"Failure rates are known to be highly correlated with drive models, manufacturers and vintages. [...] However, in this paper, we do not show a breakdown of drives per manufacturer, model, or vintage due to the proprietary nature of these data."
The meaning of a digital signature on a piece of data depends on the signing entity and the data. My signature on an email means I wrote the email. My signature on someone's public key doesn't mean I "wrote" the public key--it means I've verified that the name and picture on a state-issued ID card match the individual who has access to the associated private key and that I've known the person for a certain period of time. Very different meanings.
Whether or not it is reasonable for a distro maintainer to sign a blob from nVidia depends on what that signature would mean.
Maybe it means the distro maintainer has verified that the blob came from an employee of nVidia authorized to distribute official drivers.
Maybe it means the distro maintainer has tested the blob for a certain number of hours in certain hardware environments without experiencing crashes.
Maybe it means the distro maintainer has been granted exclusive access to the source code for the blob, has audited it, and has certified that it has passed manual and machine inspections.
Maybe it means the distro maintainer has been issued a legal document binding nVidia to provide financial compensation if the blob is later found to contain spyware.
Different people have different preferences about how distro maintainers should behave.
So what if the people buying them cannot balance their checkbooks or figure out that $50/month for 60 months is way more expensive than a $1000 one time fee
We're already there--what do you think a 2-year cell phone contract is? "Save $450 on the handset by agreeing to pay an extra $50/mo for 24 months."
You bring up a very interesting point here--one that was conspicuously absent from the article and other comments.
Could the reason for this perspective be that articles about animal experimentation are written by and for people who have already largely decided that the benefits of animal experimentation outweigh the drawbacks? Is applying labels to your audience such as "stupid", "moronic", "sociopathic" and "nutcases" the most effective way to engage them in a constructive debate that will eventually result in a reduction in this type of hurtful behavior?
This is slashdot. Everyone here understands how it feels to perceive an injustice and be emotionally overwhelmed by it. That said, please consider revising your argumentative strategy to encourage constructive conversation.
To answer one of your questions, I do think it would be a good idea to produce and distribute video footage of this research. Here are some reasons:
It would encourage researchers to think twice about whether their actions would be considered appropriate and justifiable by the general public. My ethics manual at work offers the following simple test to help employees judge the appropriateness of their actions: "Would you do it if your coworkers and family were watching?"
It would improve the accuracy of the research. When performing experiments, it is important to keep detailed notes on the procedure being performed in order to be able to duplicate the results at a later time, or for other researchers to be able to duplicate the results. Nevertheless, it is possible to make errors or omissions in these notes, and that creates problems. An objective video recording could remedy that situation.
It would improve the accessibility of the research. Many people can understand a process better by seeing it done than they can by reading about it.
If you think these are good points, please feel free to reuse them. If you intend to continue arguing in favor of this kind of oversight, it would also be a good idea to anticipate some of the arguments your opponents might make against video recording so you can be prepared to counter them. Some of those arguments might be:
Video records are more expensive and difficult to record, store and distribute than text-based notes and pictures.
In particular, scientific journals cannot publish video recordings without attaching a CD or DVD to the issue or maintaining a companion web site. Whether or not you believe scientific journals are the best way to distribute research, it cannot be denied that they are popular and important.
Video recordings that can be used to identify the individual researchers could make it easier for criminally inclined animal rights activists to perpetrate acts of violence.
If you're comfortable with scripting, you may want to write your own android app to do just what you need. Just make sure you buy an android phone that can install non-market applications. To test this in the store, ask to see a running phone of the model you're considering, and follow this procedure:
1. From the main screen, tap the menu button.
2. Tap Settings.
3. Tap Applications.
4. Look for a checkbox that says "Unknown sources - Allow installation of non-Market applications", and make sure you can enable it.
I realize this solution isn't for everyone. You would have to learn Java if you don't know it and learn Android programming, which isn't that hard. Otherwise, you could set up a special server with a regular OS and run your scripts there. If there are exceptional conditions, your server scripts could send an email to a special email address. You can quite easily configure your android phone to play a special ringtone when your "exceptional condition" email account has new mail.
As far as easily attaching sounds to an email, you may have to roll your own app for that. Someone more knowledgeable than me may know of an app to do this, though.
I'm curious: what do you mean by "1 euro max per day for domestic calls"? Do you pay by the minute for domestic calls? Also, what do you mean by "14mbit unlimited data"? In the US, carriers often claim to offer unlimited data but are usually lying. Either they start throttling speeds after a certain amount of data transfer, or they stop all data access claiming that the subscriber is abusing the network by transferring too much data.
Thank you both for sharing. As someone who pays $150/month in the US for no-frills TV, telephone and 10/1 Internet with a 50GB/month transfer cap, it's eye-opening to read about conditions in other parts of the world.
Also, I'm sorry about how the United States has been behaving towards other countries for the past decade or so.
Then it turns their desktop to crap because it won't even start because the graphics configuration is bolloxed and they end up spending half an hour screwing around reinstalling the graphics driver or hunting the NVidia site for a replacement.
Half an hour... How things change. Before cell phones could access the Internet and I had multiple computers in my house, an X failure meant trying to remember how to connect to the Internet from the console and then searching for a solution using lynx. That was not a half hour process by any stretch of the imagination.
Reading your comment, I got a feeling like I did when a kid asked me why the drive letters in Windows start with C: instead of A:.
What do you call that feeling? It's not quite nostalgia; I'm glad the days of editing XFree86Config with ed and storing data on floppies are gone for good. Thanks, I guess, for reminding me how far we've come?
I know it is futile to ask people to read an article before they comment on it, and I know it is equally futile to ask people who submit articles such as this to post links to original articles instead of second or third sources
I don't believe it's futile.
If you are submitting an article about an article in a scientific journal, please include a link to the original article in the original journal instead of a newspaper article based on a press release announcing the publication of the article.
If I were submitting an article about an article in a scientific journal to slashdot, it wouldn't occur to me to look for (or follow) a link to the original article. Whenever I search for information on a topic online and there is a link in the search results to an article in a scientific journal, that article is almost always behind a paywall. Even more frustrating, it's usually set up as a tease so that it *looks* like it's a link to the full article but turns out to be a page trying to sell me a subscription to their service (or $19.99 for three days of access to just that one article). I avoid these links due to how frustrating that experience is most of the time.
the official press release from the journal is available [bmj.com] and the full article itself [bmj.com] are available online
You seem to be informed about this sort of thing, and given how astonishing and tantalizing the prospect of full-text access is to me, I would appreciate some suggestions. Is there a way I can change my searching behavior so that I don't run into paywalls disguised as the full article? Is there a way I can change my behavior in general so that more full-text articles are available for free online?
I wish it was possible to require SSH keys for some (or even all) users to have a passphrase, and enforce this requirement on the server.
As it stands right now, even if you generate a key for someone with a pass phrase, they can remove it easily on the client side and the server has no way of knowing. This means you could have passwordless logins to remote systems. Not good.
Such a policy would require the server to take the client's word for it that the private key was encrypted with a passphrase.
At least with modern systems and key agents you can get passwordless ease of use once you log into your local account, and if someone happens to get your private key they don't immediately have instant access to the machines you can log into. You should have a little time to secure the machines. [Think lost/stolen laptop or backup drive.]
Agreed. If someone is removing the passphrase from their private key, there is some other problem that needs to be solved. Personally, I like ecryptfs for my home directory and LUKS for my backup drive with the LUKS passphrase inside my login keyring.
I can't speak for other people, but it trips me out to be walking through the supermarket chatting online. My inner ten-year-old is so happy. Other than that, I like using it as a music player on trips, an audiobook player, an ebook reader and a geocaching toy,
I had no idea you could move the menu, minimize, maximize, and close buttons back to where they were in previous releases. I've been suffering with the new system since Ubuntu 10.04. Thanks for the tip!
I'll take your word for it that the ballot has been rigged so that other parties have to waste efforts trying to get approval to appear. However, the reasoning I have used to vote for one of the two major parties goes like this:
1. Either major party X or major party Y is definitely going to win the election. 2. Both X and Y are pretty bad, but X is better than Y most of the time. 3. I'll vote for the X, the lesser of two evils.
I'm certain this is the way I make the decision. I'm pretty sure this is the way the rest of my family does it as well. When I talk to people who are disappointed by the current two party system, this is the reasoning that they articulate to me.
The could have budgeted some money for future updates to the Linux game. Just look at what happens to most closed source/commerical Linux native games:
1. The game is released. It may run well on most contemporary systems. 2. Time passes and critical libraries break ABI compatibility. (I'm looking at you, glibc). 3. The game no longer runs on modern systems.
For good longevity, a closed source game would need some kind of ABI shim between the closed executable and the multitude of unstable open source ABIs. For most games, this shim is called "wine" and that state of affairs makes me sad.
(Yes, believe it or not, there ARE other distros; although it is hard to tell since so many stories and postings say "Ubuntu" in place of the word "Linux" or "Linux distribution")
Isn't it great? I can't wait until the days of users asking, "So I should try Linux. Which distro should I use?" and getting useless or contradictory answers are long forgotten.
nVidia was the best for a while and in my opinion still is. Now Intel has great FOSS drivers. If you want performance, ATI's FOSS drivers are coming along. Once they're fully functional, I see no reason to buy nVidia cards ever again.
Hmm. I didn't demand the wallet from the other person, although I do see your point.
I guess it depends on where you grew up. I could have said (and did say) something like that without the expectation that it would come to blows. My "fight" response was totally not engaged when I opened my mouth.
If the person had pulled a knife and said, "Oh yeah? Come take it from me!" then I would have let him have it, same as any armed robbery.
5) "We're not splitting the money. That's stealing, and if you try it with the other guy I'm calling the cops. Now, let's find a phone book and call the owner."
And before you say nobody would ever do that, I've done exactly that in the past. I'm not sure if it was going to be a con, but it worked out well enough in the end.
For a long time Microsoft has had a package called Services For Unix that you can install on Windows. It allows Windows to act as a server but not a client with respect to standard *nix protocols like NFS.
Microsoft wants to replace *nix in the server space by breaking into purely *nix environments and replacing an entrenched server operating system with their operating system.
Whether this is done by making Windows interoperable with the protocols that are already on the clients or changing the clients to interoperate with Windows as a server is immaterial.
Unless they're making it easy for people to replace Windows AD servers with Samba servers running on Linux, this is not a big deal.
Unmanned missions may be cheaper and safer, but sending out real people to expand the horizons of human activity in space is much more important. It gets people excited! That brings in money and inspires young people.
Then, when NASA has a huge group of talented experts and tons of cash, they can do real science instead of worrying every day about whether the budget will get slashed before they can complete the current round of experiments.
As a Linux user I have the opposite frustrations when I come to use Windows. "Why do I have to search the web to find a piece of software to download? Why can't I just go to 'Add/Remove Programs', type in the name (or a keyword) and click install?", "Why can't I chose a different desktop environment when I log in?", "Why can't I use the command line to do even basic stuff?"
Disclaimer: Ubuntu is the only OS on my personal computer.
Answers: You have to search the web rather than using a built-in application catalog for a couple of reasons. First, Windows developers aren't forced to beg the Ubuntu repo maintainers for their holy blessing in order to get an app into the catalog that appears when you launch the "Add/Remove Programs" utility. Second, because Windows dependably installs a particular set of libraries by default, there aren't the same dependency hassles on Windows as there are on Linux*, so Windows doesn't need an elaborate dependency-handling system backed by internet package repositories. Third, Windows doesn't need an app to be recompiled when you upgrade, say, from DirectX 8 to DirectX 9, so there's no need to constantly refresh the binaries on your system from a central repo where all of these programs have been meticulously twiddles to cooperate with eachother.
The enormous Debian/Ubuntu repository system that gobbles up people-hours in maintenance time and the graphical package search/installer tool is a band-aid on top of horrendous dependency and compatibility issues endemic to GNU/X11/Linux operating systems.
*Seriously, folks. Put away your Windows 95 stories and try downloading any ol' app for Windows XP. Double click the installer. Run the program. Simple. Done.
The reason you can't choose a different desktop environment when you log in is that you can choose a different desktop environment when you log in. If you really want to install (whatever the win32 port of GNUStep is called), it has instructions for making it the default shell. I think KDE has a (really old) windows port as well.
The reason that alternative graphical shells aren't installed by default, or the reason that selecting a different graphical shell is not a big button on the login screen, is because the default one is really, really good. It's the little touches, like not having default panels that suck up precious screen real estate and not having a schizophrenic clock applet that jumps around the panel each time I try to log in. And multiple optional graphical shells would only fragment the user base.
The reason you can't use the command line to do even basic stuff is that the Linux command line is a whole lot better than the Windows command line.
Slashdot Mirror
Here's my favorite part of the paper:
:/
"Failure rates are known to be highly correlated with drive models, manufacturers and vintages. [...] However, in this paper, we do not show a breakdown of drives per manufacturer, model, or vintage due to the proprietary nature of these data."
Thanks, Google.
The meaning of a digital signature on a piece of data depends on the signing entity and the data. My signature on an email means I wrote the email. My signature on someone's public key doesn't mean I "wrote" the public key--it means I've verified that the name and picture on a state-issued ID card match the individual who has access to the associated private key and that I've known the person for a certain period of time. Very different meanings.
Whether or not it is reasonable for a distro maintainer to sign a blob from nVidia depends on what that signature would mean.
Maybe it means the distro maintainer has verified that the blob came from an employee of nVidia authorized to distribute official drivers.
Maybe it means the distro maintainer has tested the blob for a certain number of hours in certain hardware environments without experiencing crashes.
Maybe it means the distro maintainer has been granted exclusive access to the source code for the blob, has audited it, and has certified that it has passed manual and machine inspections.
Maybe it means the distro maintainer has been issued a legal document binding nVidia to provide financial compensation if the blob is later found to contain spyware.
Different people have different preferences about how distro maintainers should behave.
So what if the people buying them cannot balance their checkbooks or figure out that $50/month for 60 months is way more expensive than a $1000 one time fee
We're already there--what do you think a 2-year cell phone contract is? "Save $450 on the handset by agreeing to pay an extra $50/mo for 24 months."
Could the reason for this perspective be that articles about animal experimentation are written by and for people who have already largely decided that the benefits of animal experimentation outweigh the drawbacks? Is applying labels to your audience such as "stupid", "moronic", "sociopathic" and "nutcases" the most effective way to engage them in a constructive debate that will eventually result in a reduction in this type of hurtful behavior?
This is slashdot. Everyone here understands how it feels to perceive an injustice and be emotionally overwhelmed by it. That said, please consider revising your argumentative strategy to encourage constructive conversation.
To answer one of your questions, I do think it would be a good idea to produce and distribute video footage of this research. Here are some reasons:
If you think these are good points, please feel free to reuse them. If you intend to continue arguing in favor of this kind of oversight, it would also be a good idea to anticipate some of the arguments your opponents might make against video recording so you can be prepared to counter them. Some of those arguments might be:
Good luck in your future debates!
If you're comfortable with scripting, you may want to write your own android app to do just what you need. Just make sure you buy an android phone that can install non-market applications. To test this in the store, ask to see a running phone of the model you're considering, and follow this procedure:
1. From the main screen, tap the menu button.
2. Tap Settings.
3. Tap Applications.
4. Look for a checkbox that says "Unknown sources - Allow installation of non-Market applications", and make sure you can enable it.
I realize this solution isn't for everyone. You would have to learn Java if you don't know it and learn Android programming, which isn't that hard. Otherwise, you could set up a special server with a regular OS and run your scripts there. If there are exceptional conditions, your server scripts could send an email to a special email address. You can quite easily configure your android phone to play a special ringtone when your "exceptional condition" email account has new mail.
As far as easily attaching sounds to an email, you may have to roll your own app for that. Someone more knowledgeable than me may know of an app to do this, though.
I'm curious: what do you mean by "1 euro max per day for domestic calls"? Do you pay by the minute for domestic calls? Also, what do you mean by "14mbit unlimited data"? In the US, carriers often claim to offer unlimited data but are usually lying. Either they start throttling speeds after a certain amount of data transfer, or they stop all data access claiming that the subscriber is abusing the network by transferring too much data.
Thank you both for sharing. As someone who pays $150/month in the US for no-frills TV, telephone and 10/1 Internet with a 50GB/month transfer cap, it's eye-opening to read about conditions in other parts of the world. Also, I'm sorry about how the United States has been behaving towards other countries for the past decade or so.
Then it turns their desktop to crap because it won't even start because the graphics configuration is bolloxed and they end up spending half an hour screwing around reinstalling the graphics driver or hunting the NVidia site for a replacement.
Half an hour... How things change. Before cell phones could access the Internet and I had multiple computers in my house, an X failure meant trying to remember how to connect to the Internet from the console and then searching for a solution using lynx. That was not a half hour process by any stretch of the imagination.
Reading your comment, I got a feeling like I did when a kid asked me why the drive letters in Windows start with C: instead of A:.
What do you call that feeling? It's not quite nostalgia; I'm glad the days of editing XFree86Config with ed and storing data on floppies are gone for good. Thanks, I guess, for reminding me how far we've come?
I know it is futile to ask people to read an article before they comment on it, and I know it is equally futile to ask people who submit articles such as this to post links to original articles instead of second or third sources
I don't believe it's futile.
If you are submitting an article about an article in a scientific journal, please include a link to the original article in the original journal instead of a newspaper article based on a press release announcing the publication of the article.
If I were submitting an article about an article in a scientific journal to slashdot, it wouldn't occur to me to look for (or follow) a link to the original article. Whenever I search for information on a topic online and there is a link in the search results to an article in a scientific journal, that article is almost always behind a paywall. Even more frustrating, it's usually set up as a tease so that it *looks* like it's a link to the full article but turns out to be a page trying to sell me a subscription to their service (or $19.99 for three days of access to just that one article). I avoid these links due to how frustrating that experience is most of the time.
the official press release from the journal is available [bmj.com] and the full article itself [bmj.com] are available online
You seem to be informed about this sort of thing, and given how astonishing and tantalizing the prospect of full-text access is to me, I would appreciate some suggestions. Is there a way I can change my searching behavior so that I don't run into paywalls disguised as the full article? Is there a way I can change my behavior in general so that more full-text articles are available for free online?
I wish it was possible to require SSH keys for some (or even all) users to have a passphrase, and enforce this requirement on the server.
As it stands right now, even if you generate a key for someone with a pass phrase, they can remove it easily on the client side and the server has no way of knowing. This means you could have passwordless logins to remote systems. Not good.
Such a policy would require the server to take the client's word for it that the private key was encrypted with a passphrase.
At least with modern systems and key agents you can get passwordless ease of use once you log into your local account, and if someone happens to get your private key they don't immediately have instant access to the machines you can log into. You should have a little time to secure the machines. [Think lost/stolen laptop or backup drive.]
Agreed. If someone is removing the passphrase from their private key, there is some other problem that needs to be solved. Personally, I like ecryptfs for my home directory and LUKS for my backup drive with the LUKS passphrase inside my login keyring.
I can't speak for other people, but it trips me out to be walking through the supermarket chatting online. My inner ten-year-old is so happy. Other than that, I like using it as a music player on trips, an audiobook player, an ebook reader and a geocaching toy,
This is really cool. Now when one of these filesystems becomes stable in Linux, we'll be ready to go. Look out 2025--here I come!
I had no idea you could move the menu, minimize, maximize, and close buttons back to where they were in previous releases. I've been suffering with the new system since Ubuntu 10.04. Thanks for the tip!
I'll take your word for it that the ballot has been rigged so that other parties have to waste efforts trying to get approval to appear. However, the reasoning I have used to vote for one of the two major parties goes like this:
1. Either major party X or major party Y is definitely going to win the election.
2. Both X and Y are pretty bad, but X is better than Y most of the time.
3. I'll vote for the X, the lesser of two evils.
I'm certain this is the way I make the decision. I'm pretty sure this is the way the rest of my family does it as well. When I talk to people who are disappointed by the current two party system, this is the reasoning that they articulate to me.
The could have budgeted some money for future updates to the Linux game. Just look at what happens to most closed source/commerical Linux native games:
1. The game is released. It may run well on most contemporary systems.
2. Time passes and critical libraries break ABI compatibility. (I'm looking at you, glibc).
3. The game no longer runs on modern systems.
For good longevity, a closed source game would need some kind of ABI shim between the closed executable and the multitude of unstable open source ABIs. For most games, this shim is called "wine" and that state of affairs makes me sad.
Yes, choice, variety, and competition are horrible things aren't they?
They have real disadvantages. http://www.columbia.edu/~ss957/whenchoice.html
(Yes, believe it or not, there ARE other distros; although it is hard to tell since so many stories and postings say "Ubuntu" in place of the word "Linux" or "Linux distribution")
Isn't it great? I can't wait until the days of users asking, "So I should try Linux. Which distro should I use?" and getting useless or contradictory answers are long forgotten.
nVidia was the best for a while and in my opinion still is. Now Intel has great FOSS drivers. If you want performance, ATI's FOSS drivers are coming along. Once they're fully functional, I see no reason to buy nVidia cards ever again.
Hmm. I didn't demand the wallet from the other person, although I do see your point.
I guess it depends on where you grew up. I could have said (and did say) something like that without the expectation that it would come to blows. My "fight" response was totally not engaged when I opened my mouth.
If the person had pulled a knife and said, "Oh yeah? Come take it from me!" then I would have let him have it, same as any armed robbery.
You're missing a choice:
5) "We're not splitting the money. That's stealing, and if you try it with the other guy I'm calling the cops. Now, let's find a phone book and call the owner."
And before you say nobody would ever do that, I've done exactly that in the past. I'm not sure if it was going to be a con, but it worked out well enough in the end.
I'm not suprised.
For a long time Microsoft has had a package called Services For Unix that you can install on Windows. It allows Windows to act as a server but not a client with respect to standard *nix protocols like NFS.
Microsoft wants to replace *nix in the server space by breaking into purely *nix environments and replacing an entrenched server operating system with their operating system.
Whether this is done by making Windows interoperable with the protocols that are already on the clients or changing the clients to interoperate with Windows as a server is immaterial.
Unless they're making it easy for people to replace Windows AD servers with Samba servers running on Linux, this is not a big deal.
Now think of the names of the current ISS inhabitants.
Just being in orbit isn't cool anymore. That's why missions like the one in this story are important.
There's no question the robots get you more science for your buck
The problem is that if you don't have enough bucks, you can't do much science. Manned missions, on the other hand, get you more buck for your buck.
Unmanned missions may be cheaper and safer, but sending out real people to expand the horizons of human activity in space is much more important. It gets people excited! That brings in money and inspires young people.
Then, when NASA has a huge group of talented experts and tons of cash, they can do real science instead of worrying every day about whether the budget will get slashed before they can complete the current round of experiments.
Disclaimer: Ubuntu is the only OS on my personal computer.
Answers:
You have to search the web rather than using a built-in application catalog for a couple of reasons. First, Windows developers aren't forced to beg the Ubuntu repo maintainers for their holy blessing in order to get an app into the catalog that appears when you launch the "Add/Remove Programs" utility. Second, because Windows dependably installs a particular set of libraries by default, there aren't the same dependency hassles on Windows as there are on Linux*, so Windows doesn't need an elaborate dependency-handling system backed by internet package repositories. Third, Windows doesn't need an app to be recompiled when you upgrade, say, from DirectX 8 to DirectX 9, so there's no need to constantly refresh the binaries on your system from a central repo where all of these programs have been meticulously twiddles to cooperate with eachother.
The enormous Debian/Ubuntu repository system that gobbles up people-hours in maintenance time and the graphical package search/installer tool is a band-aid on top of horrendous dependency and compatibility issues endemic to GNU/X11/Linux operating systems.
*Seriously, folks. Put away your Windows 95 stories and try downloading any ol' app for Windows XP. Double click the installer. Run the program. Simple. Done.
The reason you can't choose a different desktop environment when you log in is that you can choose a different desktop environment when you log in. If you really want to install (whatever the win32 port of GNUStep is called), it has instructions for making it the default shell. I think KDE has a (really old) windows port as well.
The reason that alternative graphical shells aren't installed by default, or the reason that selecting a different graphical shell is not a big button on the login screen, is because the default one is really, really good. It's the little touches, like not having default panels that suck up precious screen real estate and not having a schizophrenic clock applet that jumps around the panel each time I try to log in. And multiple optional graphical shells would only fragment the user base.
The reason you can't use the command line to do even basic stuff is that the Linux command line is a whole lot better than the Windows command line.