As my nom de plume may be traceable by LEOs, I post as AC.
They know your IP; they see the traffic and they record it and they will have zero trouble identifying you if they decide they want to. There's a trail of breadcrumbs that leads directly to your computer for everything you do on the net. That's without the tricks of textual analysis they can apply to your known writing style.
Anonymous posting on the net at this point in time shields you only from your average citizen's inclination to take action. In no way does it shield you from the government if the government wants to know. And they don't have to want to know today. They can decide they want to know retroactively, and there you are -- bent over, pants down, no lube in sight. And they've got a pineapple.
Here's a bit, suitably edited, I wrote about this just the other day in the context of some fellow who rode in on the "just don't commit a crime and you'll be ok" horse:
The web of law is now so complex and deep that anyone is bound to make a miss-step fairly often or otherwise unexpectedly end up on the wrong side of the law and so turn out to be, regardless of intentionality (remember, ignorance of the law is no excuse), "a [darned] criminal."
From commonly ignored law (speeding, spitting, littering, jaywalking, "sharing" music or software, smoking somewhere you aren't allowed to, smoking pot, privately accepting or giving a prescription medication) to unintentional errors (walking by a partially unblocked window undressed, backing into someone else's vehicle, parking in a prohibited location) to intentional but with no knowledge of criminality (intentionally connecting to anyone's open wifi network without permission, collecting vertebrate fossils on public land, possessing too many bottles of NyQuil, having both bleach and ammonia under your sink, buying and storing too much of certain fertilizers) to being drawn in without intent (defending yourself, your family, your friend, defending your home, finding drugs or other prohibited materials in your home that you did put there)... and of course this is just the tip of the iceberg for all those categories and there are others such as a whole host of entrapment mechanisms, from chat room honeypots to speed traps and so on.
At this point, interaction with the authorities can arise; and that in itself is fraught with various types of risk, including the commission, or just the accusation, of further criminal activity. Resisting arrest, flight, striking a LEO, etc. These may not even be true accusations, but generally speaking, they might as well be because when it comes to difference in reporting between you and a LEO in court, or even for the benefit of a grand jury (which is run by the prosecutor already after you, btw), your word will weigh far less than theirs. You'll need something else, like a video (or many) and even then, you may not find yourself out of legal hot water.
Another problem is that both the feds and the states have begun implementing ex post facto laws. These are laws that make something illegal after the act, and/or increase the punishment for an act after one has been sentenced for it, and/or alter the rules of evidence from those in effect when the act occurred. So among other things, what you do today that is legal, may not only be illegal tomorrow, but could conceivably make you a criminal after the fact; likewise, the punishment they assign you today may mutate into something far more draconian tomorrow (this kind of thing is a matter of record at this point.)
All of these things, and an astonishingly large number of others that fall into the same categories, can create an "Ooops, I'm a criminal" facet of our lives -- even when our intentions are clearly and strongly -- other.
Act accordingly in this case means choosing whether or not to live in fear as a quisling cowards
Yes, that's one of the choices. Others include deciding whether putting your family and friends and those you love at risk for a fight you almost certainly cannot win is worth the candle; determining if the act is cowardice, or the best of a set of poor choices; dealing with the knowledge that if you manage to irritate those in power enough, and you fail to take them down, they may crush you -- without even noticing. You won't be a hero. You won't be a martyr. You won't be honored. You won't be feted. You'll just languish in their prisons, while your family and friends pay the very steep price of the choice you made. When you get out, your reputation and employment options will both be at zero, and the deck massively stacked against you repairing either one.
You really need to keep in mind that many people have tried to fight this particular war, and the result is that things are as bad, or worse, as they've ever been on most fronts. The battlefield, as it were, is littered with people killed and maimed, broken families, destroyed fortunes and crucified reputations. You really want to be clear-headed and have looked at every aspect before you go assigning labels like "quisling" to yourself, or others. And you really need to consider if it would be more prudent and effective to fight the small fight for your fellow citizen's minds instead of going head-on with a system that can, and will, eat you for breakfast.
I've tried to butt my head directly against the system. I did what I thought was right, and I insisted it was right, and I presented arguments that made my case and remain to this day undefeated by any sane counter. For my trouble, I learned, first hand, a great deal about the US prison system, the "justice" system, "rights", the ultimate fragility -- and importance -- of reputation, and in the end, found myself utterly checkmated in precisely the areas I felt needed attention the most. Odds are excellent that I'm as smart as, or smarter than, you. I'm almost certainly far wealthier than you, likewise so was (note the past tense) my family, and brother -- they squashed me like a bug. They're doing it right now to others. It won't inconvenience them in the slightest to do it to you.
This isn't about how you or your mom use explorer or finder or midnight commander to traverse the directory tree, reading file names and/or looking at icons.
This is about how someone actually trained to find stuff, finds stuff. And aside from outright encryption or sophisticated codes, you can't hide something from such people. Heck, you can't even hide the encrypted result; all you can do is prevent entry (and that, only until a judge says "you will open that door" because at that point, if you don't, the consequences are likely to be worse than if you had opened it in the first place. Like indefinite imprisonment until you do give it up.)
You make the assumption that religion is outside the "natural world". That's not very scientific of you.
LOL. If it's not outside the natural world, then it has every characteristic of the most dishonest bunkum, and no characteristics of something -- anything -- to do with objective reality. In other words, if you remove the "disjoint magisteria" claim from the assessment of religion, you don't have anything left worth a plugged nickle.
Which is not to say you have much with the "disjoint magisteria" argument; but at least you have something.
The whole argument boils down to "there's no scientific proof of religion because science has no access to religion, and that's the way God wants it." As soon as you assert science does have access to religion... game over, because now you require consensually experiential, repeatable evidence to back your assertion -- and no one's been able to meet that standard since day one. Not that it wouldn't be super if you could do it; but all of human experience lands on the side of the scale that says you won't.
The thing is, the Echo provides a lot already, in a great form-factor and with at least decent fidelity. To the point that it has earned a welcome place in our household; and it gets more use every day as the family gets accustomed to the idea that it's always there to talk to. To-do lists, shopping lists, weather, timers, alarms, time, music (listening, ordering), math, biographical data, cooking tips, even spelling and word definitions. They've really done a yeoman's job of pulling together a huge amount of functionality in a small, decent sounding form factor -- and the audio discrimination and lack of output directionality is outstanding. It's very good hardware.
The up-front problems that the Echo has decent solutions for already include hearing you (and understanding you!) from just about anywhere in a very large potential space, under conditions ranging from quiet to really pretty noisy; decent sounding omnidirectional music playback that does not prevent it from hearing you continue to interact with it; a non-ugly put-it-anywhere form factor so it integrates well with almost any conceivable decor; low power consumption; low cost; network connectivity; TTS; STT; and the above broad categories of useful interactivity.
To duplicate -- never mind expand -- the Echo's base functionality using a more general purpose platform would be a great deal of work -- there's little motivation to do it, as it's already been done, and inexpensively, too (our Echo cost us $100 as "Amazon Prime" members.) What I was talking about, in context of the/. post, is opening an application space where we could expand the Echo's already broad feature set.
For instance, you want Echo to know about minerals? I already have an expert system that can be queried in natural language. Colors, habitat, crystal form(s), collecting sites, Moh's hardness, specific gravity, chemical composition, weathering behavior, common pseudomorphs, uses, and so on. You can ask it about a fair number of geological issues, classes of minerals, specific minerals, or you can give it characteristics and it'll identify a list of candidates for you, along with suggestions for differentiating among them. Plugging that into Echo -- with the interface I suggested -- could be done in basically no significant amount of time.
I also have expert systems on domestic cats; one on amateur radio issues; one on keeping a reef-centric salt tank; and one on Python. They are each basically 99.9% of an Echo app/expansion waiting to happen -- all that's missing is a very tiny API on the part of the Echo, and a similarly tiny -- and uniform -- wrapper around the expert systems.
Your idea is fine for a stand-alone voice in and out wrapper, as would be any other TTS/STT solution (OS X has had TTS built right in for some time, and as of 10.7 has built-in STT as well.) But I just don't see that it has the built-in attraction the Echo does. Not that I am discouraging you from following your idea, of course.:)
Amazon's Echo -- Echo, for anyone who doesn't know, is a device that sits around and waits for you to ask it, or tell it, things. Like "set a timer for 22 minutes" or "what's the weather" or "what is 22 divided by 7" or "Play some classical music" and a whole host of similarly useful things. We have one and are most impressed with it -- as a first step.
The way it works is you wake it up with its name, then ask, it sends the audio to servers on the net, which speech-to-text it, and then they figure out how to answer you, send it back, and there you go. Which means that the questions and answers it can handle will be limited to generally known or knowable data of only a somewhat local nature (e.g. the current weather.) So "Alexa, what's the weather?" gets you "In Podunk, it's -12 degrees with mostly sunny skies."
So I wrote them and suggested an API over the local wifi network (it's on wifi already in every installation) where a local computer could hook in, and either pre-or post Amazon's evaluation after the speech-to-text, take a swing at answering the question or obeying the command.
So you could say, "Dad will be at the gym from 5pm till 6 pm today", Amazon would evaluate that as WTF, it'd get passed to the local system, which would store the info and tell the Echo that the input was handled, and pass back a confirmation like "Got it" or "Ok, amending dad's schedule today to add the gym from 5 to 6 pm." Later, little Seymour could ask (between four and five) "Where's daddy?" and the app could respond with "Probably at the gym until 6pm", and Echo would then text-to-speech the answer and hand it over to little Seymour.
Likewise, local in-house temperature, alarm status, variously local DB driven inquiries (inventories, expert systems, etc.)
All Amazon really needs for this is an SSH port hooked to the result of WTF (currently WTF gets you "I'm sorry, I didn't understand the question")
Then we could build our own Echo application ecosystem.
I already have extensive code that parses questions in text form and appropriately queries an expert DB, returning variously deep answers in text. Given the hooks as described, I think I could have Echo answering custom queries and taking custom data in under an hour, most of which would be spent reading the new docs to figure out talking to the Echo itself.:)
I got a nice letter back thanking me for the idea. Here's hoping.
Experiments Create New Particles From Atom and Neutrino Interaction
...the implication that the particle arose from vacuum is what I read; but it arose consequent to an energy to matter conversion, it seems to me. Ol' Albert had something to say about this, IIRC.
Especially remote working across 8 time zones (i.e. you only actually get to chat to each other for 1-2 hours a day)
This is just the kind of thinking I was speaking of. It makes no sense whatsoever when you actually consider it, yet people throw it up as if it was a real thing you had to worry about.
If someone hired me and was X time zones away, I would simply adjust my hours by X, or X +/- Y as they considered useful to them. This is no different than, for instance, having to work the night shift (or any other fixed shift, or a swing shift.)
This is a factor that should be of absolutely no relevance unless the prospective employee actually refuses to work the required hours, in which case, I guess they didn't want the job anyway, and you're better off moving right along.
I think programmers are just too lazy to really use the hardware
Not everyone. I write in C -- large applications, too -- and I write as close to the metal as I can get. I don't mind assembler, but the processors in use move under our feet too often: there's just no practical way to keep up without a compiler in between my code and the actual CPU instructions.
For example the smart thing to do would be to make sure that the user interface is never swapped to disk. That would reduce available RAM only slightly but would dramatically improve performance.
Agreed, that sounds like it'd be worthy. The problem I would anticipate is that a lot of the OS/UI code may be contained in huge "black boxes" that, if all loaded all the time, would consume much more RAM than we might otherwise think would be needed. OTOH, maybe we should all have 32 GB of RAM like you do. It sure has gotten inexpensive. On the OTHER other hand, if we did, the bloody OS would probably balloon to 32 GB, so... lol
In real life, my 6-core, 32 GB-RAM box swaps even the tiniest process to disk (which is of course SSD) so that even opening the KDE-menu takes ages after some time.
Concur with davydagger. You're either doing something really resource-intensive you didn't mention, or your OS is configured wrong.
If you have 32 GB of ram, unless you're running software that makes demands on that scale, you probably don't need swap at all. I've only got 8 GB of ram and my system does really well unless I actually use it up -- although mine's OS X, so your swap algorithms and so forth are different. Still, I'm almost certain you can set the box up to behave better.
In the past, I know linux had a really annoying bias for using up all the ram with buffers and cache, and would pig out if you actually tried to use that ram yourself once all the RAM was used that way, despite the supposed ability to throw out the cache and the buffers if the RAM was needed, but I am under the impression that time has passed.
davydagger offered some specifics there... sounds like the right place to at least start reading some man pages.:)
Isn't the Windows 8 framework - user interface based on CSS and Javascript already?
While we're not all listening in anechoic chambers, studies show narrowband signals are detectable as much as 20 dB below a broadband noise floor
Detectability is not an adequate metric for musical reproduction, narrow-band or otherwise. For that, we must operate above the noise floor.
[amps not reaching the ear's dynamic range] means that the suggestion that solid state and tube amps that are properly designed sound the same is incorrect.
No, it doesn't mean anything of the kind. All it means is that our ear has more dynamic range. It doesn't mean it has more resolution within any range segment (and in fact, it doesn't -- we kind of suck at absolute amplitude discrimination.) Anyway, regardless of our ear's dynamic range, all we will hear is the dynamic range the audio system presents. And the most that will be with a CD is about 90 dB.
There is no "linear range"; only an approximately linear one, which is still not completely linear with respect to psychoacoustics.
It's linear in human terms. It's not linear in absolute terms. Other than that, no -- given the same tone contours, filter orders and Q, or the lack of any, as long as the amps are running in the range they were designed to, and the input and output devices are identical, and match the amplifier's output impedances, the audio will sound the same within the limits of the caveats I already provided, primarily damping factor.
As the subject line hopefully intimated, I consider this horse beaten to death and condemned as unfit for consumption before I even posted. Therefore, the last word is something you are most welcome to.:)
Nothing is perfectly linear in this regime. However, and as should have been obvious, the question, and the answer, resides in the realm of "linear enough." What I said is exactly correct as far as it applies to humans listening to audio. And that's all I meant to address. Within that scope, my presentation was not only accurate, it was conservative.
I will now leave you to argue the merits of your case as you understand them; however, I won't be picking up the argument. I am only interested in the facts, and I have already presented those, so I'm done. Cheers!
If speakers are so awful, what's the point in preferring vinyl for "audio quality" reasons?
First of all, speaker issues almost uniformly impact the music in quite different ways than vinyl does. So this isn't really a "thing" as you present it.
More generally, speakers are "what we have", or at least, most of us (there are some cool things in the labs, and these days, the term "speaker" covers a lot more ground than "moving coil transducer" when you start spending uncomfortable amount of money.) So almost everyone has to put up with them, at least at this point. Not true for vinyl at all.
The irredeemable issues that speakers do have that we run into audibly are in the harmonic distortion class of problems; the character and amount of these distortions vary enormously with about every factor that relates to speakers at all: Environment, placement angle, distance and height, frequency response, number of drivers, amp damping factor, crossover technology and frequency and order and nearness to each driver's 3 dB down point, cabinet and baffle and damping and porting and directionality and phase linearity and... well, you get the idea.
Some speakers also don't get down into the lower end very well, or become highly directional at higher frequencies. For instance, I have a pair of JBL-L100's, completely restored, that you'd think sounded great... until I A:B them for you with my Marantz HD-880's and suddenly you realize that the JBL's were completely losing a lot of the low and high end program material.
The end result is that our choice of speakers will sonically "color", that is, audibly alter, the music we hear. That's generally a bad thing (unless you're selling speakers and you can convince your audience that your color is better than the other guy's color or lie to them and claim you have none) but it is a constant thing and because we all have various tastes -- as amply demonstrated by the up-front differences in settings of simple things like bass and treble and loudness and volume and so forth -- we will generally prefer one set of speakers over another when they are fairly compared. Once you pick and install, that's the color you're going to get. (which, because of your installation environment, probably isn't the color you heard in the showroom.... argh... this is where consultants can make big bucks.)
Something else many people don't realize is that with a turntable, moving-coil and moving-lever pickups have exactly the same kind of problems for many of the same reasons, and produce their own THD, and therefore audible color. Changing a pickup -- not just the stylus -- can change your whole perception of a familiar recording. All kinds of ways. Transient response, frequency response, THD, stereo separation... it can be profound. And that adds to what the speaker is going to do on the output end of all this.
The recording techniques used on older vinyl were generally either very light-handed with, or completely lacking in, intentional compression. Some recordings -- the aforementioned Beatles recordings are a good example -- were unintentionally compressed by virtue of being recorded on tape just a bit too "hot", which causes the tape to gently waltz out of the linear recording zone and slowly begin to decrease the changes in amplitude that exceed the linear zone threshold. Basically, light compression, and entirely a good thing or the music would have been badly damaged. This lack of heavy compression can make such a huge difference in how a performance "gets into your ear" as to be perceived as almost entirely another bit of music as compared to music compressed for playback on top-40 (and many other) stations these days*. And when a recording is only available on vinyl, or only available uncompressed on vinyl, it won't be the system coloration (call it stylus+speaker distortion) that you notice, believe me. What you'll notice is it can actually sound like you are there. That is absol
Yeah, can't help you there. Was just trying to head off the audiophile tube argument at the pass.
What you're dealing with, I think, would be fairly characterized as things sounding like you're accustomed to hearing them sound, which pleases you, and there isn't squat I can say about that unless your actual preferences change. I can't do much about that remotely, much as I would like to try.
I will tell you this, though: Early Beatles recordings, suitably remastered, played back on my system, sometimes leave me in tears. Some of which is from the sheer pleasure of it, some of which is frustration that we've gone so far down the compress-it rabbit hole, and some of which is purest, reeking nostalgia, I have to admit.
Slashdot Mirror
They know your IP; they see the traffic and they record it and they will have zero trouble identifying you if they decide they want to. There's a trail of breadcrumbs that leads directly to your computer for everything you do on the net. That's without the tricks of textual analysis they can apply to your known writing style.
Anonymous posting on the net at this point in time shields you only from your average citizen's inclination to take action. In no way does it shield you from the government if the government wants to know. And they don't have to want to know today. They can decide they want to know retroactively, and there you are -- bent over, pants down, no lube in sight. And they've got a pineapple.
Here's a bit, suitably edited, I wrote about this just the other day in the context of some fellow who rode in on the "just don't commit a crime and you'll be ok" horse:
The web of law is now so complex and deep that anyone is bound to make a miss-step fairly often or otherwise unexpectedly end up on the wrong side of the law and so turn out to be, regardless of intentionality (remember, ignorance of the law is no excuse), "a [darned] criminal."
From commonly ignored law (speeding, spitting, littering, jaywalking, "sharing" music or software, smoking somewhere you aren't allowed to, smoking pot, privately accepting or giving a prescription medication) to unintentional errors (walking by a partially unblocked window undressed, backing into someone else's vehicle, parking in a prohibited location) to intentional but with no knowledge of criminality (intentionally connecting to anyone's open wifi network without permission, collecting vertebrate fossils on public land, possessing too many bottles of NyQuil, having both bleach and ammonia under your sink, buying and storing too much of certain fertilizers) to being drawn in without intent (defending yourself, your family, your friend, defending your home, finding drugs or other prohibited materials in your home that you did put there)... and of course this is just the tip of the iceberg for all those categories and there are others such as a whole host of entrapment mechanisms, from chat room honeypots to speed traps and so on.
At this point, interaction with the authorities can arise; and that in itself is fraught with various types of risk, including the commission, or just the accusation, of further criminal activity. Resisting arrest, flight, striking a LEO, etc. These may not even be true accusations, but generally speaking, they might as well be because when it comes to difference in reporting between you and a LEO in court, or even for the benefit of a grand jury (which is run by the prosecutor already after you, btw), your word will weigh far less than theirs. You'll need something else, like a video (or many) and even then, you may not find yourself out of legal hot water.
Another problem is that both the feds and the states have begun implementing ex post facto laws. These are laws that make something illegal after the act, and/or increase the punishment for an act after one has been sentenced for it, and/or alter the rules of evidence from those in effect when the act occurred. So among other things, what you do today that is legal, may not only be illegal tomorrow, but could conceivably make you a criminal after the fact; likewise, the punishment they assign you today may mutate into something far more draconian tomorrow (this kind of thing is a matter of record at this point.)
All of these things, and an astonishingly large number of others that fall into the same categories, can create an "Ooops, I'm a criminal" facet of our lives -- even when our intentions are clearly and strongly -- other.
Yes, that's one of the choices. Others include deciding whether putting your family and friends and those you love at risk for a fight you almost certainly cannot win is worth the candle; determining if the act is cowardice, or the best of a set of poor choices; dealing with the knowledge that if you manage to irritate those in power enough, and you fail to take them down, they may crush you -- without even noticing. You won't be a hero. You won't be a martyr. You won't be honored. You won't be feted. You'll just languish in their prisons, while your family and friends pay the very steep price of the choice you made. When you get out, your reputation and employment options will both be at zero, and the deck massively stacked against you repairing either one.
You really need to keep in mind that many people have tried to fight this particular war, and the result is that things are as bad, or worse, as they've ever been on most fronts. The battlefield, as it were, is littered with people killed and maimed, broken families, destroyed fortunes and crucified reputations. You really want to be clear-headed and have looked at every aspect before you go assigning labels like "quisling" to yourself, or others. And you really need to consider if it would be more prudent and effective to fight the small fight for your fellow citizen's minds instead of going head-on with a system that can, and will, eat you for breakfast.
I've tried to butt my head directly against the system. I did what I thought was right, and I insisted it was right, and I presented arguments that made my case and remain to this day undefeated by any sane counter. For my trouble, I learned, first hand, a great deal about the US prison system, the "justice" system, "rights", the ultimate fragility -- and importance -- of reputation, and in the end, found myself utterly checkmated in precisely the areas I felt needed attention the most. Odds are excellent that I'm as smart as, or smarter than, you. I'm almost certainly far wealthier than you, likewise so was (note the past tense) my family, and brother -- they squashed me like a bug. They're doing it right now to others. It won't inconvenience them in the slightest to do it to you.
You said:
The idea of rights is a lovely philosophical butterfly.
The actuality of rights is that they are only rules that those in power agree to enforce and are able to enforce.
The direct consequences of these facts are obvious. Act accordingly.
Orwell was an optimist.
This isn't about how you or your mom use explorer or finder or midnight commander to traverse the directory tree, reading file names and/or looking at icons.
This is about how someone actually trained to find stuff, finds stuff. And aside from outright encryption or sophisticated codes, you can't hide something from such people. Heck, you can't even hide the encrypted result; all you can do is prevent entry (and that, only until a judge says "you will open that door" because at that point, if you don't, the consequences are likely to be worse than if you had opened it in the first place. Like indefinite imprisonment until you do give it up.)
LOL. If it's not outside the natural world, then it has every characteristic of the most dishonest bunkum, and no characteristics of something -- anything -- to do with objective reality. In other words, if you remove the "disjoint magisteria" claim from the assessment of religion, you don't have anything left worth a plugged nickle.
Which is not to say you have much with the "disjoint magisteria" argument; but at least you have something.
The whole argument boils down to "there's no scientific proof of religion because science has no access to religion, and that's the way God wants it." As soon as you assert science does have access to religion... game over, because now you require consensually experiential, repeatable evidence to back your assertion -- and no one's been able to meet that standard since day one. Not that it wouldn't be super if you could do it; but all of human experience lands on the side of the scale that says you won't.
Um. Well, Apples and oranges, really.
The thing is, the Echo provides a lot already, in a great form-factor and with at least decent fidelity. To the point that it has earned a welcome place in our household; and it gets more use every day as the family gets accustomed to the idea that it's always there to talk to. To-do lists, shopping lists, weather, timers, alarms, time, music (listening, ordering), math, biographical data, cooking tips, even spelling and word definitions. They've really done a yeoman's job of pulling together a huge amount of functionality in a small, decent sounding form factor -- and the audio discrimination and lack of output directionality is outstanding. It's very good hardware.
The up-front problems that the Echo has decent solutions for already include hearing you (and understanding you!) from just about anywhere in a very large potential space, under conditions ranging from quiet to really pretty noisy; decent sounding omnidirectional music playback that does not prevent it from hearing you continue to interact with it; a non-ugly put-it-anywhere form factor so it integrates well with almost any conceivable decor; low power consumption; low cost; network connectivity; TTS; STT; and the above broad categories of useful interactivity.
To duplicate -- never mind expand -- the Echo's base functionality using a more general purpose platform would be a great deal of work -- there's little motivation to do it, as it's already been done, and inexpensively, too (our Echo cost us $100 as "Amazon Prime" members.) What I was talking about, in context of the /. post, is opening an application space where we could expand the Echo's already broad feature set.
For instance, you want Echo to know about minerals? I already have an expert system that can be queried in natural language. Colors, habitat, crystal form(s), collecting sites, Moh's hardness, specific gravity, chemical composition, weathering behavior, common pseudomorphs, uses, and so on. You can ask it about a fair number of geological issues, classes of minerals, specific minerals, or you can give it characteristics and it'll identify a list of candidates for you, along with suggestions for differentiating among them. Plugging that into Echo -- with the interface I suggested -- could be done in basically no significant amount of time.
I also have expert systems on domestic cats; one on amateur radio issues; one on keeping a reef-centric salt tank; and one on Python. They are each basically 99.9% of an Echo app/expansion waiting to happen -- all that's missing is a very tiny API on the part of the Echo, and a similarly tiny -- and uniform -- wrapper around the expert systems.
Your idea is fine for a stand-alone voice in and out wrapper, as would be any other TTS/STT solution (OS X has had TTS built right in for some time, and as of 10.7 has built-in STT as well.) But I just don't see that it has the built-in attraction the Echo does. Not that I am discouraging you from following your idea, of course. :)
Amazon's Echo -- Echo, for anyone who doesn't know, is a device that sits around and waits for you to ask it, or tell it, things. Like "set a timer for 22 minutes" or "what's the weather" or "what is 22 divided by 7" or "Play some classical music" and a whole host of similarly useful things. We have one and are most impressed with it -- as a first step.
The way it works is you wake it up with its name, then ask, it sends the audio to servers on the net, which speech-to-text it, and then they figure out how to answer you, send it back, and there you go. Which means that the questions and answers it can handle will be limited to generally known or knowable data of only a somewhat local nature (e.g. the current weather.) So "Alexa, what's the weather?" gets you "In Podunk, it's -12 degrees with mostly sunny skies."
So I wrote them and suggested an API over the local wifi network (it's on wifi already in every installation) where a local computer could hook in, and either pre-or post Amazon's evaluation after the speech-to-text, take a swing at answering the question or obeying the command.
So you could say, "Dad will be at the gym from 5pm till 6 pm today", Amazon would evaluate that as WTF, it'd get passed to the local system, which would store the info and tell the Echo that the input was handled, and pass back a confirmation like "Got it" or "Ok, amending dad's schedule today to add the gym from 5 to 6 pm." Later, little Seymour could ask (between four and five) "Where's daddy?" and the app could respond with "Probably at the gym until 6pm", and Echo would then text-to-speech the answer and hand it over to little Seymour.
Likewise, local in-house temperature, alarm status, variously local DB driven inquiries (inventories, expert systems, etc.)
All Amazon really needs for this is an SSH port hooked to the result of WTF (currently WTF gets you "I'm sorry, I didn't understand the question")
Then we could build our own Echo application ecosystem.
I already have extensive code that parses questions in text form and appropriately queries an expert DB, returning variously deep answers in text. Given the hooks as described, I think I could have Echo answering custom queries and taking custom data in under an hour, most of which would be spent reading the new docs to figure out talking to the Echo itself. :)
I got a nice letter back thanking me for the idea. Here's hoping.
Stop right there.
Why should I be interested in the rhythms of cicadas?
Sorry.
I'm an Android hipster, you insensitive clod!
Sure there is. You just didn't set the app up correctly.
Toto's on his way to be tutored. They caught him playing with unleashed atoms, now they have to figure out who's going to take all the little pions.
Pretty sure the title of TFS should have been:
This is just the kind of thinking I was speaking of. It makes no sense whatsoever when you actually consider it, yet people throw it up as if it was a real thing you had to worry about.
If someone hired me and was X time zones away, I would simply adjust my hours by X, or X +/- Y as they considered useful to them. This is no different than, for instance, having to work the night shift (or any other fixed shift, or a swing shift.)
This is a factor that should be of absolutely no relevance unless the prospective employee actually refuses to work the required hours, in which case, I guess they didn't want the job anyway, and you're better off moving right along.
Also, stop being anal about degrees, credit scores, old convictions, age, and health.
There's no programmer shortage. That's utter BS.
There's just a hiring pathology.
Not everyone. I write in C -- large applications, too -- and I write as close to the metal as I can get. I don't mind assembler, but the processors in use move under our feet too often: there's just no practical way to keep up without a compiler in between my code and the actual CPU instructions.
Agreed, that sounds like it'd be worthy. The problem I would anticipate is that a lot of the OS/UI code may be contained in huge "black boxes" that, if all loaded all the time, would consume much more RAM than we might otherwise think would be needed. OTOH, maybe we should all have 32 GB of RAM like you do. It sure has gotten inexpensive. On the OTHER other hand, if we did, the bloody OS would probably balloon to 32 GB, so... lol
Concur with davydagger. You're either doing something really resource-intensive you didn't mention, or your OS is configured wrong.
If you have 32 GB of ram, unless you're running software that makes demands on that scale, you probably don't need swap at all. I've only got 8 GB of ram and my system does really well unless I actually use it up -- although mine's OS X, so your swap algorithms and so forth are different. Still, I'm almost certain you can set the box up to behave better.
In the past, I know linux had a really annoying bias for using up all the ram with buffers and cache, and would pig out if you actually tried to use that ram yourself once all the RAM was used that way, despite the supposed ability to throw out the cache and the buffers if the RAM was needed, but I am under the impression that time has passed.
davydagger offered some specifics there... sounds like the right place to at least start reading some man pages. :)
No idea. Microsoft is dead to me. :)
Detectability is not an adequate metric for musical reproduction, narrow-band or otherwise. For that, we must operate above the noise floor.
No, it doesn't mean anything of the kind. All it means is that our ear has more dynamic range. It doesn't mean it has more resolution within any range segment (and in fact, it doesn't -- we kind of suck at absolute amplitude discrimination.) Anyway, regardless of our ear's dynamic range, all we will hear is the dynamic range the audio system presents. And the most that will be with a CD is about 90 dB.
It's linear in human terms. It's not linear in absolute terms. Other than that, no -- given the same tone contours, filter orders and Q, or the lack of any, as long as the amps are running in the range they were designed to, and the input and output devices are identical, and match the amplifier's output impedances, the audio will sound the same within the limits of the caveats I already provided, primarily damping factor.
As the subject line hopefully intimated, I consider this horse beaten to death and condemned as unfit for consumption before I even posted. Therefore, the last word is something you are most welcome to. :)
Nothing is perfectly linear in this regime. However, and as should have been obvious, the question, and the answer, resides in the realm of "linear enough." What I said is exactly correct as far as it applies to humans listening to audio. And that's all I meant to address. Within that scope, my presentation was not only accurate, it was conservative.
I will now leave you to argue the merits of your case as you understand them; however, I won't be picking up the argument. I am only interested in the facts, and I have already presented those, so I'm done. Cheers!
Your investment in mythology in way alters the facts. Sorry to have to break it to you.
Better range for your speakers. Guaranteed.
First of all, speaker issues almost uniformly impact the music in quite different ways than vinyl does. So this isn't really a "thing" as you present it.
More generally, speakers are "what we have", or at least, most of us (there are some cool things in the labs, and these days, the term "speaker" covers a lot more ground than "moving coil transducer" when you start spending uncomfortable amount of money.) So almost everyone has to put up with them, at least at this point. Not true for vinyl at all.
The irredeemable issues that speakers do have that we run into audibly are in the harmonic distortion class of problems; the character and amount of these distortions vary enormously with about every factor that relates to speakers at all: Environment, placement angle, distance and height, frequency response, number of drivers, amp damping factor, crossover technology and frequency and order and nearness to each driver's 3 dB down point, cabinet and baffle and damping and porting and directionality and phase linearity and... well, you get the idea.
Some speakers also don't get down into the lower end very well, or become highly directional at higher frequencies. For instance, I have a pair of JBL-L100's, completely restored, that you'd think sounded great... until I A:B them for you with my Marantz HD-880's and suddenly you realize that the JBL's were completely losing a lot of the low and high end program material.
The end result is that our choice of speakers will sonically "color", that is, audibly alter, the music we hear. That's generally a bad thing (unless you're selling speakers and you can convince your audience that your color is better than the other guy's color or lie to them and claim you have none) but it is a constant thing and because we all have various tastes -- as amply demonstrated by the up-front differences in settings of simple things like bass and treble and loudness and volume and so forth -- we will generally prefer one set of speakers over another when they are fairly compared. Once you pick and install, that's the color you're going to get. (which, because of your installation environment, probably isn't the color you heard in the showroom.... argh... this is where consultants can make big bucks.)
Something else many people don't realize is that with a turntable, moving-coil and moving-lever pickups have exactly the same kind of problems for many of the same reasons, and produce their own THD, and therefore audible color. Changing a pickup -- not just the stylus -- can change your whole perception of a familiar recording. All kinds of ways. Transient response, frequency response, THD, stereo separation... it can be profound. And that adds to what the speaker is going to do on the output end of all this.
The recording techniques used on older vinyl were generally either very light-handed with, or completely lacking in, intentional compression. Some recordings -- the aforementioned Beatles recordings are a good example -- were unintentionally compressed by virtue of being recorded on tape just a bit too "hot", which causes the tape to gently waltz out of the linear recording zone and slowly begin to decrease the changes in amplitude that exceed the linear zone threshold. Basically, light compression, and entirely a good thing or the music would have been badly damaged. This lack of heavy compression can make such a huge difference in how a performance "gets into your ear" as to be perceived as almost entirely another bit of music as compared to music compressed for playback on top-40 (and many other) stations these days*. And when a recording is only available on vinyl, or only available uncompressed on vinyl, it won't be the system coloration (call it stylus+speaker distortion) that you notice, believe me. What you'll notice is it can actually sound like you are there. That is absol
Yeah, can't help you there. Was just trying to head off the audiophile tube argument at the pass.
What you're dealing with, I think, would be fairly characterized as things sounding like you're accustomed to hearing them sound, which pleases you, and there isn't squat I can say about that unless your actual preferences change. I can't do much about that remotely, much as I would like to try.
I will tell you this, though: Early Beatles recordings, suitably remastered, played back on my system, sometimes leave me in tears. Some of which is from the sheer pleasure of it, some of which is frustration that we've gone so far down the compress-it rabbit hole, and some of which is purest, reeking nostalgia, I have to admit.