Hm. Is that really that surprising? Many more mature languages have been standardized with multiple competing implementations, some proprietary, some free.
In fact, I think among popular languages, there are more of them with multiple competing implementations than there are with just one. Java, C, C++, Objective-C, C#, Fortran, JavaScript, Common Lisp, and Scheme all have multiple implementations, with their own implementations of the standard library.
The main exceptions are the collection of "scripting languages" which got popular in the late '90s and early 2000s like Perl, Python, Ruby, and PHP, which have mostly grown communities around a single implementation. But Ruby and Python at least (I'm not sure about the others, being less familiar with them) have been working towards multiple implementations and at least somewhat standardizing the core language and library to make multiple implementations more viable.
I'm not sure that's worrying. Remember, this is hardware. Hardware takes time to turn on and get into a steady state. Many hardware random number generators are based on thermal noise. But in order to get in the correct range for the values to flip, you need to wait for the circuit to warm up to operating temperature.
The impression I got was that the existence of a hardware random number generator was unexpected.
The rest was just demonstrating that it works, and produces pretty reliable random output. In order to do that, they used a standard tool that tests randomness. But it's nice to make sure your randomness test tool is actually good; so using it on a known-bad RNG, and seeing that it does fail, is a good way to demonstrate that.
Some art is ephemeral. Not all art needs to preserved for all time. Have you ever seen a Tibetan Buddhist sand mandala? The monks spend weeks creating an amazing detailed geometric piece of art, and then carefully destroy it, pouring the sand into a nearby river.
That said, I'm sure that the actual music will be released in a more portable/permanent format at some point.
This really isn't that much different from other album concepts that depend on a particular medium. The Who's Tommy is a rock opera; but more people know just the album. I recall various artists releasing albums on CD-ROM with interactive graphics and animation back in the mid '90s; and of course those were platform specific, for now obsolete platforms (though you may be able to run them in an emulator if you want to try playing them).
Enjoy the art now, in the context it was created for. Maybe it will exist in the future, maybe it won't. Don't worry about it too much.
> This really isn't that much different from other album concepts that depend on a particular medium. The Who's Tommy is a rock opera; but more people know just the album.
Just the album? Tommy is a rock opera album. Rock opera is a genre, or subgenre. How does a rock opera depend on a specific medium?
2.4 is really old, originally released in 2001. 2.6 has also been around for a long time, and the release cycle changed during the 2.6 timeframe, so the old release number system no longer makes sense.
Rather than having a major stable release like 2.4, and then a long period of development in 2.5, followed by a stable 2.6 kernel later on, it switched to much shorter merge windows and release candidate cycles that last about 2-3 months. So you had 2.6.x kernels which were new stable branches, and would get patches such as 2.6.x.y.
After a while, this got a little silly. When numbers were getting up to things like 2.6.32.60, it gets hard to keep track of the numbers. Since the first two numbers were just sitting there unused, at some point Linus decided it was silly, and they would cycle to 3, and the second number would be for each stable release.
So yeah, 2.2-2.4 was about 2 years between new stable versions. 2.4-2.6 was about 3 years between new stable versions. But then there have been new stable versions about once every 2-3 months, and new long term supported versions picked from those about once a year (for distributions which prefer to do long-term support on a single kernel version, rather than updating to the latest stable).
> After a while, this got a little silly. When numbers were getting up to things like 2.6.32.60, it gets hard to keep track of the numbers. Since the first two numbers were just sitting there unused, at some point Linus decided it was silly, and they would cycle to 3, and the second number would be for each stable release.
And now we have only one number sitting there unused! Maybe they'll finish the job next time.
One of the main threats that people have been worried about recently is global-scale passive attacks by the NSA. We know that they are gathering and retaining vast quantities of data; that they don't consider it illegal to simply gather and retain data without a warrant of any sort, and only actually look at it later once they have a warrant or have a 51% confidence that at least one of the parties in the communication is foreign.
For this threat, most active attacks are not particularly useful. They may capture encrypted data at a different point in transit than the one that originated it, making most side channel attacks difficult or impossible.
But a compromised RNG can be almost impossible to detect, and yet render almost any crypto system trivially breakable offline after the fact. The NSA has already once created a DRBG for which there could be a secret key that could determine the internal state of the generator with only 32 bytes of its output http://www.schneier.com/essay-198.html
I agree that being too wary of your hardware can lead you down a rabbit hole of paranoia, and that on the whole, we have much bigger problems to solve than hypothetical backdoored hardware random number generators (I have not seen any evidence that Intel has included a backdoor, merely people saying that because it's a black box we can't tell that they haven't). But since a good RNG is absolutely essential to so much crypto, and a compromised RNG would make it so easy for an adversary who passively monitors all communications and would like to have the option to decrypt them at their leave at a later date, I think that this is more worth worrying about than problems with the AES instructions.
> The Human Rights Watch representative used this opportunity to tell Mr Snowden that on her way to the airport she had received a call from the US Ambassador to Russia, who asked her to relay to Mr Snowden that the US Government does not categorise Mr Snowden as a whistleblower and that he has broken United States law.
Am I the only one who found this whole statement pretty odd?
Let's see. A "whistleblower" is a term for someone who blows a whistle. This is generally meant to indicate someone who is making some kind of loud noise to report a crime; someone (maybe a cop, maybe just someone who is a victim of a crime) trying to get attention that a crime has been committed.
Now, when reporting a crime, do you generally expect the perpetrator of that crime to be the appropriate person to determine whether or not a crime has been committed, and thus whether or not labeling the person a "whistleblower" is appropriate?
The whole idea of whistleblower laws that don't protect someone from reporting to the public about crimes committed by the administration is laughable. What the hell is the point of any kind of whistleblower laws or protection if you can't actually blow the whistle on your bosses?
But the whole idea of government is that one group of individuals should be the "top level" of authority, creating, enforcing, and interpreting its own laws, reserving the authority to keep certain types of secrets (and, of course, reserving the authority to determine which types of secrets qualify), etc.
Um, no. The whole idea of democracy is that there is no "top level" of authority.
In the US system, there are representatives of the people, passing laws. There is a single executive, also a representative of the people, implementing said laws. And there are the courts, interpreting said laws when there are disagreements over them.
But the whole point of representative democracy is that the people ultimately get to decide who is doing this, so that a runaway government cannot be a single "top level" of authority that creates, enforces, and interprets its own laws. The whole point is that the people are the highest authority, and the government serves at their will.
But if the government hides substantial information from the people, a democracy cannot effectively function. The people need to know what the government is doing. There are good reasons for limited amounts of secrecy, for particularly sensitive information or for privacy purposes. But keeping vast amounts of information on things that affect everyone, outside of a wartime environment, is anti-democratic.
This type of secrecy, this type of executive action which is seen as being unquestionable by even elected representatives, let alone the people, is turning the government from a representative democracy to an oligarchy.
> The whole idea of democracy is that there is no "top level" of authority.
Not true in any practical sense. If the government physically does something to me (like surveils me, arrests me, evicts me, etc.) I don't get to vote my way out of it. They are the top level of authority in any meaningful sense.
You appear to be confusing individual authority with collective authority. Of course you, personally, cannot get out of being arrested; if anyone could get out of being arrested by voting or saying that they are a member of the populace and thus are the ultimate authority, that would be pretty silly.
But rather, the people, collectively, are supposed to be the top level of authority. If the people want to vote out the current administration, they can. If they want to change the laws, they can vote for representatives who will do so. If they want to change the constitution, they can call a constitutional convention. Now sure, no individual can do this alone; they have to get a substantial fraction of the populace to do it along with them.
However, there are several factors that work against this. One major one is secrecy. If what the government is doing is hidden, then they cannot provide oversight (nor can their elected representatives, or the courts).
> Anything half the price will outperform what Bose has to offer.
So, if "anything half the price" is supposed to outperform Bose, would you be willing to point out something half the price of an equivalent Bose product and show me some kind of unbiased test that demonstrates it?
I feel like I hear this kind of sentiment a lot, but I almost never see any evidence to back it up.
Bose doesn't cater to the high-end audiophile market. They don't sell big, bulky, expensive equipment that audiophiles go for. Instead, they sell small stuff that's expensive for it's size but cheap compared to audiophile equipment. And I think that it it does pretty well for its size and price. Sure, you can get better sound quality out of bigger, bulkier, more expensive equipment. And you can get reasonable audio quality out of other small, cheaper gear. But I'm wondering if you can get gear that beats it in every way; smaller, cheaper, and better?
I have never done any kind of blinded experiments, or measurements of frequency responses, or other such scientific comparisons, so I can't say for sure. And that's why I'm asking if you know better. Is there something out there, at a similar size and price point, that is better than what Bose offers? Do you have good evidence for this?
At the constraints you provided (same size) the whole notion of "better" is extremely subjective. Small speakers can only reproduce a certain range of frequencies (think of it like trying to make 10 m high waves with a spatula) so they will all produce "colored" sound i.e. distort the signal. People perceive some coloration as being better than another and Bose's technology is all about making their coloration pleasant for most people. However, "most" is not "all" and some people don't like their coloration. Not to mention that any set of big speakers produces orders of magnitude less coloration.
Jellyfish won't be able to give you such a product because they don't exist. Your comment is 100% accurate. Bose is not the absolute best, but the absolute best is 10x or even 50x the price.
I would maybe agree that Bose has cultivated a reputation among ordinary consumers as "the best" which is not accurate. There are other equally good brands with a similar price point that don't get as much attention. There's also audiophile gear which is much better, but also much more expensive. What does not exist, though, are products superior to Bose as half the price. (Assuming you're not factoring in used gear or some kind of blowout sale) That goes for any brand, though.
I've never put on a pair of Bose so I can't attest to the sound quality, but I have held a few pairs and it bothers me that headphones in that price range can feel so fragile. There's this cheap plastic feel to it. My Sony MDR-V6 feels like a tank compared to them and these are plastic too.
I love my sony's over my boses. The sonys have survived 3 years of being sat on, run over by the chair, yanked on and off, cable tied around me and dragging my aeron across the room and out the door, etc. And you CAN get replacement parts for them. Heck they come with a diagram of how to take them apart and replace parts.
My bose QC's lasted 6 months of mild air travel and then the headband broke in half. Only way to fix em was buy a new pair. Also I could wear my QC's for about 4 hours and then my ear would hurt. I can wear the sonys for 12+ hours no issue. But that's just my freakish shaped head so it's personal.
That said I don't really give a crap about the sound quality. They sound good to me and they're a good price. My previous pair lasted 10 years.
All the audio engineers I've ever asked have a set of the sonys because they're supposed to be a good representation of a middle of the road system that everyone has. They're good but not amazing. But that's what you want when mixing. A representation of the middle and one of the high end. Most people will be listening on the crap to good stuff.
Hmm, interesting. My wife has a pair of the noise canceling headphones. They seem fairly sturdy to me.
Those Sonys are great phones designed to be tough. They're mean to be used in a studio, get sat on from time to time. Have the cables yanked, etc. They don't have noise canceling and they don't run on batteries, so I consider them a different piece of gear from the Bose noise canceling phones. But I have no doubt they're probably tougher than most phones, Bose included.
An audiophile tube amp is not going to take abuse either - it's meant to be treated in a delicate way. The tradeoff for that is that it will deliver pristine quality. The Bose stuff is not quite that fragile, not quite that good either. But the point is that all gear is designed for a certain purpose and comes with certain tradeoffs.
B&W MT Series, Kef Eggs, Orb Audio, Anthony Gallo, Focal Dome, the list is enormous, but most people don't get to hear (and see) them because they are not always sold in box shifting electronics stores and they don't have the Bose marketing budget. But all of them better designed and sound better.
Bose in-car and Bose noise cancelling headphones are a different matter, as their noise cancelling is first rate.
What are good sources for this kind of evidence? I'm actually interested; I find it hard to find good information on audio quality, because there's so much audiophile garbage out there.
One problem with moving away from Google is Google Maps. While OSM does have some fine maps, they don't have navigation or business search. I use Google Maps on my phone all the time to find "where's there a restaurant around here, and how can I get to it?"
Is there any privacy-preserving service that does anything similar?
OSM is data. Don't judge the data by the official website. It's power is in the freedom of the data and all the services/applications other people can build with it.
I tried OsmAnd. I needed to download a bunch of map data when I first installed it; but once it had downloaded some of the maps, I still couldn't view anything, it kept telling me that it was still downloading maps (I don't know why it couldn't display the ones it had already downloaded).
Once it had downloaded all of the maps and would show me something, I tried to get it to give me directions. I could not (and still cannot) figure out how to set a destination. I scroll to a place I want to go, touch the menu button, touch "show" or "follow", and get a message "please select a destination first". I long press on where I want to go, which pops up a bubble that shows my latitude and longitude (not, say, what businesses are there like Google Maps does), and try doing directions from the menu again. Nope, "please select a destination".
There is no search by address, or anything else. If I use Google Maps, I can usually type in the name of the place I'm trying to go, and it will find the address and from there find the route. Even being able to find the route from OsmAnd has eluded me.
Ah. Just now, as I tried it again, I found that I could long-press on a location, then touch the menu, touch "use location" (at the very bottom of the menu, almost off the screen, and which apparently means "use the location that I just long-pressed", not "use the location I'm at"), touch "destination", choose to either navigate or show the route, and then get told that it can't determine the route because the location is not available as I'm indoors without good GPS signal.
It needs some substantial work before it's actually usable. For instance, it needs a prominent search that will take the name of a business or an address and route me there with a minimal number of touches and decisions, and even if I'm indoors and don't have a good GPS signal.
I use mostly free software on the desktop, for my work. When I have time to sit down and debug something and figure out how it works, I'm willing to put up with a less than ideal interface; it's worth it that I have access to the source, that I can fix bugs when I find them, that I can get in touch with the original authors and pay them to fix a bug for me.
But when I'm trying to navigate to some new location on my phone on my vacation, I need something that just works. I don't have time to fiddle around with a lousy interface. I don't have time to look up the address of my destination elsewhere, then scroll around the map until I can find that address and go through a long press and several layers of menus before I can get directions to it.
What we really need is freedom preserving software that just works. It doesn't need to have every possible option and knob. It doesn't need to offer every feature under the sun. It just needs to make it easy for me to find where I'm going and tell me how to get there, without going through endless layers of menus and making me figure out where the destination is in the first place.
Agreed that OsmAnd's interface leaves much room for improvement. Even though I have successfully used its navigation function for specific addresses, I have also run into all the problems you outline. It does seem to get better with each new release.
The problem is that it's really hard to write unit tests to determine if you've gotten the crypto right. For one, the unit tests would have to have a standardized interface to your cryptographic protocol. But if you're using a standard cryptographic protocol, why would you be rolling your own implementation instead of using an existing, well tested and vetted one?
Second, it's quite difficult, in an automated way, to look for the kinds of security vulnerabilities you would need to audit. Timing attacks may show up in different places for different software. Poorly seeded entropy is quite hard to detect; Debian systems were generating keys with only about 16 bits of entropy for years, which is enough that it's hard to detect in an automated way unless you are really looking for that specific breakage but is a devastating security hole. You may have bugs that are not crypto related at all but simple buffer overflows or the like. And so on.
You could probably automate some of it. But on the whole, you need many good, experienced people regularly auditing the code. And it's a lot easier to get that by using an existing, widely used and widely audited library than by rolling your own.
In fact, I think among popular languages, there are more of them with multiple competing implementations than there are with just one. Java, C, C++, Objective-C, C#, Fortran, JavaScript, Common Lisp, and Scheme all have multiple implementations, with their own implementations of the standard library.
The main exceptions are the collection of "scripting languages" which got popular in the late '90s and early 2000s like Perl, Python, Ruby, and PHP, which have mostly grown communities around a single implementation. But Ruby and Python at least (I'm not sure about the others, being less familiar with them) have been working towards multiple implementations and at least somewhat standardizing the core language and library to make multiple implementations more viable.