will kill you" philosophy on drugs? How about sites that say "ritalin is bad" to kids who are diagnosed with ADD?
Should kids be exposed to non-mainstream political viewpoints? Should politically incorrect sites that, while not taking an anti-minority "violent white supremicist" stance, oppose racial preferences or hate crime laws, be allowed?
Should kids be exposed to sites that promote atheism and/or secular humanism? How about sites that promote Satanism (a religion that is most decidedly allowed under the First Amendment)?
Unfortunately, these same questions just keep coming up because they are a natural part of any attempt to restrict information "to protect the children." It's a twofold question: who watches the watchdogs, and who trains them? And it defies any easy solution, because every parent and every citizen has their own thoughts on how other people should be raising their kids.
Gee, maybe censorship isn't the right solution. It sure seems out of place in such a diverse nation. Maybe all we need is some good parenting instead. But that's never going to happen...
Slashdot Mirror
- Speed. Linux does more with less. On my Athlon XP 1500+,
Windows XP lags noticibly on many operations, but there are virtually zero
delays using Linux.
- Usability. I'll take open source desktop tools any day over the
Windows equivalent. The GNOME desktop is better than that of Windows,
BeOS, KDE, and NeXT combined. It is designed by people who actually
know what the users need from a desktop, rather than people intent
on writing a desktop that integrates Passport and spyware into every single
applet.
- Web browsing. Mozilla 0.9.7 is so compatible, reliable, and
quick that I have uninstalled IE on all of my 80 Windows clients' machines
and replaced it with Mozilla. The users loved the tabbed browsing and have
probably never even looked back.
- Accessibility. Linux supports such accessibility features as
sticky modifier keys, text to speech support (even for images, using
OCR), and many other things that make life easier for users with
disabilities. Windows has limited support, at best, for these things.
- Standardization. Linux supports all of the latest standards
that Microsoft flouts. It supports open document formats, open web page
formats, and many other encodings that are not patented or non-free. Truly
Linux sets the bar for other OSes to live up to.
Given these many reasons, it is hard to imagine that Windows will be able to offer more to the desktop user than Windows anytime soon.-all dead homiez
(For the record, we do not block porn sites, but we keep a close eye on them to make sure users are not distributing or accessing images that are illegal under current laws.)
From what I understand, it is fairly common for others in our industry to keep their finger on the pulse and maintain good logs on what their users are up to.
-all dead homiez
- Cost. With our traffic levels at the time, we were on the
verge of ordering three new T1 lines for a moderately sized (pop. 80,000)
suburb. Those lines would have cost us about $2000/month for service and
support.
- Service quality. Since the rise of KaZaA and Morpheus, our
traffic has doubled from what it was during peak Napster season.
Our upstream was especially swamped.
- Maintenance. Many file sharing clients install spyware, "ad
gators," and other software that does a splendid job of screwing up their
network stacks. These customers then required site visits for us to fix
their systems.
- Copyright violation. As a small company, we had serious
reservations about knowingly allowing such serious ethical lapses to take
place on our network.
As it turned out, we had no other choice than to start limiting service. I came up with the following plan, which the managers approved:- Block all file sharing ports at the router level. 1214, 6699,
6346, 40000-42000, and all of their cousins were history.
- Block all incoming connections to our users, so that they could not
become servers. We allowed SSH as long as it is OpenSSH >= 2.5.2.
- Block all known VPN clients. These were sucking up tremendous amounts
of bandwidth, since we are in a rural area and many people liked to
telecommute using our service.
- Cancel three of our T1 circuits.
- Institute a "one strike you're out policy" on Nimda, email virii,
spamming, and piracy. So far we have only had three disconnections.
- Charge a $209 service fee to users who have crippled their internet
access through a fault of their own.
- And, the silver lining on the cloud: Cut rates by 33%.
The result? Profits are up by 7.5%; from the $209 service charge alone, we have collected several thousand dollars. Most users report much better latencies to major sites and very good burst bandwidth. We lost a couple of users from the VPN ban but they were all above-average bandwidth hogs so we don't miss them. All is right in the world, and I'm very satisfied with how things worked out.-all dead homiez
No word on Debian AFAIK, but I don't really support it so I wouldn't know.
-all dead homiez
Just my 2c...
-all dead homiez
Although many researchers have insisted that we will run out of petroleum by the year 2050 (or earlier), their calculations are fueled by coarse-grained extrapolation and bad science run amok. What many environmentalists do to advance their anti-oil cause is extrapolate oil use increases on 25-year or 50-year periods. For instance, they will find that oil use has increased (say) 250% from 1950 through 2000. And conclude that use will increase another 250% between 2000 and 2050.
Obviously that is not the case. The usage increases were huge when the nation became industrialized and started needing more energy, but it is ludicrous to think that the rate of change will remain constant. That is like saying that since the number of homes with PCs has increased by 1500% since 1985, it will increase another 1500% by 2017.
-all dead homiez
-all dead homiez
Fortunately, AMD honored the warranty on the chip because I was using it in an "approved" configuration, and replaced it with a 1.2Ghz (!) chip for free. I'm not sure if Intel would do something like that (especially given their financial shape). But it has definitely made me a more loyal AMD customer, despite their debatable thermal issues.
-all dead homiez
Things definitely are moving in this direction. I just read on the netbsd-discuss mailing list that those folks are considering abandoning the slow BSD-style stack-based kernel calls, in favor of the quick register-based kernel call syntax favored by Linux and Solaris. If they do that, most syscalls will differ only in call number from the ones in Linux.
-all dead homiez
I believe the story on the widened probe is here or here.
-all dead homiez
What I did to counter this was to make a very quick and dirty patch to my kernel, which accepted these malformed packets as normal SYN requests. The result? Web services were back to normal and Apache is chugging away as we speak. I've been doing this since the ban and have had no problems at all.
Might be something to try...
-all dead homiez
Actually, Linus has all but endorsed including "streams" or "resource forks" in Linux in this posting. These would make the implementation of AtheOS-like attributes trivial, and certainly KDE/Gnome would provide support for them. So, despite Kurt's other arguments against KDE, there will be few technical reasons that keep it from being ported. (Someday.)
-all dead homiez
What can we do about this? We can support companies who make CD-ROM drives that are not affected by the protection. (Several of these have been "discovered" recently.) We can lobby Congress and ask for a bill that gives us our fair use rights again. We can buy a $30 Discman clone and use that to play CDs, like in the good old days. There's no easy answer, but to paraphrase the old cliche, the price of freedom is eternal vigilance.
-all dead homiez
Trusting a closed source application means that you're trusting every programmer who ever wrote a line of code for the application. When you can't see that code to make sure it's not crap, you've got a security nightmare waiting to happen.
-all dead homiez
-all dead homiez
-all dead homiez
Some company (MS, DIVX, BT Openworld, etc.) launches their rental product.
Some 16-year old somewhere in the world, who realizes that if the code can execute on the user's box, the user can save a copy, finds a way to crack the piracy protection and uploads the crack to his favorite FTP site
A bunch of freeloaders all over the U.S. start using the crack and getting the software for free
The company gets pissed off and uses the DMCA to squelch distribution of the crack's source code
The users get pissed off and distribute the code even more
Some hard-ass conservative judge decides that the best way to stop the piracy is to butcher the First Amendment
And now, source code is no longer considered protected speech
It's not the vendors' fault for deploying these schemes, but the damage to our civil liberties that inevitably results is immense.
-all dead homiez
Ville Turjanmaa: Parts of Linux was rewritten in assembly and the speed gain was 10-40%. That will give an idea.
I don't mean to flame here, but one of the first things you learn in a computer architecture class is to "make the common case fast." The parts of Linux that were rewritten in asm that improved performance by 10-40% were most likely primitives that were executed hundreds of times a second - like bcopy() and maybe some parts of the VM subsystem. Ville's response draws no distinction between rewriting bcopy() in asm, and rewriting printk() (which is slow, but rarely executed) in asm. Unfortunately, I see no point in rewriting them both if it's not necessary. Sometimes it matters but often it doesn't.
The space advantage to hand-optimized asm is clear, but the cost in portability and time almost certainly outweighs it. I really don't see what this OS offers that Linux doesn't have.
-all dead homiez
Actually this is a DVD+RW drive. There are at least seven different standards for recording/pressing DVDs and they all have their pros and cons. A summary can be found here.
Indeed, having so many different standards is sure to slow the adoption of a recordable DVD format. But hopefully, someday everyone will use the same format and the media will be cheap. Witness the price drops (over time) that occurred with CD-Rs, and then with CD-RWs - the drives did not become commodities until the media did. Back in 1996, blank CD-Rs were about $20 each, as a point of reference. Be patient; we will have cheap recordable DVDs soon enough.
-all dead homiez
See here for an in-depth explanation.
-all dead homiez
What is the solution? Well, the ideal mix of users includes a large majority who view the ads, and a small minority (this usually turns out to be the technically-inclined Slashdot crowd) who knows how to avoid them. Keep this mix, and everything is great for both groups. Let the balance get out of hand, and the result will disappoint us all.
The rise of PVRs can improve our enjoyment of TV, or it can destroy the content providers. And at this point, it could go either way.
-all dead homiez
Is there some hidden reason why we would want millions of copies of an email worm's attachment to get through? This could actually be part of the solution to two problems.
Also, do note that a common method of spamming is to connect to an open relay and have the relay take care of sending out thousands of identical messages by simply sending thousands of "RCPT TO:" commands. Checksumming spam would completely break this spamming method and would force the spammer to retransmit the entire message for every recipient in order to vary it, thus making the process more costly.
-all dead homiez
-all dead homiez
Unfortunately, these same questions just keep coming up because they are a natural part of any attempt to restrict information "to protect the children." It's a twofold question: who watches the watchdogs, and who trains them? And it defies any easy solution, because every parent and every citizen has their own thoughts on how other people should be raising their kids.
Gee, maybe censorship isn't the right solution. It sure seems out of place in such a diverse nation. Maybe all we need is some good parenting instead. But that's never going to happen...
My 2c.
-all dead homiez