He's saying we have a right to use the Earth, but we don't have a right to damage it.
That's an interesting religious position, but it makes me wonder how his religious views have affected his research. (Not that that's always a bad thing.)
Making the data available on the internet is simple enough that it's reasonable to expect them to do it. And if they're refusing to do that, it's reasonable to accuse them of actively trying to prevent sharing and review of their data.
Those things are not necessarily reasonable. Data can easily amount to hundreds of gigabytes. I know someone getting data from NASA for weather-related research who, instead of downloading the data via FTP, gets a DVD in the mail each week, because the people generating the data don't have the budget for a fast enough internet connection or a big enough FTP server. Putting it on a throttled FTP server wouldn't necessarily be useful. (Bittorrent, on the other hand, could work, if lots of people wanted to download that 1 TB of raw weather observations.)
The code and tools to process it *do* exist, but they are archaic or take some effort to recode appropriately and validate.
That's my biggest concern. Few scientists are software engineers, and I fear that computer programs used in scientific research have bugs that affect their results.
Things which we believe to be true, and for which we have evidence: (scientific facts)
Things which we believe to be true, but for which we have no evidence: (superstitions)
Things which we believe to be untrue, and for which we have (contrary) evidence: (irrationalities)
Things which we believe to be untrue, but for which we have no evidence: (superstitions)
I mostly agree, but note that one's theology greatly affects one's opinion about what constitutes "evidence", and therefore what one believes is "true" and what is "superstition". Most of what we believe, we believe because somebody or something we trust (parent, teacher, television etc.) said so, so nobody actually believes anything without evidence of some sort (though sometimes that evidence is just hearsay). Most of what we believe, we do so without proof, and worse, what we think is proof often turns out to not be so.
In reality, we have conflicting evidence about a great many things: some people say one thing is true, while other people say that it is superstition. Not all evidence is scientific evidence, scientific evidence sometimes conflicts too, and science maddeningly fails to answer lots of important questions (it does best when a question can be repeated under controlled circumstances, and most circumstances in life are uncontrolled). In the end, a better taxonomy would be this:
Things which we believe to be true, based on our personal experiences
Things which we believe to be false, based on our personal experiences
"Personal experiences" include "somebody told me this was true" as well as "I have personally performed this scientific experiment." Unfortunately that's not quite as interesting as your categorization.
There is no such thing as an "objective, external reality". All things viewed and/or reported by a human being are subjective.
Human opinions are subjective, but the things that stimulate our senses are what they are: objective. A sound wave is a sound wave, even when people disagree about what it is or what it means, and things like atoms and photons really do exist. The objective, external reality impacts me routinely, despite my attempts to ignore it. It is our "through a glass darkly" views of it that are subjective; debates over the accuracy of our opinions about that external reality have employed philosophers and theologians for many years.
It is possible that we are (or just I am) living in a simulation, but as far as anyone can tell, that doesn't matter: attempts to ignore the "simulation" behave the same as attempts to ignore reality, and have the same results, such as loss of job.
Sorry, but no self-respecting expert would ever describe themselves as, or want to be labeled "Hacker"
Hogwash. Self-respecting experts have the confidence to describe themselves however they'd like, and "hacker" has a specific non-negative meaning in some subcultures, including the one that Paul Graham is talking about.
I'm saying we, humans, have lost as soon as we have to fight.
I'd go a step farther than that: we, humans, have lost even before we begin fighting. We'll never live up to our (varying) ideals.
If we could have solved WWII before it started, getting the German people to realise Hitler was dangerously full of shit, all those millions of lives would have been saved. I fail to see how you can see that as being worse than them dying.
I agree, that would have been wonderful. Unfortunately humans will never be able to solve all of our disagreements peacefully. We're too stubborn, too shortsighted and too easily fooled by demagogues and charlatans. And for that matter, too willing to go along with demagogues and charlatans eyes wide open. Germany could have won WW2, and lots of Germans back then would have been OK with that.
Of course if you know anything about our choice of "allies" over the past 30 years or so, this means we're one hot headed revolt away from facing a militia where the locals have drone guns too.
Don't worry, they won't sell them enough spare parts to stay in operation for long. It's the American Way: get them hooked on the equipment and charge through the nose for parts. Maintenance on military equipment is tough; tanks and planes consume lots of parts.
Any time we fight, we've already lost, no matter the outcome of the war.
Any time? Who is "we"? Lots of places have fought and won defensive wars. Giving in and surrendering to people who hate you and are willing and able to kill you because of your race, creed, color or sex is not necessarily the best option.
Re:You don't need MS Office to create .doc files
on
Does ODF Have a Future?
·
· Score: 2, Insightful
This becomes a MAJOR problem in an environment where templates have not been created and/or maintained properly and efficiently. Often employees will take an existing document, ctrl-A, DEL, File-> Save As..., then start typing to create a "new" document simply to get the "corporate headers".
In an enterprise environment, this is a non-problem. The standard image will have fast save disabled, markup viewing turned on, and the print/save/send warning turned on, and these settings will be reestablished every time a user logs on.
Do you see the disconnect? The post you're responding to says that in at least some environments (I suspect it is the majority), things are done in an ad hoc way.
Yes, if they have competent sysadmins with management support their systems will be locked down and they won't be able to stick the corporate foot in the corporate mouth. How often do you suppose that happens?
So, is unknowing possession a crime in this case? Let the poor analogies begin...
It's like if you're driving your car, and you don't know if you're on a Bill Gates Windows-only road or not. Nobody asks you to pay the toll, but when you drive your Linux car that was free and gets 200 MPG you get pulled over!
I can't speak for that package system since I don't know it, but usually that's the fault of the packages not the system.
It depends where one draws the boundary between "packages" and "system". In the case of pkg-get, it isn't used for the core operating system, so there is a lot of installed software outside of its worldview. Of course it's partly the fault of gvim for requiring something different from what's installed, but if Sun were the maintainer for the entire repository, then packages would hopefully not get out of sync like that.
It would be great to see Solaris become tightly integrated with something like apt. pkg-get is ok, but it isn't currently used for all packages, and a Sun-backed and -improved version would be better. For example, I'd like to see it manage security updates in a way that meets the needs of Solaris sysadmins, with separate actions for downloading, applying and rolling back. I'd also like to see my attempts to install gvim not download 50 megabytes worth of libraries that are already on my system, in a slightly different version number.
Who wants AIX support? Seriously, what AIX systems even have an ATI 3d chip? Or did you mean AIGLX?
While the parent poster did mean AIGLX, this does display an advantage of an open source driver: not every computer is an x86, and there do exist small communities with the resources to port an open-source driver to their platform.
If the net effect of the current system is negative...
You're assuming that the current system has no positive benefits at all...
How do you get "has no positive benefits at all" out of "if the net effect is negative"? There are probably positive benefits to every bad thing.
If you scrap the whole thing you're left with only the negative effects and none of the positive ones which people are currently enjoying.
On the contrary: if you scrap the system then the primary negative effect, businesses using the system to stifle legitimate competition, goes away entirely.
Granted, but the question remains: complex to whom?
Syntax complexity can be independently measured. Someone could memorize every syntactical element of the most complex language ever devised, and it would still be the most complex language ever devised. Ideally programmers will focus on design rather than syntax.
Progress in programming language (if it has been defined by a single metric at all), has been defined by increases in expressiveness, not simplicity.
The most expressive language in the world is not very useful if its syntax is confusing, complicated and error prone. Some examples of those things:
Confusing: using the same or very similar syntax for very different operations, such as defining = to mean different things in different contexts.
Complicated: having lots more than one way to do something; having a relatively large number of different kinds of syntactical elements; having verbose syntactical sugar.
Error prone: see confusing; also, having different operations be so similar syntactically that adding or omitting a single character produces a valid operation (such as = and == in C, or the original FORTRAN loop syntax)
Syntax simplicity is certainly not the only way to measure a language (you noted expressiveness; there are others also), but it is extremely important.
So your point is that the current patent system provides insufficient protection to inventors and that the solution is to scrap it altogether and provide no protection at all ?
If so, that's not a bad point to make. If the net effect of the current system is negative, then scrapping the current system entirely will result in a net improvement.
I can't think of any product which has Y and would really benefit from X but doesn't have it.
Linux distributions would benefit from legal playback of data stored in various proprietary codecs.
Pretty soon people wouldn't share their inventions any more if they could actually keep the workings secret and if they couldn't they have trouble making any money from them so in the end no one would really bother.
I don't believe that. Most inventions made to solve a problem are obvious after the fact ("why didn't I think of that?"), and in most cases patenting an invention does not improve income. People don't learn new things by trolling patent archives; instead, developers avoid patent archives in order to minimize liability.
Religion notwithstanding, it will become microkernel-like, because a monolithic kernel just doesn't scale. It's as simple as that.
By that definition, Linux is already microkernel-like, with clearly defined subsystems that interact using "message passing" (function calls). Being "monolithic" does not require spaghetti code.
Monolithic kernels don't scale as the number of CPUs rises, nor as the number of drivers and other logical components goes up.
That depends greatly on which definition of "monolithic" one uses. Arguably, a kernel can be called monolithic if it uses a single address space for its subsystems (as opposed to putting subsystems into what Linux calls "user space" and doing context switches when communicating between them), and nothing stops such a kernel from scaling well.
Arguably, a defining feature of microkernels is putting subsystems in user space, which kills performance.
It's quite easy to avoid casual cursing in public, but given time, eventually someone runs over your foot with a bicycle or you nearly get hit in a crosswalk, and before you even have time to realise you're not really hurt, you've cursed in public.
That's only true if one is in the habit of cursing in general. If the first exclamation that comes to mind is not a prohibited word (for example, "ouch") then accidental cursing is unlikely.
There are local laws against cursing in public, here and there, often dating back to the 1800s. Enforced? Rarely, if ever. You'd have to arrest everyone, sooner or later.
Actually, it's pretty easy to avoid cursing in public, and for those who choose not to avoid it, every prohibited word or phrase has non-censored synonyms.
So it does sound like a neat trick, but what is a valid, viable use case for it?
When my PC converges with my PDA, I want to be able to walk up to any display and use it without hauling a cable around. Better yet: I want HD video in my glasses, connected wirelessly to my PDA PC.
I have an external hard drive for my work laptop. It would be nice to be able to connect to it wirelessly. I'd also like to be able to sync my laptop to a docking station wirelessly.
There are all kinds of nifty things you can do when your network is faster than your hard disk, even if that network is only a few meters across, like keep all your hard disks in the next room or build computers out of all-"external" hardware.
What's really scary is that currently, only a few 3rd party devices (some newer HD Televisions and the TiVos) use CableCards, but the tech said Comcast is planning on rolling out their own hardware that uses them. This means that instead of 1 CableCard install per month, all installs will be CableCard based.
That sounds like a good way to quickly encounter and iron out the bugs in the system. Those who sign up a couple months later might have no trouble at all.
Slashdot Mirror
That's an interesting religious position, but it makes me wonder how his religious views have affected his research. (Not that that's always a bad thing.)
Everyone consumes; not having kids allows someone to produce more.
Those things are not necessarily reasonable. Data can easily amount to hundreds of gigabytes. I know someone getting data from NASA for weather-related research who, instead of downloading the data via FTP, gets a DVD in the mail each week, because the people generating the data don't have the budget for a fast enough internet connection or a big enough FTP server. Putting it on a throttled FTP server wouldn't necessarily be useful. (Bittorrent, on the other hand, could work, if lots of people wanted to download that 1 TB of raw weather observations.)
That's my biggest concern. Few scientists are software engineers, and I fear that computer programs used in scientific research have bugs that affect their results.
I mostly agree, but note that one's theology greatly affects one's opinion about what constitutes "evidence", and therefore what one believes is "true" and what is "superstition". Most of what we believe, we believe because somebody or something we trust (parent, teacher, television etc.) said so, so nobody actually believes anything without evidence of some sort (though sometimes that evidence is just hearsay). Most of what we believe, we do so without proof, and worse, what we think is proof often turns out to not be so.
In reality, we have conflicting evidence about a great many things: some people say one thing is true, while other people say that it is superstition. Not all evidence is scientific evidence, scientific evidence sometimes conflicts too, and science maddeningly fails to answer lots of important questions (it does best when a question can be repeated under controlled circumstances, and most circumstances in life are uncontrolled). In the end, a better taxonomy would be this:
"Personal experiences" include "somebody told me this was true" as well as "I have personally performed this scientific experiment." Unfortunately that's not quite as interesting as your categorization.
Human opinions are subjective, but the things that stimulate our senses are what they are: objective. A sound wave is a sound wave, even when people disagree about what it is or what it means, and things like atoms and photons really do exist. The objective, external reality impacts me routinely, despite my attempts to ignore it. It is our "through a glass darkly" views of it that are subjective; debates over the accuracy of our opinions about that external reality have employed philosophers and theologians for many years.
It is possible that we are (or just I am) living in a simulation, but as far as anyone can tell, that doesn't matter: attempts to ignore the "simulation" behave the same as attempts to ignore reality, and have the same results, such as loss of job.
Hogwash. Self-respecting experts have the confidence to describe themselves however they'd like, and "hacker" has a specific non-negative meaning in some subcultures, including the one that Paul Graham is talking about.
I'd go a step farther than that: we, humans, have lost even before we begin fighting. We'll never live up to our (varying) ideals.
I agree, that would have been wonderful. Unfortunately humans will never be able to solve all of our disagreements peacefully. We're too stubborn, too shortsighted and too easily fooled by demagogues and charlatans. And for that matter, too willing to go along with demagogues and charlatans eyes wide open. Germany could have won WW2, and lots of Germans back then would have been OK with that.
Don't worry, they won't sell them enough spare parts to stay in operation for long. It's the American Way: get them hooked on the equipment and charge through the nose for parts. Maintenance on military equipment is tough; tanks and planes consume lots of parts.
Any time? Who is "we"? Lots of places have fought and won defensive wars. Giving in and surrendering to people who hate you and are willing and able to kill you because of your race, creed, color or sex is not necessarily the best option.
Do you see the disconnect? The post you're responding to says that in at least some environments (I suspect it is the majority), things are done in an ad hoc way.
Yes, if they have competent sysadmins with management support their systems will be locked down and they won't be able to stick the corporate foot in the corporate mouth. How often do you suppose that happens?
It's like if you're driving your car, and you don't know if you're on a Bill Gates Windows-only road or not. Nobody asks you to pay the toll, but when you drive your Linux car that was free and gets 200 MPG you get pulled over!
The latest Civ 4 expansion just came out this month.
It depends where one draws the boundary between "packages" and "system". In the case of pkg-get, it isn't used for the core operating system, so there is a lot of installed software outside of its worldview. Of course it's partly the fault of gvim for requiring something different from what's installed, but if Sun were the maintainer for the entire repository, then packages would hopefully not get out of sync like that.
It would be great to see Solaris become tightly integrated with something like apt. pkg-get is ok, but it isn't currently used for all packages, and a Sun-backed and -improved version would be better. For example, I'd like to see it manage security updates in a way that meets the needs of Solaris sysadmins, with separate actions for downloading, applying and rolling back. I'd also like to see my attempts to install gvim not download 50 megabytes worth of libraries that are already on my system, in a slightly different version number.
While the parent poster did mean AIGLX, this does display an advantage of an open source driver: not every computer is an x86, and there do exist small communities with the resources to port an open-source driver to their platform.
How do you get "has no positive benefits at all" out of "if the net effect is negative"? There are probably positive benefits to every bad thing.
On the contrary: if you scrap the system then the primary negative effect, businesses using the system to stifle legitimate competition, goes away entirely.
Sorry for the late reply.
Syntax complexity can be independently measured. Someone could memorize every syntactical element of the most complex language ever devised, and it would still be the most complex language ever devised. Ideally programmers will focus on design rather than syntax.
The most expressive language in the world is not very useful if its syntax is confusing, complicated and error prone. Some examples of those things:
Confusing: using the same or very similar syntax for very different operations, such as defining = to mean different things in different contexts.
Complicated: having lots more than one way to do something; having a relatively large number of different kinds of syntactical elements; having verbose syntactical sugar.
Error prone: see confusing; also, having different operations be so similar syntactically that adding or omitting a single character produces a valid operation (such as = and == in C, or the original FORTRAN loop syntax)
Syntax simplicity is certainly not the only way to measure a language (you noted expressiveness; there are others also), but it is extremely important.
If so, that's not a bad point to make. If the net effect of the current system is negative, then scrapping the current system entirely will result in a net improvement.
Linux distributions would benefit from legal playback of data stored in various proprietary codecs.
I don't believe that. Most inventions made to solve a problem are obvious after the fact ("why didn't I think of that?"), and in most cases patenting an invention does not improve income. People don't learn new things by trolling patent archives; instead, developers avoid patent archives in order to minimize liability.
By that definition, Linux is already microkernel-like, with clearly defined subsystems that interact using "message passing" (function calls). Being "monolithic" does not require spaghetti code.
That depends greatly on which definition of "monolithic" one uses. Arguably, a kernel can be called monolithic if it uses a single address space for its subsystems (as opposed to putting subsystems into what Linux calls "user space" and doing context switches when communicating between them), and nothing stops such a kernel from scaling well.
Arguably, a defining feature of microkernels is putting subsystems in user space, which kills performance.
That's only true if one is in the habit of cursing in general. If the first exclamation that comes to mind is not a prohibited word (for example, "ouch") then accidental cursing is unlikely.
Actually, it's pretty easy to avoid cursing in public, and for those who choose not to avoid it, every prohibited word or phrase has non-censored synonyms.
When my PC converges with my PDA, I want to be able to walk up to any display and use it without hauling a cable around. Better yet: I want HD video in my glasses, connected wirelessly to my PDA PC.
I have an external hard drive for my work laptop. It would be nice to be able to connect to it wirelessly. I'd also like to be able to sync my laptop to a docking station wirelessly.
There are all kinds of nifty things you can do when your network is faster than your hard disk, even if that network is only a few meters across, like keep all your hard disks in the next room or build computers out of all-"external" hardware.
That sounds like a good way to quickly encounter and iron out the bugs in the system. Those who sign up a couple months later might have no trouble at all.