Flash doesn't work completely reliably on any platform I have tried. I find that Adobe Flash on 32bit Linux works about as well as the OS X version (meaning: it's usable but it does have occasional problems).
The main problem people are having is that there is no 64bit Linux version of Flash, so all you can do is run it in some emulated environment.
But company-wide outages happen with regularity with non-hosted applications as well: corporate servers go down, corporate networks go down, worms infest every machine at once, etc.
Google does need to address this because people perceive it differently. But I think even given these outages, you're still better off with hosted apps. Besides, Google offers off-line features as well.
Ah, I see, Microsoft's paid corporate moderators are in full force again.
I'm not being sarcastic, though: add up the number of hours you spend installing, upgrading, and maintaining Windows and Office locally, and compare it with Google's downtime. In fact, just waiting for Microsoft Office to load probably adds up to more downtime.
Net neutrality can mean many different things. It can mean neutrality towards customers. It can mean neutrality towards service providers. It can mean neutrality towards protocols. It can mean max-out-the-line usage. People need to be specific about which they mean when they talk.
You developed the technology as part of your work, work that you got paid for. Therefore, the technology belongs to the company and they have a right to patent it. Generally, employment contracts require you to file patents.
If your employment contract doesn't require you to file patents, you can legally get away with not filing a patent, but the company may still choose to terminate you.
What you can do is negotiate about it and try to convince the company that filing bad patents is bad for business. You can also try to minimize the damage by making the patent clearer and more specific. And if the company has unreasonable patent policies, you yourself may want to look for a new job.
The amount of downtime each individual user experiences from their local Microsoft Windows and Microsoft Office installations is far higher than the few hours per year people may experience with hosted apps.
Yes, what suckage. Who on earth wants a simple, dynamic, low-cruft language with first-class functions, closures, lambdas and a prototype object system that still manages to look like the C language family? Evil, evil, evil...
There have been dozens of such languages around. The hard part is also making it (1) fast, and (2) correct. And JavaScript is neither.
When the problem is with a public entity, especially government, the truth should be shouted from the rooftops.
Nobody has shown there to be a problem.
They are hiding the fact that it is flawed badly and quite possibly BY DESIGN.
Of course it is "flawed" by design, except that it is not a flaw. Subway tokens were never unforgeable before, so why should that all of a sudden be a requirement?
So they made FREE SPEECH a TOP priority
I think the MIT students should be allowed to publish their findings, just like I think people should be allowed to publish obscenities on the web. But people should also recognize that what the MIT students did is stupid. People should also recognize that arguments like yours are stupid.
This is really CYA on behalf of the incompetent people running the Boston system.
They didn't have a secure system before (tokens), why should they have one now?
They're just pointing it out. I'm sure others are already exploiting the flaws even before the announcement.
Of course, people are exploiting it, just like they were exploiting tokens before. It's factored into the overall cost.
They made the cheap choice ( unvalidated stored value cards w/ crappy encryption of the data ) and it bit them on the ass.
Economically, I don't see a problem. In fact, the rational thing is to stick with the current system.
Of course, they have a PR problem. But Obama has a PR problem over the false claims that he is a Muslim, too. The problem here is with the accusers, not the system.
and publicises the incompetence of the administration responsible.
The only incompetence I see is by a bunch of undergraduates who don't understand the economics of mass transit.
The problem with Java and the CLR in browsers are mostly with the libraries, not the virtual machines. So, I think the thing to do would be to start with, say, the CLR and develop an open applet environment and browser integration based on it. This environment could even be emulated in Silverlight, allowing things to run without any install on Windows.
People who write bullshit like "But early binding in any dynamic code loading scenario like the web requires a prioritization or reservation mechanism to avoid early versus late binding conflicts." have no business being associated with language design. It's no wonder that JavaScript sucks so badly as a language.
I would guess GP referred to things like Core Animation.
Yeah, so after creating a bunch of animations, Apple refactored their code and made a library out of it. Big deal.
And the reason why Apple is a little ahead in having libraries for things like animation is simply because they are shipping higher end hardware at a premium price. Linux and Windows go easy on such things because a lot of people use low-end hardware.
Factoring out some animation code is not innovation, sorry. And it doesn't really solve the fundamental problems of today's WIMP interfaces, which Apple suffers from just as much as Windows, Gnome, and KDE.
I can understand you not wanting to pay a dime for the Mac Hardware to learn the tools
I have owned about a dozen Macs (initially, because of the hype, later, for work-related reasons).
Please don't talk about Glade/Python as being superior to ObjC/Cocoa and IB.
Well, I'm sorry you don't want to hear it, but after developing for both, my conclusions is that even Glade/Python is better than Objective-C/Cocoa/InterfaceBuilder.
I'm guessing you've never used Cocoa. Try it, and be amazed.
You're guessing wrong. I have both used and programmed Cocoa, and I'm underwhelmed: programming it is tedious. I think people who think that Cocoa is good must be MFC refugees and have never seen anything else.
Yes, f-stop is a big piece of it, but the other piece is the size of the pixels.
Only because f-stop is relative aperture, which ends up creating a dependency between sensor size, FOV, and aperture. If you look at it in terms of FOV and absolute aperture, it makes no difference what the sensor size is.
In different words, an f2.0 35-100 lens on a half size sensor gives you the same number of photons per pixel as an f2.8 70-200 lens on a full size sensor. And if you look up the weights and sizes of the Olympus and Canon lenses, you'll find that they're similar.
The situations of the Linux kernel and the Boston subway are completely different. In the case of the Linux kernel, people need to know because it's their security that's at stake. In the case of the Boston subway, it all comes down to the economics of fare evasion and doesn't affect anybody's security (and you can be certain that the Boston subway knew about this and accepted it when they bought the system).
Now, I think the MIT students have a first amendment right to disclose this. However, I also think that these kinds of antics deserve reproach: people should point out that this was a stupid thing to do.
1) Google can pull an Apple'ish move and push for carriers to open up the networks.
Apple hasn't pushed for carriers to open up the networks to anybody but Apple. With an iPhone, you're far more restricted than with a Palm, Nokia, or Windows Mobile on any of the major carriers.
It's not about the language. It's about the libraries. And Apple is currently second-to-none in that department for user interaction.
Really? As demonstrated by what?
Looks to me Apple has the same pushbutton/scrollbar/slider stuff as anybody else. And Objective C with XCode seems clunky and outdated compared to Glade and Python, or C# and Stetic.
You start with a tiny sensor, add a big zoom which further reduces the light hitting the sensor
You can put an f4 lens on a full frame camera and use it at ISO 400, or you can put an equivalent f2.8 lens on a half frame camera and use it at ISO 200, and you get about the same amount of noise. But the half frame camera will be smaller and lighter and have somewhat larger DOF (it will also have a worse diffraction limit, but that doesn't matter here). So, you don't really gain anything by using a larger sensor.
I think this myth that larger formats mean less noise comes from film, where it really is true due to the inherent noise of film. But for digital, noise is photon noise and that pretty much only depends on the number of photons per pixel, not pixel size.
The museum had a policy of no photographs. This is hardly uncommon: not only do many people find it annoying to stumble over photographers and deal with flashes while they're trying to look at art, but repeated exposure to light flashes can damage art.
That's a lame excuse; they could simply have a no-flash policy. If they're busy, they could also have a no-tripod policy. But there is no reason to restrict photography itself.
No, the real reason museums have a "no photographs" policy is because they want to control publication of the art in order to make money from it, and that is not OK. The copyright on the art has, in most cases, expired long ago, and it should be available to the public.
"No photography" restrictions are reprehensible for an organization whose purpose is to make art accessible to the public.
Slashdot Mirror
Flash doesn't work completely reliably on any platform I have tried. I find that Adobe Flash on 32bit Linux works about as well as the OS X version (meaning: it's usable but it does have occasional problems).
The main problem people are having is that there is no 64bit Linux version of Flash, so all you can do is run it in some emulated environment.
But company-wide outages happen with regularity with non-hosted applications as well: corporate servers go down, corporate networks go down, worms infest every machine at once, etc.
Google does need to address this because people perceive it differently. But I think even given these outages, you're still better off with hosted apps. Besides, Google offers off-line features as well.
A GPLv2 UI library is essentially useless. They either need a classpath exception or LGPL.
Ah, I see, Microsoft's paid corporate moderators are in full force again.
I'm not being sarcastic, though: add up the number of hours you spend installing, upgrading, and maintaining Windows and Office locally, and compare it with Google's downtime. In fact, just waiting for Microsoft Office to load probably adds up to more downtime.
what do you microsoft fanbois have to say about this, we really would like to hear.
Since those Microsoft fanbois are likely paid by Microsoft, it's not hard to imagine...
As long as the only source for this is Declan McCullagh, I don't believe it. King probably hung up on him because he was being rude.
Is there any other story on this? Any confirmation?
Net neutrality can mean many different things. It can mean neutrality towards customers. It can mean neutrality towards service providers. It can mean neutrality towards protocols. It can mean max-out-the-line usage. People need to be specific about which they mean when they talk.
You developed the technology as part of your work, work that you got paid for. Therefore, the technology belongs to the company and they have a right to patent it. Generally, employment contracts require you to file patents.
If your employment contract doesn't require you to file patents, you can legally get away with not filing a patent, but the company may still choose to terminate you.
What you can do is negotiate about it and try to convince the company that filing bad patents is bad for business. You can also try to minimize the damage by making the patent clearer and more specific. And if the company has unreasonable patent policies, you yourself may want to look for a new job.
The amount of downtime each individual user experiences from their local Microsoft Windows and Microsoft Office installations is far higher than the few hours per year people may experience with hosted apps.
What is stupid about it?
What is "stupid" about is their and your claim that this is a flaw.
What is stupid about knowing the truth?
Nothing. But there is something stupid about publishing the truth in some cases. This is one.
Yes, what suckage. Who on earth wants a simple, dynamic, low-cruft language with first-class functions, closures, lambdas and a prototype object system that still manages to look like the C language family? Evil, evil, evil...
There have been dozens of such languages around. The hard part is also making it (1) fast, and (2) correct. And JavaScript is neither.
And at $120+, the HP 49g+/50g also seems rather overpriced...
When the problem is with a public entity, especially government, the truth should be shouted from the rooftops.
Nobody has shown there to be a problem.
They are hiding the fact that it is flawed badly and quite possibly BY DESIGN.
Of course it is "flawed" by design, except that it is not a flaw. Subway tokens were never unforgeable before, so why should that all of a sudden be a requirement?
So they made FREE SPEECH a TOP priority
I think the MIT students should be allowed to publish their findings, just like I think people should be allowed to publish obscenities on the web. But people should also recognize that what the MIT students did is stupid. People should also recognize that arguments like yours are stupid.
This is really CYA on behalf of the incompetent people running the Boston system.
They didn't have a secure system before (tokens), why should they have one now?
They're just pointing it out. I'm sure others are already exploiting the flaws even before the announcement.
Of course, people are exploiting it, just like they were exploiting tokens before. It's factored into the overall cost.
They made the cheap choice ( unvalidated stored value cards w/ crappy encryption of the data ) and it bit them on the ass.
Economically, I don't see a problem. In fact, the rational thing is to stick with the current system.
Of course, they have a PR problem. But Obama has a PR problem over the false claims that he is a Muslim, too. The problem here is with the accusers, not the system.
and publicises the incompetence of the administration responsible.
The only incompetence I see is by a bunch of undergraduates who don't understand the economics of mass transit.
Any ideas ?
The problem with Java and the CLR in browsers are mostly with the libraries, not the virtual machines. So, I think the thing to do would be to start with, say, the CLR and develop an open applet environment and browser integration based on it. This environment could even be emulated in Silverlight, allowing things to run without any install on Windows.
People who write bullshit like "But early binding in any dynamic code loading scenario like the web requires a prioritization or reservation mechanism to avoid early versus late binding conflicts." have no business being associated with language design. It's no wonder that JavaScript sucks so badly as a language.
I would guess GP referred to things like Core Animation.
Yeah, so after creating a bunch of animations, Apple refactored their code and made a library out of it. Big deal.
And the reason why Apple is a little ahead in having libraries for things like animation is simply because they are shipping higher end hardware at a premium price. Linux and Windows go easy on such things because a lot of people use low-end hardware.
Factoring out some animation code is not innovation, sorry. And it doesn't really solve the fundamental problems of today's WIMP interfaces, which Apple suffers from just as much as Windows, Gnome, and KDE.
I can understand you not wanting to pay a dime for the Mac Hardware to learn the tools
I have owned about a dozen Macs (initially, because of the hype, later, for work-related reasons).
Please don't talk about Glade/Python as being superior to ObjC/Cocoa and IB.
Well, I'm sorry you don't want to hear it, but after developing for both, my conclusions is that even Glade/Python is better than Objective-C/Cocoa/InterfaceBuilder.
I'm guessing you've never used Cocoa. Try it, and be amazed.
You're guessing wrong. I have both used and programmed Cocoa, and I'm underwhelmed: programming it is tedious. I think people who think that Cocoa is good must be MFC refugees and have never seen anything else.
Yes, f-stop is a big piece of it, but the other piece is the size of the pixels.
Only because f-stop is relative aperture, which ends up creating a dependency between sensor size, FOV, and aperture. If you look at it in terms of FOV and absolute aperture, it makes no difference what the sensor size is.
In different words, an f2.0 35-100 lens on a half size sensor gives you the same number of photons per pixel as an f2.8 70-200 lens on a full size sensor. And if you look up the weights and sizes of the Olympus and Canon lenses, you'll find that they're similar.
The situations of the Linux kernel and the Boston subway are completely different. In the case of the Linux kernel, people need to know because it's their security that's at stake. In the case of the Boston subway, it all comes down to the economics of fare evasion and doesn't affect anybody's security (and you can be certain that the Boston subway knew about this and accepted it when they bought the system).
Now, I think the MIT students have a first amendment right to disclose this. However, I also think that these kinds of antics deserve reproach: people should point out that this was a stupid thing to do.
1) Google can pull an Apple'ish move and push for carriers to open up the networks.
Apple hasn't pushed for carriers to open up the networks to anybody but Apple. With an iPhone, you're far more restricted than with a Palm, Nokia, or Windows Mobile on any of the major carriers.
It's not about the language. It's about the libraries. And Apple is currently second-to-none in that department for user interaction.
Really? As demonstrated by what?
Looks to me Apple has the same pushbutton/scrollbar/slider stuff as anybody else. And Objective C with XCode seems clunky and outdated compared to Glade and Python, or C# and Stetic.
You start with a tiny sensor, add a big zoom which further reduces the light hitting the sensor
You can put an f4 lens on a full frame camera and use it at ISO 400, or you can put an equivalent f2.8 lens on a half frame camera and use it at ISO 200, and you get about the same amount of noise. But the half frame camera will be smaller and lighter and have somewhat larger DOF (it will also have a worse diffraction limit, but that doesn't matter here). So, you don't really gain anything by using a larger sensor.
I think this myth that larger formats mean less noise comes from film, where it really is true due to the inherent noise of film. But for digital, noise is photon noise and that pretty much only depends on the number of photons per pixel, not pixel size.
The museum had a policy of no photographs. This is hardly uncommon: not only do many people find it annoying to stumble over photographers and deal with flashes while they're trying to look at art, but repeated exposure to light flashes can damage art.
That's a lame excuse; they could simply have a no-flash policy. If they're busy, they could also have a no-tripod policy. But there is no reason to restrict photography itself.
No, the real reason museums have a "no photographs" policy is because they want to control publication of the art in order to make money from it, and that is not OK. The copyright on the art has, in most cases, expired long ago, and it should be available to the public.
"No photography" restrictions are reprehensible for an organization whose purpose is to make art accessible to the public.