"Proprietary software authors do not have to do things "properly", they just kludge things together that may or may not work in every possible weirdy case, and rely on nobody ever seeing what an awful job they made of it in the first place."
This indicates that you don't know how a proprietary software development environment works, nor do you understand how an open-source development environment works.
Yes, there are many closed-source products where code is never reviewed or audited. But if you've ever coded in the aerospace or other life-critical industries, it becomes abundantly clear that code reviews are regular practice, and that your job is put on the line very quickly if you don't measure up. Even Microsoft now has detailed security practices, and, after hearing from some of my Microsoft friends, they are taken very seriously.
As for open source, yes, code is often tightly audited, particularly in well-known projects. But many smaller projects, particularly ones with small development teams, never audit code. Even the Linux kernel has many instances of abysmally poor code.
The point is this: open source is not immune from poor coding. And a tightly managed closed-source project can produce high-quality code.
"This machine doesn't even come with Office and it doesn't even sound like it will let you install it! I bet FireFox is a no-no, too."
While the product doesn't come with Office, it does include the full-featured PlanMaker and TextMaker spreadsheet & word processing software, which is really quite excellent.
Re:I don't get it - of course you need specs
on
Linus Says No to 'Specs'
·
· Score: 0, Flamebait
This embodies Linus's bullshit attitude of not standardizing on a binary compatible ABI in the Kernel. You cannot be certain that modules written for today's kernel will work in toomorrow's kernel.
Microsoft Windows now has 5 years of driver binary compatibility. Linux doesn't even have 5 days of binary compatibility - every minor kernel revision breaks binary modules, as does changing certain kernel options. You have to recompile every module for every kernel.
You ever wonder why there are few 3rd party drivers for Linux? Because a manufacturer doesn't want to release their source code, and because they don't want to provide binary modules for every permutation of every kernel revision in existance.
Now, this alone is bad, but now you consider the fact that the underlying driver architecture is coninually changing. Did you write OSS drivers five years ago? Too bad. Were your X drivers written for XFree86? Time to update them for XOrg.
Moreover, Linux has been quoted as saying that the breaking of binary compatibility between kernel releases is intentional. That's right - he wants it to be hard to write binary modules for Linux. You don't want to use the platform the way he wants you to? Too bad. Use another platform.
Well, guess what: I use a platform that supports my hardware. That happens to be Windows.
"That idea died a horrible death, despite Microsoft's best efforts to make the Network the Computer."
You're thinking of Sun ("The Network is the Computer". Microsoft doesn't want network applications to take off because it undermines the strength of their platform.
Have you heard of the recovery console? Or WinPE/BartPE? Both can be run off a CD, so rootkits are not an issue. Unless, of course, you burn the CD with an infected system.
Typed on a Treo 650, so don't call me a Microsoft apologist.
Palm is, and always was, a crappy platform.
But, here's the thing: back in 1996, it didn't matter that it sucked. Back then, Palm still sucked, but Windows CE sucked more. Palms were cheaper, faster, and they got the PIM stuff done with ease.
But when Microsoft shipped Pocket PC, that all changed. Microsoft had a PDA operating system with multitasking. They had an OS with a real network stack, with real color, and with real networking. They had a real web browser, too, in the box.
More importantly, Microsoft now had devices with enough CPU power and memory to run their piggy Windows CE.
CE is a much, much better development platform than Palm. You don't have bullshit memory limitations to cope with. You don't have to store all of your data as 'databases'. There's an API for writing games (GAPI) that gives you quick access to the display.
Today, Palm OS is running on the same hardware as Windows Mobile. The Treo 650 is not significantly different from many Pocket PCs in terms of hardware capabilities. In many ways, it is even superior. But Palm OS has not kept up with the hardware. There is still no native API for writing ARM applications - YEARS after the shipment of ARM devices. There is still no multitasking. Bugs in applications still reboot the device.
Today, people expect more. They want a real web browser - but Palm OS doesn't have a good one. Why? Because it is easier to port to Windows Mobile than it is to port to Palm OS. Because it is hard to write a good web browser within the confines of device memory. Because the OS has no multithreading and no multitasking.
Palm's problem is that they never innovated. Every feature added to Palm OS since 1997 was added to Windows CE first. Microsoft and its partners added color first, they added expandable storage first, they added real audio first. Palm has been playing catch-up with Microsoft for the last 8 years.
Palm hardware is fine. The Treo is a nice device. I carry one. But Palm OS sucks.
I'm an ECE undergrad student at CU Boulder. Being an engineer is hard work. My roommates complain about having to read 30 pages and write an essay once a week for 'Women in Aincent Greece'. I have to read every bit as much for APPM 1360 (Calculus 2 for engineers), but the difference is that I am not told what to read explicitly, I am expected to read and understand the material myself. And, of course, instead of writing essays, I do problems. It just takes me six hours a week instead of two.
The reason that there is so much engineering attrition is that engineering is HARD. I'm not saying that other degrees aren't valid, but the fact is, at CU Boulder, the engineering majors have a higher average incoming SAT verbal score than the English majors.
If an English major screws up, little bad will come of it. When a doctor, lawyer, or engineer screws up, there is significant real world impact. I want the person defending my liberty to know the law. I want the person operating on me to know medicine. And I want the structural engineer who certified my building to know his trade cold.
That's why engineers, lawyers, and doctors are held to such strict standards. Not everyone is cut out to be a doctor, a lawyer, or an engineer. That's OK.
Typed and sent on a Treo 650. Made possible by engineers like myself. I'm proud to be a part of a trade that makes such a big difference.
"When you do things right, no one can be certain that you have done anything at all."
Palm OS's biggest user, Palm (formerly PalmOne), just switched their flagship product to Palm OS. And PalmSource just got purchased by ACCESS, a company known for their embedded web browser.
Palm OS as we know it is dead.
"TiVo is dying"
Yes, they are. Their partnership with DIRECTV essentially ended when DIRECTV was purchased by News Corp (which owns NDS who already has a PVR solution); the D20 and other upcoming DIRECTV PVRs are based on NDS technology, not TiVo. As for cable companies, their only major deal has been with Comcast, and right now it looks like they won't even have product available until well into 2006. Comcast also already has their own PVR solution, and is already testing a Microsoft TV Foundation PVR, so TiVo probably isn't getting much in the way of royalties from the Comcast deal.
What is TiVo's competitive advantage over integrated PVRs? Yes, their software is better, but most people would rather have a lame integrated PVR than have to deal with the hassle of an IR changer cable.
I hate to break it to the Slashdot zealots, but here it is: Microsoft wants one thing and one thing alone: money. It is not in Microsoft's best interests to restrict development for Windows. It is not in their best interests to break compatibility with older software. Neither of these things will EVER happen at Microsoft because the strength of their platform lies in its software library and ease of development. This news has NO implications for FOSS on Windows.
This article, in fact, is 90 percent FUD. DRM has existed for a long time in the shareware and commercial software world - this just standardizes it and provides centralized downloading and key storage. Not a bad idea, if I say so myself.
At my university (University of Colorado at Boulder), we have an on-campus 33MW power plant. Waste heat from the plant is piped around campus to heat buildings, and the electricity generated is enough to power the campus, with 8MW left over that is sold to the grid. The facility also produces chilled water through a massive vapor-phase system that is used to cool the physics and other nearby buildings.
"In monorail, the train wraps the rail. That means that it can not jump it. In contrast, think about how many of trains that we hear have jumped the track. If you follow the news, it happens every month or so."
Crap crap crappety crap. You cannot compare monorail derailments to train derailments because, if you actually compare the number of hours of service of trains vs. monorails, trains so dramatically outstrip monorails that a comparison is irrelivent.
Looking at derailments is like looking at airplane crashes. Think, for example, of how many people DIE in car crashes every day. Few of those are newsworthy. Major train derailments are newsworthy precicely because they are so rare.
Monorails simply do not have a substantial enough track record to compare their safety to conventional trains.
POP3 SUCKS for webmail. You don't get any sync features, so forget about keeping your folders consistant accross computers. Forget about having folders at all, in fact, because POP3 doesn't support them.
IMAP is a much better protocol. Until GMail supports it, I'm sticking with FastMail.
Bullshit. GSM-1900 and GSM-850 are perfectly functional. Using different frequencies doesn't make the system any less robust or functional, and quadband phones make frequency differences irrelevant.
23 since the release of 1.0. We're now on 1.07. Seven major security releases.
Is it better than IE? Probably. Firefox vulnerabilities tend to be fixed more quickly and are less secure than IE vulnerabilities. That said, IE is considerably easier to update for both corporate users and home users.
Firefox can do better. Fortunately, the update mechanism is much-improved in 1.5. But, remember, Firefox is supposed to be the darling of the Open Source movement. We can do better.
802.11b chipsets are already pretty cheap, and in many cases schools already have network infastructure instealled. If you can do an 802.11b scanner for $50, why not a wireless "clicker"?
At my university, we use H-ITT "clickers". They are the crappiest pieces of crap that I have ever used.
Which is more likely: someone cracking the encryption (which, even in 20 years, will require time, resources, and experise far beyond what most criminals are capable of), or having the card stolen through a mugging or pickpocketing. Door locks work because they require criminals to have more time and more resources. Picking a lock takes specialized equipment and knowledge. Breaking down a strong door requires time and heavy equipment. The more work a crime requies, the less attractive the payout becomes. Credit card theft is an unattractive crime because it is difficult to actually get away with: - Buying goods in a store means that you are videotaped. You also have to come up with a plausible fake card (unless you steal the original - difficult to do without the victim knowing). Moreover, if the card is reported stolen, you may be arrested on the spot. - Buying goods online requires a shipping address. - Cash advances mean getting photographed at an ATM, and they tend to raise fruad alerts - Selling stolen goods is always a problem The security behind this system accomplishes the same thing as a decent lock. Is it unbreakable? Absolutely not. Is it enough to deter most fruad? Yes.
Newer cars have cleaner emissions, more power, ride better, and are considerably safer.
ABS is a valuable system when it is used properly. It allows the driver to steer while braking on slick surfaces, hopefully avoiding a collision.
Older cars may be eaiser to fix, but they break considerably more often. And, no, they aren't "built more solid".
New cars do have crumple zones, but those are designed to decrease the forces acting on you during a crash. They are one of the most important safety advances in automotive history.
Don't give me this shit about "skill in driving". If you want to test your driving skill, try driving a F1 racecar. Otherwise, people neither want nor need "driving skill". Driving safely is about knowing and following the rules, using good judgement, and reacting properly in panic situations. It has nothing to do with whether the vehicle has fuel injection, or whether the engine is computer controlled, or whether the car has crumple zones. We are not "dependant" on airbags in the same way that an airplane is not "dependant" on parachutes.
Slashdot Mirror
"Proprietary software authors do not have to do things "properly", they just kludge things together that may or may not work in every possible weirdy case, and rely on nobody ever seeing what an awful job they made of it in the first place."
This indicates that you don't know how a proprietary software development environment works, nor do you understand how an open-source development environment works.
Yes, there are many closed-source products where code is never reviewed or audited. But if you've ever coded in the aerospace or other life-critical industries, it becomes abundantly clear that code reviews are regular practice, and that your job is put on the line very quickly if you don't measure up. Even Microsoft now has detailed security practices, and, after hearing from some of my Microsoft friends, they are taken very seriously.
As for open source, yes, code is often tightly audited, particularly in well-known projects. But many smaller projects, particularly ones with small development teams, never audit code. Even the Linux kernel has many instances of abysmally poor code.
The point is this: open source is not immune from poor coding. And a tightly managed closed-source project can produce high-quality code.
"This machine doesn't even come with Office and it doesn't even sound like it will let you install it! I bet FireFox is a no-no, too."
While the product doesn't come with Office, it does include the full-featured PlanMaker and TextMaker spreadsheet & word processing software, which is really quite excellent.
This embodies Linus's bullshit attitude of not standardizing on a binary compatible ABI in the Kernel. You cannot be certain that modules written for today's kernel will work in toomorrow's kernel.
Microsoft Windows now has 5 years of driver binary compatibility. Linux doesn't even have 5 days of binary compatibility - every minor kernel revision breaks binary modules, as does changing certain kernel options. You have to recompile every module for every kernel.
You ever wonder why there are few 3rd party drivers for Linux? Because a manufacturer doesn't want to release their source code, and because they don't want to provide binary modules for every permutation of every kernel revision in existance.
Now, this alone is bad, but now you consider the fact that the underlying driver architecture is coninually changing. Did you write OSS drivers five years ago? Too bad. Were your X drivers written for XFree86? Time to update them for XOrg.
Moreover, Linux has been quoted as saying that the breaking of binary compatibility between kernel releases is intentional. That's right - he wants it to be hard to write binary modules for Linux. You don't want to use the platform the way he wants you to? Too bad. Use another platform.
Well, guess what: I use a platform that supports my hardware. That happens to be Windows.
"That idea died a horrible death, despite Microsoft's best efforts to make the Network the Computer."
You're thinking of Sun ("The Network is the Computer". Microsoft doesn't want network applications to take off because it undermines the strength of their platform.
Have you heard of the recovery console? Or WinPE/BartPE? Both can be run off a CD, so rootkits are not an issue. Unless, of course, you burn the CD with an infected system.
Wasn't it the SRBs that caused the distruction of both shuttles?
Typed on a Treo 650, so don't call me a Microsoft apologist.
Palm is, and always was, a crappy platform.
But, here's the thing: back in 1996, it didn't matter that it sucked. Back then, Palm still sucked, but Windows CE sucked more. Palms were cheaper, faster, and they got the PIM stuff done with ease.
But when Microsoft shipped Pocket PC, that all changed. Microsoft had a PDA operating system with multitasking. They had an OS with a real network stack, with real color, and with real networking. They had a real web browser, too, in the box.
More importantly, Microsoft now had devices with enough CPU power and memory to run their piggy Windows CE.
CE is a much, much better development platform than Palm. You don't have bullshit memory limitations to cope with. You don't have to store all of your data as 'databases'. There's an API for writing games (GAPI) that gives you quick access to the display.
Today, Palm OS is running on the same hardware as Windows Mobile. The Treo 650 is not significantly different from many Pocket PCs in terms of hardware capabilities. In many ways, it is even superior. But Palm OS has not kept up with the hardware. There is still no native API for writing ARM applications - YEARS after the shipment of ARM devices. There is still no multitasking. Bugs in applications still reboot the device.
Today, people expect more. They want a real web browser - but Palm OS doesn't have a good one. Why? Because it is easier to port to Windows Mobile than it is to port to Palm OS. Because it is hard to write a good web browser within the confines of device memory. Because the OS has no multithreading and no multitasking.
Palm's problem is that they never innovated. Every feature added to Palm OS since 1997 was added to Windows CE first. Microsoft and its partners added color first, they added expandable storage first, they added real audio first. Palm has been playing catch-up with Microsoft for the last 8 years.
Palm hardware is fine. The Treo is a nice device. I carry one. But Palm OS sucks.
I'm an ECE undergrad student at CU Boulder. Being an engineer is hard work. My roommates complain about having to read 30 pages and write an essay once a week for 'Women in Aincent Greece'. I have to read every bit as much for APPM 1360 (Calculus 2 for engineers), but the difference is that I am not told what to read explicitly, I am expected to read and understand the material myself. And, of course, instead of writing essays, I do problems. It just takes me six hours a week instead of two.
The reason that there is so much engineering attrition is that engineering is HARD. I'm not saying that other degrees aren't valid, but the fact is, at CU Boulder, the engineering majors have a higher average incoming SAT verbal score than the English majors.
If an English major screws up, little bad will come of it. When a doctor, lawyer, or engineer screws up, there is significant real world impact. I want the person defending my liberty to know the law. I want the person operating on me to know medicine. And I want the structural engineer who certified my building to know his trade cold.
That's why engineers, lawyers, and doctors are held to such strict standards. Not everyone is cut out to be a doctor, a lawyer, or an engineer. That's OK.
Typed and sent on a Treo 650. Made possible by engineers like myself. I'm proud to be a part of a trade that makes such a big difference.
"When you do things right, no one can be certain that you have done anything at all."
"PalmOS ... is dying"
l )
y .epl?reqid=760974)
You might want to check that out:
"ACCESS to Extend Leadership in Mobile Device Software with Acquisition of PalmSource" (http://www.access.co.jp/english/press/050909.htm
"Palm and Microsoft Join to Bring the Palm Experience to Windows Mobile"
(http://www.palm.com/us/company/pr/news_feed_stor
Palm OS's biggest user, Palm (formerly PalmOne), just switched their flagship product to Palm OS. And PalmSource just got purchased by ACCESS, a company known for their embedded web browser.
Palm OS as we know it is dead.
"TiVo is dying"
Yes, they are. Their partnership with DIRECTV essentially ended when DIRECTV was purchased by News Corp (which owns NDS who already has a PVR solution); the D20 and other upcoming DIRECTV PVRs are based on NDS technology, not TiVo. As for cable companies, their only major deal has been with Comcast, and right now it looks like they won't even have product available until well into 2006. Comcast also already has their own PVR solution, and is already testing a Microsoft TV Foundation PVR, so TiVo probably isn't getting much in the way of royalties from the Comcast deal.
What is TiVo's competitive advantage over integrated PVRs? Yes, their software is better, but most people would rather have a lame integrated PVR than have to deal with the hassle of an IR changer cable.
I hate to break it to the Slashdot zealots, but here it is:
Microsoft wants one thing and one thing alone: money. It is not in Microsoft's best interests to restrict development for Windows. It is not in their best interests to break compatibility with older software. Neither of these things will EVER happen at Microsoft because the strength of their platform lies in its software library and ease of development. This news has NO implications for FOSS on Windows.
This article, in fact, is 90 percent FUD. DRM has existed for a long time in the shareware and commercial software world - this just standardizes it and provides centralized downloading and key storage. Not a bad idea, if I say so myself.
At my university (University of Colorado at Boulder), we have an on-campus 33MW power plant. Waste heat from the plant is piped around campus to heat buildings, and the electricity generated is enough to power the campus, with 8MW left over that is sold to the grid. The facility also produces chilled water through a massive vapor-phase system that is used to cool the physics and other nearby buildings.
"In monorail, the train wraps the rail. That means that it can not jump it. In contrast, think about how many of trains that we hear have jumped the track. If you follow the news, it happens every month or so."
Crap crap crappety crap. You cannot compare monorail derailments to train derailments because, if you actually compare the number of hours of service of trains vs. monorails, trains so dramatically outstrip monorails that a comparison is irrelivent.
Looking at derailments is like looking at airplane crashes. Think, for example, of how many people DIE in car crashes every day. Few of those are newsworthy. Major train derailments are newsworthy precicely because they are so rare.
Monorails simply do not have a substantial enough track record to compare their safety to conventional trains.
POP3 SUCKS for webmail. You don't get any sync features, so forget about keeping your folders consistant accross computers. Forget about having folders at all, in fact, because POP3 doesn't support them.
IMAP is a much better protocol. Until GMail supports it, I'm sticking with FastMail.
Bullshit. GSM-1900 and GSM-850 are perfectly functional. Using different frequencies doesn't make the system any less robust or functional, and quadband phones make frequency differences irrelevant.
I hate to break it to you, but Firefox has had way more than one hole in the last year:
http://secunia.com/product/4227/
23 since the release of 1.0. We're now on 1.07. Seven major security releases.
Is it better than IE? Probably. Firefox vulnerabilities tend to be fixed more quickly and are less secure than IE vulnerabilities. That said, IE is considerably easier to update for both corporate users and home users.
Firefox can do better. Fortunately, the update mechanism is much-improved in 1.5. But, remember, Firefox is supposed to be the darling of the Open Source movement. We can do better.
802.11b chipsets are already pretty cheap, and in many cases schools already have network infastructure instealled. If you can do an 802.11b scanner for $50, why not a wireless "clicker"?
At my university, we use H-ITT "clickers". They are the crappiest pieces of crap that I have ever used.
Don't you mean iPod Flea?
Which is more likely: someone cracking the encryption (which, even in 20 years, will require time, resources, and experise far beyond what most criminals are capable of), or having the card stolen through a mugging or pickpocketing.
Door locks work because they require criminals to have more time and more resources. Picking a lock takes specialized equipment and knowledge. Breaking down a strong door requires time and heavy equipment.
The more work a crime requies, the less attractive the payout becomes. Credit card theft is an unattractive crime because it is difficult to actually get away with:
- Buying goods in a store means that you are videotaped. You also have to come up with a plausible fake card (unless you steal the original - difficult to do without the victim knowing). Moreover, if the card is reported stolen, you may be arrested on the spot.
- Buying goods online requires a shipping address.
- Cash advances mean getting photographed at an ATM, and they tend to raise fruad alerts
- Selling stolen goods is always a problem
The security behind this system accomplishes the same thing as a decent lock. Is it unbreakable? Absolutely not. Is it enough to deter most fruad? Yes.
ISO/IEC 14443 has two-factor authentication. You can't steal the card number because the card doesn't transmit the card number.
12-year-old busted after realizing that ISO/IEC 14443 uses two-factor authentication: Classic.
The RF component of these cards is considerably more secure than even the magstripe component.
The MasterCard system, like all of its type, uses the ISO/IEC 14443 contactless smartcard standard.
ISO 14443, unlike most RFID standards, is a cryptographically strong system that renders easedropping useless.
Yes, but AMD has a history of smooth transitions. They aren't the first, but they ramp more quickly and with fewer roadblocks than nearly anyone else.
Oh, and Fab 30 isn't going away anytime soon. Not until Fab 36 can meet capacity, anyway.
Nintendo is no longer a player in the industry.
The new controller makes porting XBOX 360/PS3 games extremely difficult. Don't expect to see Madden for the Revolution.
Newer cars have cleaner emissions, more power, ride better, and are considerably safer.
ABS is a valuable system when it is used properly. It allows the driver to steer while braking on slick surfaces, hopefully avoiding a collision.
Older cars may be eaiser to fix, but they break considerably more often. And, no, they aren't "built more solid".
New cars do have crumple zones, but those are designed to decrease the forces acting on you during a crash. They are one of the most important safety advances in automotive history.
Don't give me this shit about "skill in driving". If you want to test your driving skill, try driving a F1 racecar. Otherwise, people neither want nor need "driving skill". Driving safely is about knowing and following the rules, using good judgement, and reacting properly in panic situations. It has nothing to do with whether the vehicle has fuel injection, or whether the engine is computer controlled, or whether the car has crumple zones. We are not "dependant" on airbags in the same way that an airplane is not "dependant" on parachutes.
"two years before Vista launch"
RTM is November 15, 2006. 428 days, or 1.17 years.
And, if the entire OS is like Microsoft Max, perhaps it won't suck afterall.