I honestly have no idea how this would work. It seems like if it did, it would be a magic bullet against all computer criminal charges.
Which is why it makes it difficult. It's ridiculously easy to generate a logfile which says whatever you want it to say. Why are logfiles considered proof in criminal cases?
Depends, especially, whether what they said is independently verifiable, and how much effort you're willing to put into that independent verification.
I mean, what are you going to do? Look it up on Snopes?
We ended up not going for completely different reasons, but I suspect that everyone in my (admittedly small) company would agree that this was funny.
Might not be funny if they cut off a larger speech, but if it was just a bunch of booths with demos running, we can always restart the demos. And understand, it's a whole wall of TVs going off -- it seems incredibly unlikely that prospective customers or partners would assume it's our fault, when all of our neighbors had TVs go off, too.
From now on, no one with an infrared-controlled device at a tradeshow is going to leave it exposed. A few tabs of black electrical tape will thwart TV-B-Gones.
On the good side, these "schemes" tend to be found and revealed really quickly these days.
On the bad side, damage has still been done. There are generally going to be more people who hear the initial message than hear of the source being debunked.
I'm inviting a flame war here, but isn't math - at least in the sense that we teach it - an artificial construct that we invented to describe our observations? Math can contain theories, but I don't think it could ever be classified as one because we actually know what it is.
Math can contain theorums. It might contain theories, but these are irrelevant -- in pure mathematics, if you know something, you know it, and you have a mathematical proof.
I've said before that there are certain truths that are not a matter of point of view -- not open for reasonable debate. Two plus two equals four. If you disagree with that statement, you're either stupid or insane.
Well, look into mathematical proofs. Not only is two plus two equal to four, but people have actually proved the principles of basic addition, or at least worked out what basic assumptions you need for them to be true.
Isn't gravity demonstrable? If I understand correctly, there's still a lot of uncertainty in the scientific community about how gravity works exactly, but it's clearly an observable and demonstrable fact that it does.
Not quite. Here's how it works:
We have observed things to fall in the past. If we were to be rigorous, we'd say that on a certain date, at a certain time, in a certain location, we dropped a certain item and watched it fall towards the Earth.
Newton came up with some mathematical equations which describe this phenomenon. Note: Not "explain", but "describe". But of course, they describe more than one object falling in one place at one time -- they describe that any object will fall anywhere, and, more generally, that objects attract.
We call this a "hypothesis" -- it predicts something (things fall, and they accelerate at this rate). Most people consider gravity to be sufficiently tested -- we've worked out (with Newton's equations) that some particular thing should fall, and how fast -- and, similarly, that something we launch into space will orbit (or not), and how fast, and whether it will stay even with the Earth.
When a hypothesis is sufficiently tested, we accept it tentatively as a "theory". That's all gravity is -- a temporary theory until we find something better. And we did -- Einstein's General Relativity actually disproved Newton's Law of Universal Gravitation.
Maybe I'm living under a rock here, but I've never really seen evolution demonstrated.
Well, correcting you again, but this is the "hypothesis" of evolution tested.
And it can and does provide very specific predictions. It predicts that certain fossils will be found -- and we then went and found those fossils, and they were pretty much what we expected. It predicts that if you develop a really effective mosquito spray, and any of the mosquitoes are immune, those mosquitoes will survive, thus showing that the mosquito population as a whole has adapted to that spray.
I'm sorry if you were expecting a Borg-like adaptation, where, in the span of a few minutes of screen time, an organism adapts to a particular threat.
That's why it's a theory, because we can only infer, we can't demonstrate, and unfortunately no one actually witnessed it.
Well, no one has witnessed quite a lot of things that we have theories for. You know, crack open a science textbook -- do you think anyone has actually seen an electron? No, we can only infer.
For all you know, there's infinite monkeys in your computer making it go. Or maybe it's the Hand of God Himself. But we consider computers, among other things, to be evidence for -- not proof of, but evidence for -- the existence of electrons.
And it's generally not useful to suggest alternate ideas that have no testable predictions. For instance, if you said my computer operates because of the hand of God, that's a fine idea, but there's absolutely nothing that predicts. It doesn't tell me w
The very nature of Science is debate and inquiry. We know what we know because of how well it survives that debate.
Evolution works because there is no other theory that explains what we've observed. Intelligent Design does not count, because it provides no testable predictions.
But when you say things like:
You can't reject evolution any more than you can reject combustion, or gravity. If people DO reject it, they are simply being ignorant and stupid.
All depends on why you reject it. Everyone I've seen reject evolution has done so for stupid, dogmatic reasons, but you are making stupid, dogmatic statements about science.
I happen to agree with your conclusions, somewhat: Creationism has no place in science. But the reasons you've given for that are not scientific.
gimme a break, for me this would be a huge no-no, it would be the hallmark of somebody going after every possible latest fad, instead of focusing on proven tools for the job. Yes, there ARE cases where the bleeding edge is needed,
The point is not that the bleeding edge is needed, but that it teaches you things -- people take the good ideas from Ruby on Rails, for instance, and fold them back into their own favorite environments, even Java.
You're IT department consists of you and you're imaginary friend doesn't it.
Actually, just me, because, as I said, there is no IT department where I work. Everyone is responsible for their own machine. We can do this because there are five of us, and we're all developers.
Have you actually worked with real employees? My guess is no
Guess again.
and they probably would want to kick your ass if you did.
That depends on whether I get to set the policy.
Well-run corporations will make their IT department the final word on what goes on a computer and what doesn't. Maybe the CEO can come in and demand that something be done, but users don't have that luxury. If they're on Windows, they are on the corporate domain, their auto-updates are enabled and cannot be disabled, and they will update from the internal company WSUS server, which means IT decides when how, and to whom each update is applied.
The difference here, if I understand this correctly, is that in Linux, you have to run through every computer and add cron job by hand.
Except in Linux, just about any task you can do by hand, you can automate. There are many scripts for deploying configuration to a large number of Linux machines. (Directly -- the lingo is "push", not "pull".) Ruby On Rails seems to like Capistrano, though that's more designed around deploying a Rails app.
But hey, you already control the repository, why not roll your own install disc? I know I could never admin a Windows domain without a custom (nLite'd) install disc, and/or a standard disk image to start from...
Or if you wanted just one place to do it, add that cron job to the base install provided by your repository...
In Windows, when you join corporate domain this all is done automatically.
Wait, so on Linux, all I have to do is add the cron job and change the repository. On Windows, I have to join the corporate domain.
Again, I've never done this on Windows, so I don't really know how long it takes to join the corporate domain, on each computer. On Linux, it would take me maybe five minutes per machine, including boot time.
No matter what solution you adopt to deploy stuff, it's going to take some setup.
mismatches between them may provide a way of catching thinkos
I kind of doubt that, if you're using something which formats automatically. Even vim can autoindent, which reduces them back to useless redundancy.
And it helps (in properly formatted code) readability
I'd debate that. I find Python plenty readable.
There is one feature I'd miss: bouncing between matching brackets in vim.
The bad think about having brackets as syntax and whitespace as just formatting is that it is redundant, and you can produce things that look wrong but are wrong but look right if you don't have something automatically formatting.
Right. And if you do have something automatically formatting, I just don't see how the extra curlies make a difference (except bouncing in vim).
Personally, it's a moot point for me -- there are too many other big things wrong with Python (the GIL comes to mind) for the whitespace to make much difference.
Is the bookmarking implemented by the DVD player firmware or by the java/fancy menu stuff that comes on the dvd?
By the Javascript. (Yes, there is a huge difference. I love Javascript, can't stomach Java.)
It sounds great if it works the same on every dvd. But if the implementation is different on each title (or from each studio) then it might be more confusing then useful?
Maybe so, but the point is, the players did not have this feature to begin with. It is therefore possible for studios to invent and deliver new and useful features. Some of them are tied to the movie, most of them are gimmicky (like Transformers having a HUD showing the "health" of each robot)...
Until the server goes away. Or they decide that what you paid for it originally is no longer enough to warrant further downloading.
It's possible to design a Blu-Ray disc that does both of these, although generally, the first part won't be true.
Or wait -- maybe it's not possible yet? I know it's possible to design an HD-DVD disc that way...
Storage and maintenance and management and constant connectivity and device usability and persistence of the company selling your product.
Ah, but you also said something about "will not help you obtain Sanctuary any faster". I don't really see how anything other than bandwidth contributes to the speed at which I obtain Sanctuary.
Nope. Otherwise you couldn't burn your own Blu-Ray media. Does it really sound like companies would do that, the same companies that make camcorders?
Yes, it does. Sony has released DVDs with copy protection that their own DVD players couldn't deal with.
It's also possible that you could burn your own Blu-Ray discs, but only through their software, which applies some level of DRM to them anyway.
I know how it works for HD-DVD: There are some features which you can only access (local storage, network) if your disc is AACS-encrypted. There were plans to make it possible to only sign a disc, but access the same feaures.
Nor do I, the "value" is the content. In the end I do enjoy some content produced and all things being equal I'd rather reward the people that produce it as best I can.
So would I. And when I can view that content on Linux on my desktop, maybe I will. That's one reason Sanctuary has my enthusiastic support.
Like what - using standard USB for connected devices? Supporting Linux on the PS3? Perhaps there are also reasons to support Sony.
Well, they've done far too much at this point for me to trust them. It's going to take a lot more to gain that trust back.
Specifically: Standard USB is what the Xbox 360 uses as well, and the Wii uses Bluetooth. Supporting Linux on the PS3 in a hypervisor which denies access to the 3D hardware... what's the point, really? I can think of some academic uses for it, but as a consumer, there's not really a reason to buy a PS3 over a 360, and there's more than one reason to buy a 360 over a PS3. (Not even getting started with the Wii...)
Ubuntu have been experimenting with upstart as a replacement for the traditional init lately, and I don't know for sure how compatible this is with the existing syntax.
Pretty close. They haven't been using upstart nearly as much as I think they should -- most things are still following the old SysV design.
It's possible to continue the boot process manually if you somehow find yourself within the initrd environment, including launching a root shell.
I know, just pointing out, you can't do it with the same commands. The "init=/bin/sh" line will probably work in an initrd, but the process of getting to the "real" filesystem is a lot harder. (Fortunately, for the most common case, it's possible to simply unmount the root filesystem, sync, and then ctrl+alt+del or hard reboot.)
It doesn't really matter about init not running. The command as given will launch init, going straight into runlevel 6 (i.e. reboot).
Ah, I see. Still, does it work well, considering that was not an "exec" command, and therefore, init won't be running as PID 1? I kind of didn't like the idea of leaving the bash process around, even only for a reboot.
You're right that it would be possible to remount / read-only and # exec/sbin/init with no need for a reboot -- I hadn't thought of that. Must be some vestiges of Windows badness still glooping around in my brain!
Just be thankful that you didn't have to mess with initrds as much as I did. (This was before I moved to ubuntu, where I can generate/update a stock initramfs with one command.)
they didn't develop the browser, don't have the source code for the browser,
Wait, what? Don't have source code for Firefox?
I'm waiting for an explanation as to how I'm not understanding, because if you just said what I think you said, you're claiming that your techs don't have source code to the most widely-known open source program?
And if the name of the program isn't a big deal, then how come many Linux distributions opt for generic names for programs in the launch menu?
Because unless your name is "Internet Explorer", people won't know, just from the name, what kind of program it is. I think most of us have generally accepted names like Firefox, but unless someone already knows what it is, how would you know Firefox is a web browser? Should we just be naming the first project "Web Browser", and subsequent projects like Konqueror, iCab, etc can fade into obscurity?
And how does a name like "GwenView" not violate the principle of least surprise?
I suspect GwenView is a response to a Windows image viewer, IrfanView.
iThink you may be just a bit iPredisposed to think of iLinux apps as having weird iNames.
Honestly, why is iCrap better than Krap?
Woop-de-doo...why don't we suffix the name with "Application" just to let everyone know that this program is indeed an application?
What program isn't an application?
Or at least, it's generally possible to know it's an Application from context; by knowing what it does. It's generally not possible to know, without the naming scheme, which desktop environment (or graphics library) a given application uses.
It gdoes make it a bit geasier to Kfigure out gwhich iApps go to Kwhich iDesktop Kenvironment.
My guess is that the whole reason for releasing now is the desperate hope that they'll have something decent ready for Kubuntu.
And I imagine that in another few months, it's going to be better than KDE3. By that I mean functionality; KDE3 just hasn't been particularly stable for me. (The ACID3 test crashed Konqueror, and Konqueror quite frequently crashes while I'm typing comments on Slashdot...)
Slashdot Mirror
Which is why it makes it difficult. It's ridiculously easy to generate a logfile which says whatever you want it to say. Why are logfiles considered proof in criminal cases?
Depends, especially, whether what they said is independently verifiable, and how much effort you're willing to put into that independent verification. I mean, what are you going to do? Look it up on Snopes?
Why would you want to cover a show which can't recover from as simple a prank as turning TVs off?
Generally, any story they do directly about OSTG/SourceForge/etc includes that disclaimer.
They tend to be good stories anyway.
We ended up not going for completely different reasons, but I suspect that everyone in my (admittedly small) company would agree that this was funny.
Might not be funny if they cut off a larger speech, but if it was just a bunch of booths with demos running, we can always restart the demos. And understand, it's a whole wall of TVs going off -- it seems incredibly unlikely that prospective customers or partners would assume it's our fault, when all of our neighbors had TVs go off, too.
From TFA:
On the bad side, damage has still been done. There are generally going to be more people who hear the initial message than hear of the source being debunked.
Math can contain theorums. It might contain theories, but these are irrelevant -- in pure mathematics, if you know something, you know it, and you have a mathematical proof.
I've said before that there are certain truths that are not a matter of point of view -- not open for reasonable debate. Two plus two equals four. If you disagree with that statement, you're either stupid or insane.
Well, look into mathematical proofs. Not only is two plus two equal to four, but people have actually proved the principles of basic addition, or at least worked out what basic assumptions you need for them to be true.
Not quite. Here's how it works:
We have observed things to fall in the past. If we were to be rigorous, we'd say that on a certain date, at a certain time, in a certain location, we dropped a certain item and watched it fall towards the Earth.
Newton came up with some mathematical equations which describe this phenomenon. Note: Not "explain", but "describe". But of course, they describe more than one object falling in one place at one time -- they describe that any object will fall anywhere, and, more generally, that objects attract.
We call this a "hypothesis" -- it predicts something (things fall, and they accelerate at this rate). Most people consider gravity to be sufficiently tested -- we've worked out (with Newton's equations) that some particular thing should fall, and how fast -- and, similarly, that something we launch into space will orbit (or not), and how fast, and whether it will stay even with the Earth.
When a hypothesis is sufficiently tested, we accept it tentatively as a "theory". That's all gravity is -- a temporary theory until we find something better. And we did -- Einstein's General Relativity actually disproved Newton's Law of Universal Gravitation.
Well, correcting you again, but this is the "hypothesis" of evolution tested.
And it can and does provide very specific predictions. It predicts that certain fossils will be found -- and we then went and found those fossils, and they were pretty much what we expected. It predicts that if you develop a really effective mosquito spray, and any of the mosquitoes are immune, those mosquitoes will survive, thus showing that the mosquito population as a whole has adapted to that spray.
I'm sorry if you were expecting a Borg-like adaptation, where, in the span of a few minutes of screen time, an organism adapts to a particular threat.
Well, no one has witnessed quite a lot of things that we have theories for. You know, crack open a science textbook -- do you think anyone has actually seen an electron? No, we can only infer.
For all you know, there's infinite monkeys in your computer making it go. Or maybe it's the Hand of God Himself. But we consider computers, among other things, to be evidence for -- not proof of, but evidence for -- the existence of electrons.
And it's generally not useful to suggest alternate ideas that have no testable predictions. For instance, if you said my computer operates because of the hand of God, that's a fine idea, but there's absolutely nothing that predicts. It doesn't tell me w
...to Evolution.
The very nature of Science is debate and inquiry. We know what we know because of how well it survives that debate.
Evolution works because there is no other theory that explains what we've observed. Intelligent Design does not count, because it provides no testable predictions.
But when you say things like:
All depends on why you reject it. Everyone I've seen reject evolution has done so for stupid, dogmatic reasons, but you are making stupid, dogmatic statements about science.
I happen to agree with your conclusions, somewhat: Creationism has no place in science. But the reasons you've given for that are not scientific.
Question: Why'd you choose the technology you did?
This guide could also be useful for a manager who has no idea what a good technology is.
Agreed, but the author did want to correct anyone who would hire dumb people, mistaking their stupidity for a lack of social skills.
The point is not that the bleeding edge is needed, but that it teaches you things -- people take the good ideas from Ruby on Rails, for instance, and fold them back into their own favorite environments, even Java.
Also, it's an indicator of passion.
Actually, just me, because, as I said, there is no IT department where I work. Everyone is responsible for their own machine. We can do this because there are five of us, and we're all developers.
Guess again.
That depends on whether I get to set the policy.
Well-run corporations will make their IT department the final word on what goes on a computer and what doesn't. Maybe the CEO can come in and demand that something be done, but users don't have that luxury. If they're on Windows, they are on the corporate domain, their auto-updates are enabled and cannot be disabled, and they will update from the internal company WSUS server, which means IT decides when how, and to whom each update is applied.
Except in Linux, just about any task you can do by hand, you can automate. There are many scripts for deploying configuration to a large number of Linux machines. (Directly -- the lingo is "push", not "pull".) Ruby On Rails seems to like Capistrano, though that's more designed around deploying a Rails app.
But hey, you already control the repository, why not roll your own install disc? I know I could never admin a Windows domain without a custom (nLite'd) install disc, and/or a standard disk image to start from...
Or if you wanted just one place to do it, add that cron job to the base install provided by your repository...
Wait, so on Linux, all I have to do is add the cron job and change the repository. On Windows, I have to join the corporate domain.
Again, I've never done this on Windows, so I don't really know how long it takes to join the corporate domain, on each computer. On Linux, it would take me maybe five minutes per machine, including boot time.
No matter what solution you adopt to deploy stuff, it's going to take some setup.
A heavily-patched Linux. I'm fairly sure one can configure SELinux to disallow this.
Or Linux running in some sort of virtual machine (usermode Linux, XEN, etc) which disallowed access to the real raw disk.
Oh, sorry, didn't see that. You're right.
Those are true in pretty much any case which relies solely on IP addresses to prove guilt.
No, the real problem would be making sure your computer is completely clean (or apparently so) when they pick it up for discovery.
I kind of doubt that, if you're using something which formats automatically. Even vim can autoindent, which reduces them back to useless redundancy.
I'd debate that. I find Python plenty readable.
There is one feature I'd miss: bouncing between matching brackets in vim.
Right. And if you do have something automatically formatting, I just don't see how the extra curlies make a difference (except bouncing in vim).
Personally, it's a moot point for me -- there are too many other big things wrong with Python (the GIL comes to mind) for the whitespace to make much difference.
By the Javascript. (Yes, there is a huge difference. I love Javascript, can't stomach Java.)
Maybe so, but the point is, the players did not have this feature to begin with. It is therefore possible for studios to invent and deliver new and useful features. Some of them are tied to the movie, most of them are gimmicky (like Transformers having a HUD showing the "health" of each robot)...
It's possible to design a Blu-Ray disc that does both of these, although generally, the first part won't be true.
Or wait -- maybe it's not possible yet? I know it's possible to design an HD-DVD disc that way...
Ah, but you also said something about "will not help you obtain Sanctuary any faster". I don't really see how anything other than bandwidth contributes to the speed at which I obtain Sanctuary.
Yes, it does. Sony has released DVDs with copy protection that their own DVD players couldn't deal with.
It's also possible that you could burn your own Blu-Ray discs, but only through their software, which applies some level of DRM to them anyway.
I know how it works for HD-DVD: There are some features which you can only access (local storage, network) if your disc is AACS-encrypted. There were plans to make it possible to only sign a disc, but access the same feaures.
So would I. And when I can view that content on Linux on my desktop, maybe I will. That's one reason Sanctuary has my enthusiastic support.
Well, they've done far too much at this point for me to trust them. It's going to take a lot more to gain that trust back.
Specifically: Standard USB is what the Xbox 360 uses as well, and the Wii uses Bluetooth. Supporting Linux on the PS3 in a hypervisor which denies access to the 3D hardware... what's the point, really? I can think of some academic uses for it, but as a consumer, there's not really a reason to buy a PS3 over a 360, and there's more than one reason to buy a 360 over a PS3. (Not even getting started with the Wii...)
Pretty close. They haven't been using upstart nearly as much as I think they should -- most things are still following the old SysV design.
I know, just pointing out, you can't do it with the same commands. The "init=/bin/sh" line will probably work in an initrd, but the process of getting to the "real" filesystem is a lot harder. (Fortunately, for the most common case, it's possible to simply unmount the root filesystem, sync, and then ctrl+alt+del or hard reboot.)
Ah, I see. Still, does it work well, considering that was not an "exec" command, and therefore, init won't be running as PID 1? I kind of didn't like the idea of leaving the bash process around, even only for a reboot.
Just be thankful that you didn't have to mess with initrds as much as I did. (This was before I moved to ubuntu, where I can generate/update a stock initramfs with one command.)
Yes, that's it. Thanks.
Wait, what? Don't have source code for Firefox?
I'm waiting for an explanation as to how I'm not understanding, because if you just said what I think you said, you're claiming that your techs don't have source code to the most widely-known open source program?
Because unless your name is "Internet Explorer", people won't know, just from the name, what kind of program it is. I think most of us have generally accepted names like Firefox, but unless someone already knows what it is, how would you know Firefox is a web browser? Should we just be naming the first project "Web Browser", and subsequent projects like Konqueror, iCab, etc can fade into obscurity?
I suspect GwenView is a response to a Windows image viewer, IrfanView.
iThink you may be just a bit iPredisposed to think of iLinux apps as having weird iNames.
Honestly, why is iCrap better than Krap?
What program isn't an application?
Or at least, it's generally possible to know it's an Application from context; by knowing what it does. It's generally not possible to know, without the naming scheme, which desktop environment (or graphics library) a given application uses.
It gdoes make it a bit geasier to Kfigure out gwhich iApps go to Kwhich iDesktop Kenvironment.
My guess is that the whole reason for releasing now is the desperate hope that they'll have something decent ready for Kubuntu.
And I imagine that in another few months, it's going to be better than KDE3. By that I mean functionality; KDE3 just hasn't been particularly stable for me. (The ACID3 test crashed Konqueror, and Konqueror quite frequently crashes while I'm typing comments on Slashdot...)
Ripping off Sudo was a good start, but they really need to learn some lessons from Linux package managers.
OS X has the same problem, by the way. Linux distros are really the only place you see a system-wide package manager.