That's because if done incorrectly a chimney is dangerous. With an open door (or no door) it behaves like a blast furnace. (not a great idea in a wooden house). A straw roof by contrast allows most of the smoke to pass through without danger.
Because I can write an OS or application myself (and distribute worldwide), but not build a car myself (and even if I could, I couldn't distribute it worldwide)? Patents on cars restrict the freedom of a few corporations and maybe a few real people, but not very many. Patents on software directly restricts the freedom of millions of real people. Not that I am necessartily advocating patents on cars, but anyway...
It already runs (slowly) on average PC hardware with PearPC. Why won't it run more quickly on average PC hardware with PearPC when the Mac CPU is also x86?
Jimi Hendrix, Janis Joplin, Elvis Presley. If only extended copyrights had been available back in the sixties and seventies, these talented artistes would have been saved from the despair of pondering a penniless future, and not been driven to waste their lives.
Such negativity I have never seen. Judging by all the responses to this article (the ones not advocating fraud that is) complaining about this or that failure with assorted Palms, can someone explain why would anybody even want a new Palm device? (or is it just a case of "hey, free stuff"?)
The reason we haven't gone to space is because there is nowhere to go, and nothing for a person to do that isn't a waste of time and resources. Machines will be doing all the interesting work in space for the forseeable future.
It is C's fault that programs make core dumps (or equivalent). It's not like core dumps, or rather the "undefined behaviour" that causes them when you do something nasty to a pointer, are not an intrinsic feature of the language. They are a flaw in the language that require the developer to do a lot of work to make sure the code doesn't trigger any undefined behaviour.
C, as a language exposes the developer to a whole class of errors that another language does not, and so it's perfectly reasonable to say that C is a flawed language in this respect. I know there are reasons (good or otherwise) why C has chosen to behave like this, but it's still a language flaw rather than an implementation flaw, and one that leads to the majority of the security vulnerabilities in your favourite OS (whichever it happens to be), and most of the random crashes and resource leakeage in your favourite applications.
Languages can be poorly designed just as much as implementations can be.
The trouble is, people (companies, politicians, open source visionaries, journalists, etc...) want open source to be something it isn't. Like it or not, as a whole, it's not a charity, it's not working for the third world, it's not business friendly, it's not anti-business, it's not trying to make the world a better place, it's not trying to provide a unified desktop, it's not trying to undermine capitalism, it's not trying to stop microsoft. Some particular individual projects might be any of those things, but for the most part, it's people writing something that they find useful or interesting for themselves, and don't mind sharing with anybody who is bothered enough to use it. As a whole it is directionless, except in the sense of general expansion of the number of both useful and useless projects out there.
But what's a successful open source project? Many of the developers (unpaid type, which is the vast majority) would consider it successful if the project does whatever they wanted it to do when they started. No need for all those other folks at all. What you are describing is a popular project. If popularity wasn't one of the goals of the developer (and since he is not getting paid, popularity is a cost) then a popular project is not necessarily successful from his point of view.
It's difficult to do well though, because you have to design the interface to every widget that people are going to want to invent before they invent it. So you either get an extremely limited set of widgets, or something too generic to be much use.
I think the alternative is that you would/should have Kpilot (KDE), gpilot (GNOME), xpilot (X11), iPilot (MacOSX), WinPilot (Win32), pilot (console), etc... variants of the same program, so that people could use whichever one happened to match whatever they already had installed. No need to reinvent the wheel, just make N different wheel mountings. Unfortunately there's no easy way to do that without multiplying the workload by something related to N. That is why people just use something like Qt or Gtk, and unfortunately using stuff like those is more or less what the guy was complaining about in the first place.
Slashdot Mirror
That's because if done incorrectly a chimney is dangerous. With an open door (or no door) it behaves like a blast furnace. (not a great idea in a wooden house). A straw roof by contrast allows most of the smoke to pass through without danger.
Yeah, but they covered it up.
Because I can write an OS or application myself (and distribute worldwide), but not build a car myself (and even if I could, I couldn't distribute it worldwide)? Patents on cars restrict the freedom of a few corporations and maybe a few real people, but not very many. Patents on software directly restricts the freedom of millions of real people. Not that I am necessartily advocating patents on cars, but anyway...
It already runs (slowly) on average PC hardware with PearPC. Why won't it run more quickly on average PC hardware with PearPC when the Mac CPU is also x86?
If I worked in a shop I'd want my children to benefit financially from it in the future too, doesn't mean I'd get my way.
Jimi Hendrix, Janis Joplin, Elvis Presley. If only extended copyrights had been available back in the sixties and seventies, these talented artistes would have been saved from the despair of pondering a penniless future, and not been driven to waste their lives.
What would you compress with? I found that e.g. bzip2-ing the .iso doesn't yield any saving for a dvd.
What if I have a digital door?
Such negativity I have never seen. Judging by all the responses to this article (the ones not advocating fraud that is) complaining about this or that failure with assorted Palms, can someone explain why would anybody even want a new Palm device? (or is it just a case of "hey, free stuff"?)
Perhaps because they had an ideolgy that caused them to believe in the revolutionary overthrow of all non communist governments around the world?
The reason we haven't gone to space is because there is nowhere to go, and nothing for a person to do that isn't a waste of time and resources. Machines will be doing all the interesting work in space for the forseeable future.
But a lawyer and whatever he recommends to do is likely to cost a whole lot more money than a LAN party can afford.
They are also correlated with the beer drinking vs wine drinking countries.
It is C's fault that programs make core dumps (or equivalent). It's not like core dumps, or rather the "undefined behaviour" that causes them when you do something nasty to a pointer, are not an intrinsic feature of the language. They are a flaw in the language that require the developer to do a lot of work to make sure the code doesn't trigger any undefined behaviour.
C, as a language exposes the developer to a whole class of errors that another language does not, and so it's perfectly reasonable to say that C is a flawed language in this respect. I know there are reasons (good or otherwise) why C has chosen to behave like this, but it's still a language flaw rather than an implementation flaw, and one that leads to the majority of the security vulnerabilities in your favourite OS (whichever it happens to be), and most of the random crashes and resource leakeage in your favourite applications.
Languages can be poorly designed just as much as implementations can be.
The trouble is, people (companies, politicians, open source visionaries, journalists, etc...) want open source to be something it isn't. Like it or not, as a whole, it's not a charity, it's not working for the third world, it's not business friendly, it's not anti-business, it's not trying to make the world a better place, it's not trying to provide a unified desktop, it's not trying to undermine capitalism, it's not trying to stop microsoft. Some particular individual projects might be any of those things, but for the most part, it's people writing something that they find useful or interesting for themselves, and don't mind sharing with anybody who is bothered enough to use it. As a whole it is directionless, except in the sense of general expansion of the number of both useful and useless projects out there.
Is it not possible that there's a memory leak/hog in the javascript code on the particular page, or does javascript not work like that?
Congress or somebody appointed by them. If Iran wants to do something they can force Iranian sites to be in .whatever.ir
If they ship your code without a license to do so, why don't you sue them? It's what they'd do to you.
But what's a successful open source project? Many of the developers (unpaid type, which is the vast majority) would consider it successful if the project does whatever they wanted it to do when they started. No need for all those other folks at all. What you are describing is a popular project. If popularity wasn't one of the goals of the developer (and since he is not getting paid, popularity is a cost) then a popular project is not necessarily successful from his point of view.
It's difficult to do well though, because you have to design the interface to every widget that people are going to want to invent before they invent it. So you either get an extremely limited set of widgets, or something too generic to be much use.
I think the alternative is that you would/should have Kpilot (KDE), gpilot (GNOME), xpilot (X11), iPilot (MacOSX), WinPilot (Win32), pilot (console), etc... variants of the same program, so that people could use whichever one happened to match whatever they already had installed. No need to reinvent the wheel, just make N different wheel mountings. Unfortunately there's no easy way to do that without multiplying the workload by something related to N. That is why people just use something like Qt or Gtk, and unfortunately using stuff like those is more or less what the guy was complaining about in the first place.
Happiness is not foolish.
No, it would be a tragedy all round.