Possibly, but most people here are only against stupid, trivial patents, patents for ideas, software, business models and genes, and patents for stuff that's already covered by copyright. If there's anything that deserves to be patentable, I'd say this is it.
The only assumption I'm making is that Google isn't completely retarded when it comes to security.
When a seemingly competent company releases an OS based on a pretty good kernel, and the OS appears to have a permission-based security system, I consider it pretty safe to assume that that's what it is. Sure, it could be merely a facade without any actual security to back it up, but that's not something I'd expect from Google. I mean, why even bother with notifying users with the permission an app needs, when there are no permissions and the security is completely absent?
I know Google employs a lot of really smart people, and I feel it's safe to trust they don't want Android to be a joke. My assumptions would have been different had this been a Microsoft product.
If a permission is not computed by usage, then what in that statement would make me think a runtime check is performed?
Because it's called "permission". If there's no actual runtime check whether you have permission for what you're trying to do, then it's not really much of a permission, is it? It'd just be a meaningless flag that doesn't do anything.
If this isn't obvious to you, then I hope you don't do anything that involves any kind of security.
It was extremely obvious from what he said. The XML specifies what permissions the program needs. Any permission not specified in the XML, it doesn't get. So it won't have that permission, and won't be making any phone calls that it wanted to.
It sounds like a rather simple and obvious way to handle security.
There's absolutely no reason that this should be the case. I can't speak for Windows Mobile, but the Symbian kernel has a capability model that makes it relatively easy to protect against this kind of thing. Applications, by default, can only read a few system locations (shared libraries and so forth) and can only write into their own directory. Each shared library and each application has a set of capability bits.
I don't know the technical details of it, but it's quite possible that Android does exactly the same thing. When you install an app, you get to see a list of the permissions it needs. At that point, an informed user will stop and consider if that app really needs all those permissions, and will not install it if he doesn't trust it. Unfortunately, most users don't do that and just blindly assume that every app can be trusted, giving it all the permissions it requests, whether they make sense or not.
Meanwhile, in the little town where I lived in Southern Oregon a few years ago, a Natural Gas 500MW power plant cost something like $80-100 million to build.
It's true. Fossil fuel plants are far cheaper to build, but their fuel usage costs a lot more.
I'm not so sure. Years ago I heard that French nuclear power is heavily subsidized because France considers nuclear independence strategically important. (Although I doubt that subsidy is 10 cents per kWh.)
I don't see why it's a problem distinguish between levels of failure just like your system apparently distinguishes between levels of passing.
In Netherland, schools give grades from 1 to 10. 5 and below are failing grades, 6 and up are passing grades. A 5 is bad, but doesn't drag your average down much. It can easily be compensated by a few slightly better grades. A 1 drags your average way down, and requires spectacular results on your other tests if you want to end up with a passing grade.
If you're just interested in fail/pass, then why not simply have a single failing grade and a single passing grade? I've never quite understood the sense behind the American alphabet soup.
Exactly. Ceres (and some other planetoids) lost their planet status because people started to discover more and more similar objects in the same orbit. Exactly what happened to Pluto.
Yeah, I liked Pluto. It was my favourite planet when I was a kid. But I completely agree with its reclassification.
Unfortunately the categorisation isn't quite so straightforward - we've got gas giant being a subset of planet for example, but "dwarf planet" is not a subset of "planet" (aside from being misleading from the name, it also seems unclear why say Mercury and Jupiter are subsets of one thing, distinct to Pluto being something else).
What's unclear about that? Mercury and Jupiter have an orbit of their own, whereas Pluto is part of a belt of similar objects. Just like Ceres.
Of course we knew they had to have chemical weapons, because Donald Rumsfeld sold them to Saddam in the '80s. But from what I remember, the chemicals had degraded over the years and wouldn't be terribly effective anymore.
They weren't hiding any of this if you were paying attention. The marketing speak became about terrorism when they tried to sell it to the American public, and it became about WMD when they tried to sell it to the United Nations.
That's the real problem in their execution: that they couldn't rely on the truth, but needed to sell their plans through lies.
Lying to the enemy is one thing, but lying to your own people and your allies is completely wrong. It makes you a crook and your cause unjust.
There's a huge difference between revealing secret operational plans, and revealing secret reports about stuff that already happened.
If on June 3 1944 he publishes the plans for Operation Overlord, that's treason and he needs to hang. If on that same date he publishes about death squads in Italy in 1943, he's doing a public service. Accountability after the facts is never wrong in my book.
Who cares what you fork? It's not about the code, it's about the developers and the customers. Sure, you can keep using the existing stuff for free, but Oracle would determine the direction of all meaningful future development.
That's good to hear. I had an iPhone 3G, and never liked its mail client.
What didn't you like, out of curiosity? I thought it was quite acceptable and not much different from iPhone OS 3's mail client.
I remember having to go through lots of menus every time I wanted to check a different mail account. It's possible that Android's standard mail client is no different, but having a separate app for gmail and one for my other email account basically solves the problem.
You'd think there'd be an easier way to do that, but so far, this works well enough for me.
Someone else already pointed out the settings configuration on Android, which is massive, hard to navigate, and hard to find things in.
There's a lot of stuff in it alright, but I also remember having trouble finding stuff on the iPhone settings.
Scrolling is, imo, terrible on Android. It feels like I'm working hard to scroll, and there are delays from when I move my finger to when the scrolling starts. This seems like a minor complaint, but this sort of thing (input lag from when you make the input to when you see results) is really a big deal in HCI.
I agree. I'm having similar problems with scrolling on my big screen iMac. Really annoying. On Android too, scrolling could have been a bit snappier. Should have been. UI needs to run at the highest priority, and the scheduler needs to keep user interaction in mind.
That reminds me: doesn't the linux kernel always schedule server style: efficiency first instead of responsiveness first? Does anyone know if that's been fixed for Android? It seems like a rather obvious issue.
The mail client is simply atrocious.
It's buggy and unreliable, but other than that, I prefer the mail client over the gmail client. And I prefer both over iPhone's mail client, which I really hated (whereas Mail.app is one of my favourite mail clients).
But the big advantage of Android here is: you can install other mail clients. Not possible on the iPhone.
Incidentally, iOS4 has a much better mail client than iPhone OS 3.
That's good to hear. I had an iPhone 3G, and never liked its mail client.
Tap to zoom is far better on iOS4.
Do people tap to zoom? I always use pinch and zoom. Works fine, though (like scrolling) just a tiny bit too slow.
There are some people who give really awful handshakes. There's your early squeezers, but also people who squeeze too hard, as if they want to squeeze you into submission. And people who shake hands at a really odd angle so you're basically at their mercy.
I don't know why these people can't do a normal handshake, but the awful travesty they make of it leaves me with a desire to kick them. They seem to be mostly men, and business/manager types. Really odd, because it seems to me that in that line of work, handshakes can be pretty important. Yet they're some of the most awful handshakers I've ever met.
It seems silly to suggest lessons for something as trivial as a handshake, but some people seem to need it.
Slashdot Mirror
User operation prohibition on DVDs. If your DVD player ignores them, it may be in violation of the DVD format license.
... and I'd like to know where you got it, because I would also like a DVD player that does what I want.
It never ceases to amaze me how legal corruption in US politics is.
And acceptable too. How come nobody votes these people out of office?
Isn't that an argument for patents, though?
Possibly, but most people here are only against stupid, trivial patents, patents for ideas, software, business models and genes, and patents for stuff that's already covered by copyright. If there's anything that deserves to be patentable, I'd say this is it.
The only assumption I'm making is that Google isn't completely retarded when it comes to security.
When a seemingly competent company releases an OS based on a pretty good kernel, and the OS appears to have a permission-based security system, I consider it pretty safe to assume that that's what it is. Sure, it could be merely a facade without any actual security to back it up, but that's not something I'd expect from Google. I mean, why even bother with notifying users with the permission an app needs, when there are no permissions and the security is completely absent?
I know Google employs a lot of really smart people, and I feel it's safe to trust they don't want Android to be a joke. My assumptions would have been different had this been a Microsoft product.
If a permission is not computed by usage, then what in that statement would make me think a runtime check is performed?
Because it's called "permission". If there's no actual runtime check whether you have permission for what you're trying to do, then it's not really much of a permission, is it? It'd just be a meaningless flag that doesn't do anything.
If this isn't obvious to you, then I hope you don't do anything that involves any kind of security.
It was extremely obvious from what he said. The XML specifies what permissions the program needs. Any permission not specified in the XML, it doesn't get. So it won't have that permission, and won't be making any phone calls that it wanted to.
It sounds like a rather simple and obvious way to handle security.
Unfortunately I noticed that there is always someone ready to do the same job than you for cheaper. It will always be this way.
But will they do it equally well? That is always the question.
The stock traders make what they do because they know what buttons to push and when.
No, they make so much money because they control money. People who control lots of money always make lots of money. They can make sure of that.
There's absolutely no reason that this should be the case. I can't speak for Windows Mobile, but the Symbian kernel has a capability model that makes it relatively easy to protect against this kind of thing. Applications, by default, can only read a few system locations (shared libraries and so forth) and can only write into their own directory. Each shared library and each application has a set of capability bits.
I don't know the technical details of it, but it's quite possible that Android does exactly the same thing. When you install an app, you get to see a list of the permissions it needs. At that point, an informed user will stop and consider if that app really needs all those permissions, and will not install it if he doesn't trust it. Unfortunately, most users don't do that and just blindly assume that every app can be trusted, giving it all the permissions it requests, whether they make sense or not.
78 miles per gallon is about 3 liter for 100 km.
That'd be 33 km per liter. That's pretty good!
Meanwhile, in the little town where I lived in Southern Oregon a few years ago, a Natural Gas 500MW power plant cost something like $80-100 million to build.
It's true. Fossil fuel plants are far cheaper to build, but their fuel usage costs a lot more.
I'm not so sure. Years ago I heard that French nuclear power is heavily subsidized because France considers nuclear independence strategically important. (Although I doubt that subsidy is 10 cents per kWh.)
I don't see why it's a problem distinguish between levels of failure just like your system apparently distinguishes between levels of passing.
In Netherland, schools give grades from 1 to 10. 5 and below are failing grades, 6 and up are passing grades. A 5 is bad, but doesn't drag your average down much. It can easily be compensated by a few slightly better grades. A 1 drags your average way down, and requires spectacular results on your other tests if you want to end up with a passing grade.
If you're just interested in fail/pass, then why not simply have a single failing grade and a single passing grade? I've never quite understood the sense behind the American alphabet soup.
Exactly. Ceres (and some other planetoids) lost their planet status because people started to discover more and more similar objects in the same orbit. Exactly what happened to Pluto.
Yeah, I liked Pluto. It was my favourite planet when I was a kid. But I completely agree with its reclassification.
Unfortunately the categorisation isn't quite so straightforward - we've got gas giant being a subset of planet for example, but "dwarf planet" is not a subset of "planet" (aside from being misleading from the name, it also seems unclear why say Mercury and Jupiter are subsets of one thing, distinct to Pluto being something else).
What's unclear about that? Mercury and Jupiter have an orbit of their own, whereas Pluto is part of a belt of similar objects. Just like Ceres.
Of course we knew they had to have chemical weapons, because Donald Rumsfeld sold them to Saddam in the '80s. But from what I remember, the chemicals had degraded over the years and wouldn't be terribly effective anymore.
They weren't hiding any of this if you were paying attention. The marketing speak became about terrorism when they tried to sell it to the American public, and it became about WMD when they tried to sell it to the United Nations.
That's the real problem in their execution: that they couldn't rely on the truth, but needed to sell their plans through lies.
Lying to the enemy is one thing, but lying to your own people and your allies is completely wrong. It makes you a crook and your cause unjust.
There's a huge difference between revealing secret operational plans, and revealing secret reports about stuff that already happened.
If on June 3 1944 he publishes the plans for Operation Overlord, that's treason and he needs to hang. If on that same date he publishes about death squads in Italy in 1943, he's doing a public service. Accountability after the facts is never wrong in my book.
What you're basically saying is that it's a disgrace that a comedy show turns out to be the best news show of the US.
No, he takes people's money to wage war on other people's behalf. He's a politician.
Who cares what you fork? It's not about the code, it's about the developers and the customers. Sure, you can keep using the existing stuff for free, but Oracle would determine the direction of all meaningful future development.
That's good to hear. I had an iPhone 3G, and never liked its mail client.
What didn't you like, out of curiosity? I thought it was quite acceptable and not much different from iPhone OS 3's mail client.
I remember having to go through lots of menus every time I wanted to check a different mail account. It's possible that Android's standard mail client is no different, but having a separate app for gmail and one for my other email account basically solves the problem.
You'd think there'd be an easier way to do that, but so far, this works well enough for me.
Someone else already pointed out the settings configuration on Android, which is massive, hard to navigate, and hard to find things in.
There's a lot of stuff in it alright, but I also remember having trouble finding stuff on the iPhone settings.
Scrolling is, imo, terrible on Android. It feels like I'm working hard to scroll, and there are delays from when I move my finger to when the scrolling starts. This seems like a minor complaint, but this sort of thing (input lag from when you make the input to when you see results) is really a big deal in HCI.
I agree. I'm having similar problems with scrolling on my big screen iMac. Really annoying. On Android too, scrolling could have been a bit snappier. Should have been. UI needs to run at the highest priority, and the scheduler needs to keep user interaction in mind.
That reminds me: doesn't the linux kernel always schedule server style: efficiency first instead of responsiveness first? Does anyone know if that's been fixed for Android? It seems like a rather obvious issue.
The mail client is simply atrocious.
It's buggy and unreliable, but other than that, I prefer the mail client over the gmail client. And I prefer both over iPhone's mail client, which I really hated (whereas Mail.app is one of my favourite mail clients).
But the big advantage of Android here is: you can install other mail clients. Not possible on the iPhone.
Incidentally, iOS4 has a much better mail client than iPhone OS 3.
That's good to hear. I had an iPhone 3G, and never liked its mail client.
Tap to zoom is far better on iOS4.
Do people tap to zoom? I always use pinch and zoom. Works fine, though (like scrolling) just a tiny bit too slow.
There are some people who give really awful handshakes. There's your early squeezers, but also people who squeeze too hard, as if they want to squeeze you into submission. And people who shake hands at a really odd angle so you're basically at their mercy.
I don't know why these people can't do a normal handshake, but the awful travesty they make of it leaves me with a desire to kick them. They seem to be mostly men, and business/manager types. Really odd, because it seems to me that in that line of work, handshakes can be pretty important. Yet they're some of the most awful handshakers I've ever met.
It seems silly to suggest lessons for something as trivial as a handshake, but some people seem to need it.
Probably 68% of Nexus One users already knew there's not going to be a Nexus Two.