Most big components should be a part of any distribution, and those should be enough for 90% of applications.
Little stuff can just be part of the bundle. I don't mind it if a bundle is 5 megs larger than it would otherwise need to be if it makes my life easier.
Also, there are no more DMG exploits. There is nothing wrong with having a few XML files around that belong to an application you no longer have, if it it really irks you, or if programs leave behind large caches, there are plenty of pieces of software that will delete preferences and caches that belong to software you no longer have.
Most applications shouldn't need to modify the OS to run, and for that minority that do, OS X still does have packages. This is how haxies and so forth work.
The only valid objection I've seen to bundles is the one about how a user shouldn't be able to install random software from the internet. This is a pretty good point, but I fail to see how that, even in a system that uses an apt repository, you would be able to prevent a user from downloading and installing some random RPM from a website. You would have to have a severely crippled OS.
Hard drive space is cheap. And so is bandwidth. People on 56k connections can install software from CDs. Don't make things broken for everyone to help them.
Do appfolders (bundles), and if you want the functionality of a shared library, include it in the bundle. Unless it's something like 100 megs. Which it won't be.
If dozens of programs each end up including the same shared code in their appfolders, who cares? Again, hard drive space is cheap, and that's the price to pay for easier system management, and knowing that applications won't suddenly stop working when one of their dependencies has been changed.
There is no earthly reason why a GUI application should scatter files hither and yon across a hard drive, and why installing a program should require some package or installer or whatever. I cannot believe the hassle that I have to go through to install software on my Linux box as opposed to my Mac.
An OS X application consists of one file--- really a bundle. It is a directory that acts like a single executable file. Everything it needs to run that is not part of the basic OS X setup is in that file.
You don't even need to install the application. You can just run it from its compressed disk image that is still sitting in your downloads directory, if you like. Or you can copy it to your hard drive wherever you like. When you tire of it, you delete it.
Now, "Linux" is not capable of doing this because no one runs just Linux. But there is no reason why, say, Gnome apps can't be distributed this way. If there are technical issues in the way, they need to be resolved. Because the OS X way is better that the Linux and the Windows methods, and ought to be copied.
(ps: I do know that Unix programs are often installed via packages in OS X, as well as software that for whatever reason needs to modify the OS. But these are very rare and approached warily by seasoned OS X users.)
You obviously have never used a Mac. Terminal.app is right there for you to run. Wow! I'm running it right now! Bash is not Linux, and Linux didn't invent terminals. The free iTerm is much better than any terminal available for Linux, by the way.
And yes, focus follows clicks in OS X, and it does so in a consistent way. Weird, I thought that Linux nerds were always complaining about the lack of focus follows mouse. (Which you can have in X programs. I can run Fluxbox on one monitor and regular Aqua on another.)
OS X has two major UI advantages over any other GUI: proper window management, where programs own windows instead of windows owning programs. No "main windows," no wondering if you have two windows in one program or two instances of one program, etc. And second, file management. Applications consisting of only one file. To "install," copy that file to \Applications. To "uninstall," delete that file.
There is no reason for a program to spew detritus (other than plain text\XML config files that are user-specific) hither and yon all over your harddrive. Only OS X behaves properly in this regard. Linux is the worst.
"The hugeness of the present day online population came primarily from the great opening up of the Web"
Its privitization, you mean. Government founded, privately popularized. Maybe something so open couldn't have been founded by private companies. But it sure isn't goverment that made the internet what it is today.
-j
He raises good points, especially ESR's contradictory stance on Microsoft. But he says that philosophical arguments are impractical, and therefore irrelevant. While I agree that "no regulation" is not an option in the current state of affairs, it is still an important point to be discussed. Philosophical points may never be fully implemented, but they are hugely influential nonetheless. A focus on underlying philosophy is not childish.
Frequently I hear that the internet is the product of regulation. It is true that the internet is not the product of a free market. But- it did not become "huge" until it was somewhat privatized. AOL, Compuserve, et al. demonstrate that private, large scale computer networks are practical. If it weren't for the internet, I'n sure we'd all be on some such network. The demands of consumers such as Slashdot readers would make sure it wasn't like AOL. Given the points on both sides, I would propose that the existence and success of the internet be used neither to defend nor attack regulation. It's absurd that both sides are basing their general arguments on one special case.
-j
s.d.i. is possible, and it always has been. the 'scientific' objections to it were politically motivated.
current plans for defense systems owe quite a bit to s.d.i. research. difficult does not equal impossible.
by the way- what the chair force needs to do is stop wasting money on hyperexpensive fighters of limited use. that money would be better spent elsewhere in the military. I am thinking specifically of the crusader artillery system that the army is going to get Real Soon Now.
Tactical Mobile Hinged Mechanical Interior/Exterior Personnel Transfer Apparatus a.k.a., a door on a truck. I came up with this one on drill this weekend with the Guard.
"For instance exploiting child labor in a third world country and paying them dirt" I suppose these children who are so poor that they have to take jobs to support their families, would be much better off if they all got fired?
America survived on child labor for years. We needed to. It was only outlawed when it was no longer economically necessary. Sadly, for these nations, it still is.
In third world countries, the trend is for the parents to have as many kids as they can, so there are more to work the fields and factories. I suppose the humane thing would be to cut them off until they change their ways, thus shoving them deeper into poverty?
Or how about we just let them develop naturally as we did, until there's no such thing as a "third world"?
> I have no problem with Third World labor, as > long as the workers are given decent wages and > safe conditions. I would agree with/humane/ conditions, but wages should be determined by the market.
If it weren't for the lower wages, no one would build factories in underdeveloped countries. The real choice for them isn't low wages or high wages, it's low wages or no wages.
Let the workers decide what they're willing to do. Rich, bored Americans need to learn to mind their own business, and be thankful that someone overseas isn't trying to put them out of a job.
this does not preclude charity, as Rand's definition of "value" was not exclusively monetary. romantic relationships, for example, involve exchanged values.
Rand was a second-rate romance novelist
romantic (in the true sense) does not equal "romance". I won't bother to defend the merits of her novels, except to mention that it's through the government system of education that I learned of Rand. quite ironic. philosophers ignore her, but literary types do not. that's quite an argument from authority, and unneeded. her novels stand up for themselves. even if you have problems with her philosophy, she has said more for artistic integrity than any other writer I know.
you're clearly trolling by comparing Objectivism to socialism. Leonard Peikoff, heir of Ayn Rand, is leader of the Objectivist cult. I would like for him and his goombahs to leave Objectivism, which is a philosophical school, and not a fixed canon (usually misinterpreted). there are exiled, saner groups such as David Kelley's , but they do not control the all important "For more information" cards which fall out of the novels. Ayn Rand made many foolish decisions in her life, and thinking Peikoff somehow worthy was her biggest.
her philosophy was not reductionist, but unfinished. she never wrote a systematic exposition of it. what she did write was not always on target. but her ideas, being the kernel of the libertarian movement (and much more), were hardly irrelevant.
forced altruism is immoral, and it is incorrect to have altruism as the basis of your morality.
however, this does not make "altruism"- in the sense of helping others- evil. the frequent denunciations of altruism were directed at ethics systems, not acts of human kindness.
One the one hand, there are zealots who do not understand Ayn Rand's philosophy, but think they are for it. On the other hand, there are people who do not understand Ayn Rand's philosophy, but think they are against it.
This situation is understandable, given the incompetent fools (Peikoff and his cronies) now the official rulers of the Objectivist kingdom, and given the foolishness of Rand's own later years.
Either way. "The fact that a man has no claim on others....does not preclude or prohibit good will among men and does not make it immoral to offer or to accept voluntary, non-sacrificial assistance." -Ayn Rand, in "The Objectivist", June 1966.
Charity is ethically neutral. If you value literature, would like to see it preserved, and are a benevolent person, it is in your own rational self interest to help Project Gutenberg, if you can and want to. There is no room to argue the point. Even rational, objective people can order their values differently- reason many be objective, but temperament is as subjective as ever.
Please, all you psuedo-Randites out there, stop your ignorant posting. You damage a great philosopher's reputation. If you would like to belong to a cult, try Catholicism, and leave Objectivism to what few serious thinkers haven't yet abandonded it. And thank you, for reminding me why I usually avoid reading message boards.
Slashdot Mirror
Most big components should be a part of any distribution, and those should be enough for 90% of applications.
Little stuff can just be part of the bundle. I don't mind it if a bundle is 5 megs larger than it would otherwise need to be if it makes my life easier.
Also, there are no more DMG exploits. There is nothing wrong with having a few XML files around that belong to an application you no longer have, if it it really irks you, or if programs leave behind large caches, there are plenty of pieces of software that will delete preferences and caches that belong to software you no longer have.
Most applications shouldn't need to modify the OS to run, and for that minority that do, OS X still does have packages. This is how haxies and so forth work.
The only valid objection I've seen to bundles is the one about how a user shouldn't be able to install random software from the internet. This is a pretty good point, but I fail to see how that, even in a system that uses an apt repository, you would be able to prevent a user from downloading and installing some random RPM from a website. You would have to have a severely crippled OS.
Hard drive space is cheap. And so is bandwidth. People on 56k connections can install software from CDs. Don't make things broken for everyone to help them.
Do appfolders (bundles), and if you want the functionality of a shared library, include it in the bundle. Unless it's something like 100 megs. Which it won't be.
If dozens of programs each end up including the same shared code in their appfolders, who cares? Again, hard drive space is cheap, and that's the price to pay for easier system management, and knowing that applications won't suddenly stop working when one of their dependencies has been changed.
Bundles are better.
For distributing user applications, at least.
There is no earthly reason why a GUI application should scatter files hither and yon across a hard drive, and why installing a program should require some package or installer or whatever.
I cannot believe the hassle that I have to go through to install software on my Linux box as opposed to my Mac.
An OS X application consists of one file--- really a bundle. It is a directory that acts like a single executable file. Everything it needs to run that is not part of the basic OS X setup is in that file.
You don't even need to install the application. You can just run it from its compressed disk image that is still sitting in your downloads directory, if you like. Or you can copy it to your hard drive wherever you like. When you tire of it, you delete it.
Now, "Linux" is not capable of doing this because no one runs just Linux. But there is no reason why, say, Gnome apps can't be distributed this way. If there are technical issues in the way, they need to be resolved. Because the OS X way is better that the Linux and the Windows methods, and ought to be copied.
(ps: I do know that Unix programs are often installed via packages in OS X, as well as software that for whatever reason needs to modify the OS. But these are very rare and approached warily by seasoned OS X users.)
You obviously have never used a Mac. Terminal.app is right there for you to run. Wow! I'm running it right now! Bash is not Linux, and Linux didn't invent terminals. The free iTerm is much better than any terminal available for Linux, by the way.
And yes, focus follows clicks in OS X, and it does so in a consistent way. Weird, I thought that Linux nerds were always complaining about the lack of focus follows mouse. (Which you can have in X programs. I can run Fluxbox on one monitor and regular Aqua on another.)
OS X has two major UI advantages over any other GUI: proper window management, where programs own windows instead of windows owning programs. No "main windows," no wondering if you have two windows in one program or two instances of one program, etc. And second, file management. Applications consisting of only one file. To "install," copy that file to \Applications. To "uninstall," delete that file.
There is no reason for a program to spew detritus (other than plain text\XML config files that are user-specific) hither and yon all over your harddrive. Only OS X behaves properly in this regard. Linux is the worst.
"The hugeness of the present day online population came primarily from the great opening up of the Web" Its privitization, you mean. Government founded, privately popularized. Maybe something so open couldn't have been founded by private companies. But it sure isn't goverment that made the internet what it is today. -j
He raises good points, especially ESR's contradictory stance on Microsoft. But he says that philosophical arguments are impractical, and therefore irrelevant. While I agree that "no regulation" is not an option in the current state of affairs, it is still an important point to be discussed. Philosophical points may never be fully implemented, but they are hugely influential nonetheless. A focus on underlying philosophy is not childish.
Frequently I hear that the internet is the product of regulation. It is true that the internet is not the product of a free market. But- it did not become "huge" until it was somewhat privatized. AOL, Compuserve, et al. demonstrate that private, large scale computer networks are practical. If it weren't for the internet, I'n sure we'd all be on some such network. The demands of consumers such as Slashdot readers would make sure it wasn't like AOL. Given the points on both sides, I would propose that the existence and success of the internet be used neither to defend nor attack regulation. It's absurd that both sides are basing their general arguments on one special case. -j
s.d.i. is possible, and it always has been. the 'scientific' objections to it were politically motivated.
current plans for defense systems owe quite a bit to s.d.i. research. difficult does not equal impossible.
by the way- what the chair force needs to do is stop wasting money on hyperexpensive fighters of limited use. that money would be better spent elsewhere in the military. I am thinking specifically of the crusader artillery system that the army is going to get Real Soon Now.
Tactical Mobile Hinged Mechanical Interior/Exterior Personnel Transfer Apparatus a.k.a., a door on a truck. I came up with this one on drill this weekend with the Guard.
"For instance exploiting child labor in a third world country and paying them dirt"
I suppose these children who are so poor that they have to take jobs to support their families, would be much better off if they all got fired?
America survived on child labor for years. We needed to. It was only outlawed when it was no longer economically necessary. Sadly, for these nations, it still is.
In third world countries, the trend is for the parents to have as many kids as they can, so there are more to work the fields and factories. I suppose the humane thing would be to cut them off until they change their ways, thus shoving them deeper into poverty?
Or how about we just let them develop naturally as we did, until there's no such thing as a "third world"?
> I have no problem with Third World labor, as > long as the workers are given decent wages and > safe conditions. I would agree with /humane/ conditions, but wages should be determined by the market.
If it weren't for the lower wages, no one would build factories in underdeveloped countries. The real choice for them isn't low wages or high wages, it's low wages or no wages.
Let the workers decide what they're willing to do. Rich, bored Americans need to learn to mind their own business, and be thankful that someone overseas isn't trying to put them out of a job.
wow. suburb bashing. that's fresh.
it was a well made movie. but i can't deem a work with tired old, false platitudes as its theme oscar worthy.
the edge city is dead, long live exurbia.
"value must always be given for value received."
this does not preclude charity, as Rand's definition of "value" was not exclusively monetary. romantic relationships, for example, involve exchanged values.
Rand was a second-rate romance novelist
romantic (in the true sense) does not equal "romance". I won't bother to defend the merits of her novels, except to mention that it's through the government system of education that I learned of Rand. quite ironic. philosophers ignore her, but literary types do not. that's quite an argument from authority, and unneeded. her novels stand up for themselves. even if you have problems with her philosophy, she has said more for artistic integrity than any other writer I know.
you're clearly trolling by comparing Objectivism to socialism. Leonard Peikoff, heir of Ayn Rand, is leader of the Objectivist cult. I would like for him and his goombahs to leave Objectivism, which is a philosophical school, and not a fixed canon (usually misinterpreted). there are exiled, saner groups such as David Kelley's , but they do not control the all important "For more information" cards which fall out of the novels. Ayn Rand made many foolish decisions in her life, and thinking Peikoff somehow worthy was her biggest.
her philosophy was not reductionist, but unfinished. she never wrote a systematic exposition of it. what she did write was not always on target. but her ideas, being the kernel of the libertarian movement (and much more), were hardly irrelevant.
forced altruism is immoral, and it is incorrect to have altruism as the basis of your morality.
however, this does not make "altruism"- in the sense of helping others- evil. the frequent denunciations of altruism were directed at ethics systems, not acts of human kindness.
One the one hand, there are zealots who do not understand Ayn Rand's philosophy, but think they are for it. On the other hand, there are people who do not understand Ayn Rand's philosophy, but think they are against it.
This situation is understandable, given the incompetent fools (Peikoff and his cronies) now the official rulers of the Objectivist kingdom, and given the foolishness of Rand's own later years.
Either way. "The fact that a man has no claim on others....does not preclude or prohibit good will among men and does not make it immoral to offer or to accept voluntary, non-sacrificial assistance." -Ayn Rand, in "The Objectivist", June 1966.
Charity is ethically neutral. If you value literature, would like to see it preserved, and are a benevolent person, it is in your own rational self interest to help Project Gutenberg, if you can and want to. There is no room to argue the point. Even rational, objective people can order their values differently- reason many be objective, but temperament is as subjective as ever.
Please, all you psuedo-Randites out there, stop your ignorant posting. You damage a great philosopher's reputation. If you would like to belong to a cult, try Catholicism, and leave Objectivism to what few serious thinkers haven't yet abandonded it. And thank you, for reminding me why I usually avoid reading message boards.