only if the therapy was for sale on the market at a reasonable price
So if the government doesn't like your price, they strike down your patent?
I'd be thinking more along the lines of a non-commercial-use exception, as you allude to later in your comment (which, iiuc, doesn't exist in any country's patent law today).
I'm not terribly familiar with this particular exploit, but from the summary it looks as though it might have been avoided if they'd used automated reference-counting like C++ smart pointers.
You mean JVMs? They're generally written in C++. No, this doesn't entitle you to grin smugly.
Although it's essentially impossible for a well-intentioned Java program to have a buffer-overflow vulnerability, the JVM itself is full of security bugs, so bytecode from an untrusted source generally shouldn't be executed.
To put that another way: the Java language has proven to be successful in preventing things like buffer-overflow attacks, but the JVM has proven unsuccessful in providing a secure sandbox in which untrusted code can be executed in a contained virtual environment. People tend to confuse these two distinct properties when they say Java is insecure.
If we wrote production JVMs in Java rather than C++, this wouldn't be a problem. (Yes, I'm serious. Unfortunately, it looks as though the most prominent such JVM, "Jikes RVM", will forever remain a research project.)
I disagree, Presto is (still today) quite usable. Opera Mobile Classic, the current name of the Presto-powered browser for Android (which is available alongside the 'real' WebKit-based Opera), breathes new life into ancient Android phones. It doesn't cope with all sites, but it's a lot better than the old Android browser. (And Chrome doesn't run on Android 2.)
The column handling is awesome, which is a particular advantage on mobile devices.
For whatever reason, the 'real' Opera browser for Android, is absolutely awful. You can't even add your own search-engines beyond the ones it ships with. (Seriously.) It's nothing more than Chrome-but-terrible.
According to the nothing to hide argument, there is no threat to privacy unless the government uncovers unlawful activity, in which case a person has no legitimate justification to claim that it remain private. The nothing to hide argument and its variants are quite prevalent, and thus are worth addressing. In this essay, Solove critiques the nothing to hide argument and exposes its faulty underpinnings.
So by all means, spy on suspects instead of bulk-collecting false positives.
I'll try my hand at Devil's advocate (sorry GCHQ, I guess you're the Devil):
If the online actions of idiotic criminals is no longer visible to GCHQ, they'll have a much harder time finding who it is that's worth spying on IRL in the first place.
I'm sure there are plenty of smugglers and dealers dumb enough to send plain-text SMS detailing their crimes. If SMS were 'opaque', that would surely deprive GCHQ of, as it were, 'low-hanging fruit'.
Microsoft WANTS directX to remain relevant in the modern multiplatform environment.
PCs have 2 major remaining market niches:
1) Enterprise(/educational) workstations (Like, for doing WORK on.)
2) PC Gaming
Those 2 count for quite a lot, though. I don't think Microsoft are too worried about Mac and Linux supplanting Windows for gaming. It could happen, but that's been the case for years now. The same goes for workstations.
Security (the greatest downside imo), inability to (ever!) run on mobile devices, increased RAM use from pulling in a whole JVM, external dependency beyond a web-browser, immaturity of JavaFX. Historically Java applets would often cause a browser crash, but that seems not be a an issue these days (presumably as we've just got the horsepower to cope).
They're way more efficient than HTML+Javascript
JavaScript JIT compilers are pretty damn good these days. I suspect that you're right, but performance can be pretty good with web technologies. There are working audio/video-decoders written in JavaScript, for instance.
If you're worried about Oracle's treatment of the Java applet platform, choose a decent launcher/updater.
You mean OpenJDK?
Are you honestly asking what's wrong with having only one platform for development?
Yes, hence why I asked.
What's wrong with ANSI C as a single platform? Win32 as a single platform? Everthing lacking in HTML+Javascript is wrong with the web as a platform.
Well the standards are ever-growing. (And the technical barrier to creating a browser becomes ever more daunting, but I guess that's just the price we pay.)
And every disagreement on the best way of doing something which can't be alternatively implemented
It's called "the web". A mish-mash of incompatible technologies is not good for the web. On the other hand I kinda agree: I'd really like to see JavaScript die and be replaced (it's just awful), but it looks like we're stuck with it. There is an upside here though: stability of the web as a platform. There's one web, and it works on all sorts of devices. That's something that would be compromised by a plugins-for-everything web.
I suspect it's because the dinosaurs who licence their stuff to Netflix think Silverlight has trustworthy DRM magic dust, where other technologies aren't to be trusted.
I found that Amazon's Silverlight player was an absolute abomination. Very rarely worked on my Windows machine. Worked a good deal better on my Mac, if I didn't mind overheating the thing for hours at a time.
Slashdot Mirror
There can't have been many situations in human history where Just launch it into space was genuinely the right (and economical) move.
Err, no. Having a Gmail account that you never use, does not dimish your privacy.
That is one of several things thing which makes the corporate system inherently evil.
I'd have gone for "amoral" personally. Public companies are structured to chase profit above all else, not to deliberately pursue evil.
only if the therapy was for sale on the market at a reasonable price
So if the government doesn't like your price, they strike down your patent?
I'd be thinking more along the lines of a non-commercial-use exception, as you allude to later in your comment (which, iiuc, doesn't exist in any country's patent law today).
I've been waiting on WebKit to catch up. Any day now, Chrome...
Smart pointers are helpful though. Might've prevented this particular exploit.
I'm not terribly familiar with this particular exploit, but from the summary it looks as though it might have been avoided if they'd used automated reference-counting like C++ smart pointers.
Does that affect your day-to-day work?
If you really want Java-written-in-Java, take a look at Jikes RVM.
You mean JVMs? They're generally written in C++. No, this doesn't entitle you to grin smugly.
Although it's essentially impossible for a well-intentioned Java program to have a buffer-overflow vulnerability, the JVM itself is full of security bugs, so bytecode from an untrusted source generally shouldn't be executed.
To put that another way: the Java language has proven to be successful in preventing things like buffer-overflow attacks, but the JVM has proven unsuccessful in providing a secure sandbox in which untrusted code can be executed in a contained virtual environment. People tend to confuse these two distinct properties when they say Java is insecure.
If we wrote production JVMs in Java rather than C++, this wouldn't be a problem. (Yes, I'm serious. Unfortunately, it looks as though the most prominent such JVM, "Jikes RVM", will forever remain a research project.)
So long as we're writing in C, this kind of thing (buffer overflows in particular) will probably continue.
(Lest I start a flame-war: C is awesome in its way, but more than almost any other language, it really does make it easy to miss things like this.)
I disagree, Presto is (still today) quite usable. Opera Mobile Classic, the current name of the Presto-powered browser for Android (which is available alongside the 'real' WebKit-based Opera), breathes new life into ancient Android phones. It doesn't cope with all sites, but it's a lot better than the old Android browser. (And Chrome doesn't run on Android 2.)
The column handling is awesome, which is a particular advantage on mobile devices.
For whatever reason, the 'real' Opera browser for Android, is absolutely awful. You can't even add your own search-engines beyond the ones it ships with. (Seriously.) It's nothing more than Chrome-but-terrible.
That seems a reasonable analysis.
A small nitpick: let's not confuse 'PC gaming' with 'gaming on Windows'.
Here is a (freely accessible) paper on the matter: 'I've Got Nothing to Hide' and Other Misunderstandings of Privacy.
And here is the Slashdot thread on the paper.
According to the nothing to hide argument, there is no threat to privacy unless the government uncovers unlawful activity, in which case a person has no legitimate justification to claim that it remain private. The nothing to hide argument and its variants are quite prevalent, and thus are worth addressing. In this essay, Solove critiques the nothing to hide argument and exposes its faulty underpinnings.
So by all means, spy on suspects instead of bulk-collecting false positives.
I'll try my hand at Devil's advocate (sorry GCHQ, I guess you're the Devil):
If the online actions of idiotic criminals is no longer visible to GCHQ, they'll have a much harder time finding who it is that's worth spying on IRL in the first place.
I'm sure there are plenty of smugglers and dealers dumb enough to send plain-text SMS detailing their crimes. If SMS were 'opaque', that would surely deprive GCHQ of, as it were, 'low-hanging fruit'.
Microsoft WANTS directX to remain relevant in the modern multiplatform environment.
PCs have 2 major remaining market niches:
1) Enterprise(/educational) workstations (Like, for doing WORK on.)
2) PC Gaming
Those 2 count for quite a lot, though. I don't think Microsoft are too worried about Mac and Linux supplanting Windows for gaming. It could happen, but that's been the case for years now. The same goes for workstations.
Yes, we're stuck on IE 6, and thus XP, at work, but it is a lot better than the alternatives.
How? The disadvantages here are enormous.
Compared to Squirrel Mail that our competitor uses, OWA is great.
Seriously? You're really arguing This product that I like uses a certain framework, therefore it is a good framework.
That doesn't mean it is fragile. It's just built for a certain version of a certain browser.
Errr.... poor browser support is a legitimate complaint when it comes to web apps.
Just so I know, you are kidding, right?
the download can include drive-by malware
Can? If memory serves, you have to opt-out of McAfee, in the Java installer.
Forgive me, but what's bad with Java applets?
Security (the greatest downside imo), inability to (ever!) run on mobile devices, increased RAM use from pulling in a whole JVM, external dependency beyond a web-browser, immaturity of JavaFX. Historically Java applets would often cause a browser crash, but that seems not be a an issue these days (presumably as we've just got the horsepower to cope).
They're way more efficient than HTML+Javascript
JavaScript JIT compilers are pretty damn good these days. I suspect that you're right, but performance can be pretty good with web technologies. There are working audio/video-decoders written in JavaScript, for instance.
If you're worried about Oracle's treatment of the Java applet platform, choose a decent launcher/updater.
You mean OpenJDK?
Are you honestly asking what's wrong with having only one platform for development?
Yes, hence why I asked.
What's wrong with ANSI C as a single platform? Win32 as a single platform? Everthing lacking in HTML+Javascript is wrong with the web as a platform.
Well the standards are ever-growing. (And the technical barrier to creating a browser becomes ever more daunting, but I guess that's just the price we pay.)
And every disagreement on the best way of doing something which can't be alternatively implemented
It's called "the web". A mish-mash of incompatible technologies is not good for the web. On the other hand I kinda agree: I'd really like to see JavaScript die and be replaced (it's just awful), but it looks like we're stuck with it. There is an upside here though: stability of the web as a platform. There's one web, and it works on all sorts of devices. That's something that would be compromised by a plugins-for-everything web.
I really am not happy with the way the web is reducing itself to Ein Language, Ein Platform, Ein Consortium
The only free and open 'non-standard web technologies' I can think of are Java applets (oh dear) and Dart.
Flash and Silverlight are proprietary.
Anyway, what's wrong with the web as a single platform? You still have your pick of browsers.
I suspect it's because the dinosaurs who licence their stuff to Netflix think Silverlight has trustworthy DRM magic dust, where other technologies aren't to be trusted.
Could Windows XP be approaching bug-free perfection?
Seems rather more likely it's just no longer worth targetting.
Apparently in the US they also support Flash.
I found that Amazon's Silverlight player was an absolute abomination. Very rarely worked on my Windows machine. Worked a good deal better on my Mac, if I didn't mind overheating the thing for hours at a time.
mission creep, while a real thing should never be an excuse to not use something worthwhile.
What's the difference between this position, and mission creep, while a real thing, should just be ignored?
But not the hardware part? Not a ringing endorsement, huh.