Free market does not require people to play by the rules or anything like that because there cannot be government rules.
Even most conservatives don't believe that free markets can work unless there is government restraint on monopolies, which tend to form in any free market due to economies of scale. Ironic that I have to point this out in the middle of a discussion about a cable company merger.
Being testable against observations is an essential characteristic of a hypothesis. If it isn't testable against observations, it isn't a "non-ideal" hypothesis, it is pseudoscience.
How adorable that you can simply throw away the observational sciences.
I said that a hypothesis has to be "testable against observations." Presumably the observational sciences have observations. If their theories aren't testable against observation, then they aren't science.
I don't really understand the rage being directed at Google here. They have fixed the issue in new versions of Android. If they back-ported the fix to 4.3 (assuming that's even possible) what would make carriers/manufacturers implement the fix when they already aren't updating the core version? Nothing. And they wouldn't. The carriers/manufacturers have financially abandoned these older models in favor or their new stuff.
They could deploy it to their own phones. Half of the Google-sold phone models are vulnerable to this bug.
People are used to a big brother company controlling everything about a software experience (Apple, Microsoft). The google approach is open. Unfortunately this requires the user to do a little bit of thinking, make an informed choice, and support the right companies with their money.
Which company would you buy an Android phone from to ensure that it received updates for the life of the contract, assuming your contract started on the last day the phone was available for sale?
What are the chances that a vendor that declines to update 4.3 to 4.4 would be willing to do an update for a 4.3.x if Google bothered to do it.
Considering that Google won't even do this for their pre-4.4 Nexus phones, I'd say that the chances are pretty low. The fact that Google still won't fix its own phones doesn't let it off the hook. They don't actually make ANY commitment to update Nexus devices at all, and have no documented end of life policy. They're basically not serious about security.
Apple and Microsoft control their own update process on all platforms; Google does not. It's the individual carriers who are getting in the way of Android updates.
They control the updates on the GSM Galaxy Nexus phone. It isn't getting the security patch.
My point was that it would not be microsoft's fault in this scenario, not that this scenario happened often. So maybe in the same way that people are not dumb enough to buy computers from comanies selling computers with windows XP in 2014, they should become smart enough not to buy phones with locked bootloaders (making them dependent on hardware vendors to get android updates).
So, people should be smart enough to not buy any phone that works on the Verizon network, any phone sold in an AT&T store as part of a contract, and any phone in a T-Mobile store sold under a purchase plan other than 1-2 models in the US?
You're basically saying that Android is great as long as you don't buy 99% of the devices on the market.
And even if you guy, eg, a Galaxy Nexus with an unlocked bootloader, the company that sold it to you (Google) only provided support for 1.5 years from the date the device FIRST went on sale. MS supports Windows for 10 years after the NEXT version of Windows goes on sale. That is why 95% of the PCs in businesses are STILL running Windows despite all the talk about the death of the desktop. I don't really have a problem with the death of the desktop, but businesses aren't going to buy into an alternative that isn't supported for a long time. They're fine with BYOD, since they're not the ones paying for support.
Agree. I use Android, but they could really benefit from something like this:
https://www.google.com/chrome/...
or
http://windows.microsoft.com/e...
or
https://access.redhat.com/supp...
or
http://www.ubuntu.com/info/rel...
The first link is Google's, so it isn't like they don't know how to do this stuff.
As for the costs, Cyanogen seems to prove that they can be pretty low. They support a lot of devices with very little funding to do so, partly because they are open source and rely on volunteers.
That, and their users don't seem to care if random small things break from time to time. Fortunately bluetooth stereo is much more mainstream now than it used to be, so the volunteer testers are far more likely to notice when it breaks. Back in the early days of CM it seemed like it only worked 70% of the time, but the average college student didn't use it so they didn't notice. That was before the M-series builds as well, so running "stable" meant just waiting for the next version of Android to come out so that you could use the last one, and it was basically abandonware.
Some companies pay them for support, which seems like a reasonable way to do long term updates.
As far as I've noticed, their paid firmware is just fine, probably because they actually give it serious QA, and of course it doesn't hurt that they have full access to the drivers/etc (which to be fair is a major handicap for their free efforts).
I also like that they have personally committed to updates for the phones they support, and they don't just say "it is up to your OEM." I get the impression that if a company like Oneplus folded that CM would still keep the OS updated for existing owners. Of course, they've yet to be tested on that.
Don't get me wrong, CM does great work. I just wouldn't say that they are without issue, or proof that the free software model works without any commercial ties. The areas where CM seems to go toe-to-toe with other OEMs in every regard is in the cases where they do have commercial ties.
Well, unlike the wireless phone companies, there where no vendors for the PCs that insist on putting their hands on the OS to customize the Android experience (mostly to detrimental effect, in my experience). So yes, Verizon, T-Mobile are on the hook for this one.
My plain vanilla Nexus 4 is still running fine with the latest and greatest, well latest, OS from Google. It is just staring to take some performance hits as compared to when it first came out.
Good thing you don't have a GSM Galaxy Nexus purchased directly from Google. I doubt they're patching those.
My impression is that hypotheses can very well begin with guesses, and once the guy with the guess can come up with some solid reasons for it it turns into a hypothesis.
More like once the guy with the guess can come up with an experiment that can demonstrate the falsehood (or lack thereof) of the guess it is a hypothesis.
Obviously if the guess is already inconsistent with observations then there is no need to run the experiment since it is already falsified.
It really doesn't.
A hypothesis has to make sense, has to be based on observation and/or our best current knowledge of the subject matter. Ideally it is testable somehow, even if only mathematically or theoretically.
A guess doesn't have to have any of those constraints. "Aliens did it" is a guess, but it's not a hypothesis.
Your statement should be embroidered, hung on the wall, and required reading before anyone is allowed to post on matters of science.
Way too many people, here and elsewhere, seem to have the idea that observation is somehow not a part of science.
You want to hang on a wall a statement that a hypothesis is "ideally" testable somehow, even if only mathematically (he did say "or"), and herald it for stressing the importance of observation?
Being testable against observations is an essential characteristic of a hypothesis. If it isn't testable against observations, it isn't a "non-ideal" hypothesis, it is pseudoscience. Sure, any hypothesis should be mathematically consistent if it relies on math, but that isn't sufficient to make it a hypothesis.
I'm fine with it being impractical to perform the experiment with current technology/resources - that is unfortunate but as long as the experiment exists I'll accept something as being a hypothesis. I certainly won't trust it as being correct though.
What's much more entertaining to me is that more often than not, the police reporting function isn't that valuable because the officer will have caught someone and moved on to a new spot by the time I see the notice.
Yes and no. Typically when I see a report I figure the police speed trap or whatever is somewhere within a few miles, which is useful information. Often the police just move up and down a stretch of highway alternating between prepared positions.
I'm saying if they have to backdoor specific firmware, there is still hope. Of course, since they have the capability to sap up nearly everyone's data, there isn't much hope to begin with.
Snowden revealed quite a bit in this space. The NSA has numerous departments and they cooperate.
You have the zero-day guys. They get lists of things that would be useful to hack, and they hack them. I'm sure that includes OSes, firmwares, peripherals, you name it. Some zero-days are held in reserve to avoid revealing them in case a high-priority target comes along.
You have the target intelligence guys. They identify systems to hack. They profile the targets - is this just a casual PC user, a company, or some government agency. They estimate how likely the target is to detect an intrusion - they don't want to use some super-secret zero-day on a guy who is ultra-paranoid and sends all their network traffic into a canary layered in 14 layers of firewalls and IDS.
You have the guys who run the wholesale hacking department. They pair up targets with zero-days and arrange to have them delivered, probably by redirecting their network traffic through a server that hands out the attack (too bad all your ad banners aren't protected by SSL, etc).
You have the rootkit guys who then take that initial foothold and exploit it, branching out into a network beyond the firewall and installing rootkits and monitoring software all over the place.
You have the intel guys who go in and harvest the information being sought.
Then you have the monitoring team. They make sure that all the compromised hosts stay compromised. Maybe you just installed some antivirus software that removed 3 out of the 14 rootkits they installed on your box, so they'll go ahead and put 5 more in and tell the CIA they need to be more generous with their symantec bribes.
All that division of labor means that they can break into vast numbers of computers very efficiently, with great expertise.
There was a beginning to the universe (which alone breaks the symmetry: you can't shift backwards in time more than ~13 billion years)...
Well, it might be better to say that we have no scientific knowledge of what came before the big bang, and as best as we can tell it is impossible to ever obtain knowledge of what came before.
It is convenient to call this a "beginning of time" or something like that, but this is a bit of a contrived definition.
But, as has been pointed out elsewhere in this thread our definitions of things like space and time are already pretty tenuous in general. We're very good at predicting the results of experiments, but we're not so good at really understanding why the universe actually works the way it seems to. The equations don't really provide much insight into what is actually happening.
Remember, this is the same NSA that intercepts Cisco shipments to install back-doored firmware and develops its own zero-day hacks for Windows.
The fact that they have to do this says a lot about their capabilities.
How would you propose hacking into a computer WITHOUT developing a zero-day for it? Well, unless you want to count using vulnerabilities from three years ago that some sysadmin is too lazy to patch. It isn't like anybody thinks the NSA has some psychic that just controls the minds of sysadmins from halfway around the globe. Engineering software and getting it to run on targeted hardware is just the physical reality of intruding on systems.
We're talking about wikileaks here. Obviously that is going to be a high-profile target for intelligence agencies anywhere. You simply can't run such an operation on some unencrypted webmail service ANYWHERE.
"The medium is the massage." -- Crazy Nigel
