That is the usual price of a pay-as-you-go movie {though you will need a picture stabiliser if you want to record it -- they stick Macrovision in the signal because if you chose when you wanted to watch it then you don't need to be able to record it, do you?}; and that is by some bizarre coincidence the same amount as I would pay to download one legally: Three pounds fifty.
* A point approximately one-third of the way into the story wherein, all characters having been established sufficiently in the mind of the reader for the purposes of the rest of the story, the action begins to speed up.
* The use of deliberately misleading language, in a situation where there are more than one possible outcome, so as to cause the reader to assume a particular one; and the subsequent revelation to the reader that their selection is in fact incorrect.
* {Most commonly encountered in a children's story} The tearful parting of the Hero's Ally, never to return again.
Even if it's using ultrasound there will be a voice coil with a capturable magnetic field. As I understand it, it's using amplitude modulation {similar to MW and LW radio} but at tens as opposed to hundreds of kilohertz, and can be demodulated by simple rectification {which occurs naturally when a pressure wave is travelling through a fluid [air] and the receiving device [ear] is moving}.
The prototypes I have seen using this method had a very noticeable "scratchy" quality to them.
There are two basic kinds of A-D converter. TYPE ONE, the slow type, works by "guessing" and consists of a D-A converter feeding into one input of a comparator, which indicates whether the voltage from the D-A is greater or less than the voltage at the other input. A suitable external circuit is used to generate "guesses", and we can determine the two points between which the comparator changes state. This type of converter is slow since it takes at least as many guesses as bits in the output, but uses fewer components.
TYPE TWO, the fast type {sometimes called a "flash converter"} uses as many comparators as there are states in the digital output -- i.e. 2 ** [the number of bits]. The input signal is compared simultaneously to voltages corresponding to each step. A logic array is used to determine where in the chain the comparator outputs change state. There is no requirement for successive "guesses", the only time constraint being due to the propagation delay in the decision logic. However, for large numbers of bits, flash converters become unwieldy.
There is also a "half-flash" converter. This uses two flash A-to-Ds, a D-to-A and an op-amp. The first A-D digitises the signal to half the desired resolution, giving the higher-order bits of the result. The D-A converts this back to analogue; but there will be a difference, since the conversion is not exact. The op-amp amplifies this "error" signal and the second A-D digitises it, giving the lower-order bits of the result. Think of it as a number with a fraction: the first converter deals with the digits before the point and the second converter deals with the digits after the point. Half-flash converters depend more than most on circuit quality.
Any kind of analogue-to-digital converter requires a well-regulated power supply, tight tolerance resistances and above all, low noise components. All connections must be sound - dry solder joints and loose or dirty mechanical contacts introduce noise. However, it is just about feasible to construct an 8-bit, half-flash converter on copper strip breadboard. For video, you will require three such converters; one each for the red, green and blue signals {which you can often get from a SCART socket; but note that domestic VCRs often don't generate RGB outputs, so you may need some kind of PAL to RGB converter}. You will also need to extract the timing signals -- the LM1881 does all this in a single 8-pin package.
Audio has a narrow enough bandwidth to be digitised using a "guessing" type converter. The CD standard is to sample to 16 bits {originally 14 bits but left-alighned in a 16-bit word with the lowest order 2 bits set to zero}, 44100 times a second.
The receiving device must, by definition, have an analogue output. It's got a freakin' loudspeaker in it, for crying out loud! Maybe even two..... all you need do is carefully cut it open, and solder some wires on to the terminals. Or, since a loudspeaker has an electromagnet in it, just place another coil around the outside of the headphone. A current will be induced in this, varying in perfect sympathy with the current flowing in the speaker's coil. Or of course you could train a microphone on the speaker itself.
Whatever method you use, any distortion it introduces will be irretrievable {and so steps should be taken to minimise it} but at least it won't get any worse with subsequent copying.
Copyrights and patents owned by a real person usually don't immediately expire when that person dies: they can be passed on to next-of-kin, unless they made a specific provision in their will. But when a company "dies" it may not necessarily have a "next-of-kin". Or it may take so long to find exactly who owns the rights in question that they expire in the meantime. Breach of copyright is a civil offence, and only the copyright holder can take you to court.
If you live on a piece of land for twelve years, and during that time nobody tries to evict you or charge you rent, you own it. The same doctrine probably could be held to cover the legal fiction of "intellectual property", especially as there is now precedent for compulsory purchase of IP.
Beside which, nVidia is not innovating much nowadays. It's a matter of time now before either their graphics cards are reverse-engineered, or a law is passed somewhere in the world making full disclosure mandatory.
AFAICT it is the fact that nVidia graphics cards contain some so-called "intellectual property" claimed to belong to SGI {as if ideas could ever belong to anyone} that is preventing nVidia to release a true open-source driver enabling them to be used to the fullest extent under the popular GNU/Linux operating system and others.
If SGI are bought out, the purchaser might be more keen to release the necessary information. Alternatively, if SGI are wound up, then the information might effectively revert to the public domain by default {since there will be no party in a position to assert a claim over it}.
{Of course, it's also possible that nVidia are using the egregiuous technique of "crippling" a "£200" graphics card by making a slight change to the firmware [so the driver for the £200 card won't work with it] and selling it for £30. If they can make a profit selling the card for £30, then why should they get away with charging £200 for it? An open-source driver would reveal this blatant deception and dog-in-the-mangerism for what it is.}
The Nestlé company sells powdered baby milk in the third world, using advertising methods that would be illegal in the West to suggest that it is "better" for babies than the natural milk that they have been drinking from their mothers' own breasts for years without ill effect. The upshot of this is twofold. Mothers who don't know better are left out-of-pocket buying an unnecessary product, and babies are exposed to health hazards from the unclean water {not to mention they don't get the massive immune system boost that comes from drinking breast milk}.
It isn't strictly Nestlé's fault that the local drinking water is so polluted, but it's still downright irresponsible of them to market their products in this way.
In 1998 I almost bought a MiniDisc recorder. But I changed my mind at the last minute; and instead, I bought enough bits to build a PC and one of the first ever ATAPI CD-R drives. Best decision I ever made.
You only need to sue if there has been a violation of civil law. What Sony have done violates criminal law on several counts -- it is deception, misuse of a computer, criminal damage and aiding and abetting the misuse of a computer. Don't call a lawyer, call the police!
I think there is a market for a small Linux distro which will load itself to RAM {thereby allowing the CD to be ejected}; then wait for a music CD, and copy the tracks to a USB-storage device or something similar. There is no need for an X server, so everything probably could be kept quite small. In fact, there probably is room on a 1GB device for the OS, the CD tracks themselves as.wav files and maybe even the.mp3s. I would imagine that in countries where mathematics cannot be patented, even a pre-compiled LAME is legal.
A simple analogue to digital converter for RGB signals can be made with a dozen 2901 quad comparators and some 74HC chips. This gives you 12 bits {4096 colours}. Sure, it's not much; but add a digital-to-analogue converter, an op-amp and the same circuit again, and you have a 24-bit {2**24 colours} circuit. You can build all this on breadboard. Stick in a 1881 sync separator, and you have a device that will capture the signal straight out of a SCART socket directly. You just need an I/O port wide enough to take it all. If you can still find a mobo with the old-style 16-bit expansion slots, and they can be overclocked to 11MHz instead of the usual 8, so much the easier for you. 32-bit expansion slots are by all standards a 'mare to interface to -- you'd almost think they didn't want us building our own homebrew appliances to plug into our own computers?!
If you are not constrained by the limitations of breadboard, then you can go for something much less messy. But I think it's important to get the point across that it's possible to build A-to-D and D-to-A from some really low-tech stuff -- well, not exactly bronze age, but certainly within the grasp of anyone who knows the way to their nearest Maplin store.
Why? No good reason, really. Why would you want to climb a mountain? Why would you want to pee really high up a wall? Some people just like doing things other people don't pay much mind to. Merionesianism, I guess.
To the uninitiated, a GNU userland on top of an OpenSolaris kernel will feel just like a GNU userland on top of a Linux Kernel, or a FreeBSD kernel, or the HURD. Come to think of it, if it was possible to get a GNU userland running on top of the Windows kernel, you would be hard pushed to tell it from a GNU userland running on anything else.
In the end, it's really probably a good thing. The Open Source world is a genuine meritocracy: there is no way to survive except to be the fittest {nobody to bribe and nothing to bribe them with}, and so the competition is fierce. Yet there is no shame {and indeed, little impediment} in failure, either: up from the ashes of disaster, grow the roses of success. Open Source can only get better, according to an equation of the form
Q = 1 - e ** (-k * t)
where Q = quality expressed on a scale 0 {no solution to problem}.. 1 {perfection}
k = some constant
t = time
Windows, by its very way of being, has instilled in people a certain culture that is at odds with the Open Source movement.
Windows brings the idea that the act of writing software is a service that must be paid for, by charging money for the privilege of installing said software. The Open Source movement believes that the act of writing software should not chargeable. After all, the programmer who wrote it was going to write it anyway irrespective whether or not you paid for it; so refusing to hand it over without payment is just playing dog-in-the-manger -- and contravening Freedom Two in the process. But Windows goes on further and directly contravenes Freedoms One and Three {both of which are contingent upon the user having access to the source code, which is not supplied with the software} and even goes some way towards infringing upon Freedom Zero by means of the -- mercifully unenforcible -- End User Licence Agreement.
If something costs money, goes the thinking, then it can be stolen. And so there has appeared a group of people who have obtained Windows and other software without paying for it. Microsoft calls these people "pirates" and "thieves"; in reality they have stolen nothing but are merely attempting unilaterally to assert Freedoms Zero and Two. This fuels an attitude amongst these people that they are "sticking it to The Man" somehow, when in fact they are just as dependent upon The Man as any paying customer. In fact, by far the vast majority of software used in the world consists of unpaid-for copies of Microsoft Windows and applications running upon it.
Back in The Days, when computer users had no choice but to be knowledgeable in the field, the Four Freedoms could to a large extent be taken for granted. Computer users were effectively a society who looked after their own. Everyone benefitted from everyone else's work, and collaboration was the only way to advance. The community shared the bond that comes from a survival situation. The incompatibilities between different kinds of computer meant that the only feasible way to share software was in the form of source code {and often, different implementations of the "same" programming language meant that even this needed significant alteration}.
Today, it is possible even for a person who knows nothing about the inner workings of a computer to use one. This situation has led to a large number of people who are ignorant of the existence of this thing called "source code", believing it to be of importance only to boffins in laboratories somewhere. This attitude is deplorable: it is like someone who lives in a city not caring about which plants are edible, or how to build a bivouac, dismissing such knowledge as of importance only to "survival nutters" and the Military. Even if you do not know these things yourself, it is important at least to have friends who know these things -- it might save your life one day.
Ignorance of the benefits of the Four Freedoms is exactly what fuels the acceptance of the way Windows, and other closed-source software, trample on them. Windows users typically have been taught not to care about Open Source software. Either they expect to have to pay for software and not get the source code, or they expect to be able to get it without paying {like they are doing something big and clever} and do not care about the source code. Windows users who go on to become developers typically perpetuate the Cycle of Abuse by releasing their software closed-source. {cf. children brought up in violent families who go on to inflict violence upon their own children}. In some cases, the abuse of users' Four Freedoms is malicious but in many cases, it may be attributed to ignorance caused by developers having no better example.
The kind of computer user who is savvy enough to understand the importance of the Four Freedoms in general, and source code in particular, typ
Sounds too much like VAT to me. Thatcher's government steadily increased VAT from 8% to 15% to 17.5% and broadened its scope, while cutting income tax for the rich. Blair is just carrying on the programme -- there are tories on his left FCOL!
You have to tax people where they earn it, because people do not spend everything they earn. Otherwise, how are you supposed to distribute the wealth of the nation equitably? Face it, the poor are not going to get any richer, unless the rich get a little poorer.
We also need a maximum wage. Nobody, absolutely nobody, needs more than £100 000 a year to live on -- I say we should tax the rich till the pips squeak. And a Basic Living Allowance, paid to everybody at a flat rate {cheapest to administrate}, which would replace all state benefits. Plus free veterinary care for cats and dogs -- it's a proven fact that people who keep pets are fitter {from the exercise, mental stimulation and company}, so it would be well worth a little money to save the NHS a lot of money.
What is this Broadcast Flag of which you speak? Is this something that indicates whether the content being broadcast is "programmes" or "adverts", so that the recorder knows when to insert chapter markers / stop recording? If so, I would love to see something like this. It would make every TV station just like the BBC!
Re:or you could give us a torrent link...
on
OpenBSD 3.8 Released
·
· Score: 2, Funny
Given that it's released under the BSD licence, which everyone on Slashdot knows is so much better and freer than the GPL, who exactly do you think is likely to stop you creating your own ISO images?
Although seeing as it's OpenBSD, which is famed for being para.... er, security-conscious, almost nobody is going to want to download it if you do create one.....
There are still a few stunts you can pull if you have several incomes on which you are paying taxes separately. Corporations pay taxes too; and it's a standard practice to set up a subsidiary to whom you sell goods at an apparent loss in order to keep your own taxes in a lower bracket, sudden jump or no sudden jump.
I still don't see why we can't do away with the old stepwise linear regression for taxes {which approximates a polynomial, and was invented purely to make it feasible to do tax calculations without the aid of then-non-existent computers} and introduce a pure polynomial regression instead. For instance, your tax would be worked out according to
a + b * x + c * x ** 2 + d * x ** 3
which is easy for the Government's computers to do for every taxpayer, yet still possible for a human being to verify their own tax figure at home.
At first this probably would mean some people would be better off and others would be worse off. But this is bound to happen with any change to the tax regime. If the coefficients were chosen carefully, the effects could be minimised, and the long-term benefits {from not just disincentivising a particular kind of tax fraud, but making it physically impossible} would outweigh the short-term disadvantages.
Also, I probably would do away with Income Tax for employed people and introduce a Wages Tax instead {which would be payable by employers in respect of wages paid to their employees}. For the working classes, who receive their wages with the tax already taken out, there would be little visible change; except that the figure quoted in job advertisements would be the actual, "take home" figure, net of all taxes. This probably would make people feel less like they were being cheated {"It said £12000 a year so how come I'm only taking home £9600?"}
1. When you move a destination object, symlinks don't follow the target . This leaves "broken" symlinks that refer to nothing. Why doesn't the mv command move these too?
Because you might not have wanted the symlink to follow the target!
For instance, say I have two scripts share_broadband {which sets up a simple firewall and NAT through my broadband line} and share_modem {which does something similar, but uses my POTS modem} on my system. Now I create a symlink share_internet which points to share_broadband. {If my ADSL packs up, all I need to do is re-point one symlink and I can maintain some internet functionality.}
If I rename share_broadband, perhaps because I am making changes to it, I might prefer share_broadband to remain as a broken symlink {which will do nothing} than to point to the file being edited {which might be catastrophic}.
2. When you symlink a symlinked folder, the root symlink is ignored.
That's right. Why should it be any other way? You might have removed that symlink for a reason. It is really none of the computer's business either way.
3. Symlinks cause all kinds of weirds around chrooted file systems
A chrooted file system deliberately cannot see anything outside of the chroot. If it could, that would ruin the whole point of chroot! Look into bindmount if you really want to access stuff outside a chroot from within the chroot.
Symbolic links make the Unix file system non-hierarchical, resulting in multiple valid path names for a given file
Rigidly hierarchical file systems are a bad idea anyway. Well, they're OK for machines, but not for us human beings. The future of filing systems is going to be thematic. With something physical like a stamp collection, you are pretty much stuck with one categorisation. If you have organised your stamps by country, you have to look through several pages to find all the stamps with flowers on them. But with a filing system, you can easily have multiple listings. Caveat: you can run out of "theme tags" if you do not design it carefully.
NT *was* going to have executables that pretended to be files, i.e. when you opened the executable to get the contents it would run and return the output rather than the by bytes of the executable, with a special NT syscall to read the *real* contents
You mean like a CGI-script? This sounds like a bad idea. At least CGI scripts only behave that way when run through httpd. For that to be default behaviour at the OS level sounds kind of dodgy. Especially when all we need to do is put `backticks` around a filename to get that behaviour. eg. {quoting from memory, so I may have this wrong}
$ for i in *mp3; do mv $i `echo $i |sed -e's/\.wav//'`; done
for changing foo.wav.mp3, bar.wav.mp3 and so on to foo.mp3, bar.mp3 and so on. Try doing that in a GUI!
Those are hard links, which aren't the same thing as symbolic links.
A hard link is when you have two directory entries which refers to the same file. In effect you have one file with two or more names. GDBM files are created this way; you have foo.pag and foo.dir which are hard links to each other. Obviously both files must exist within the same physical device, since they both resolve to a starting cylinder/head/sector position.
A symbolic link is a directory entry which does not contain the usual cylinder/head/sector starting position. Instead it resolves to a pathname, which can refer to a file anywhere on any mounted file system {or even nowhere}. The pathname referred to must then be further resolved. This may take longer to resolve, but the file referred to need not be on the same physical device as the symlink itself.
Virtual folders - hmm. Is this a bit like what Konqueror does when you type audiocd:/ in the address bar with a music CD in the drive, and it shows you a bunch of directories apparently containing the tracks in.wav,.mp3 and.ogg formats {and possibly more, if additional codecs are installed} ?
Slashdot Mirror
That is the usual price of a pay-as-you-go movie {though you will need a picture stabiliser if you want to record it -- they stick Macrovision in the signal because if you chose when you wanted to watch it then you don't need to be able to record it, do you?}; and that is by some bizarre coincidence the same amount as I would pay to download one legally: Three pounds fifty.
* A point approximately one-third of the way into the story wherein, all characters having been established sufficiently in the mind of the reader for the purposes of the rest of the story, the action begins to speed up.
* The use of deliberately misleading language, in a situation where there are more than one possible outcome, so as to cause the reader to assume a particular one; and the subsequent revelation to the reader that their selection is in fact incorrect.
* {Most commonly encountered in a children's story} The tearful parting of the Hero's Ally, never to return again.
Burrows: Well, absolutely, and what makes it worse, sometimes at the end of a sentence I'll come out with the wrong fusebox!
Even if it's using ultrasound there will be a voice coil with a capturable magnetic field. As I understand it, it's using amplitude modulation {similar to MW and LW radio} but at tens as opposed to hundreds of kilohertz, and can be demodulated by simple rectification {which occurs naturally when a pressure wave is travelling through a fluid [air] and the receiving device [ear] is moving}.
The prototypes I have seen using this method had a very noticeable "scratchy" quality to them.
There are two basic kinds of A-D converter. TYPE ONE, the slow type, works by "guessing" and consists of a D-A converter feeding into one input of a comparator, which indicates whether the voltage from the D-A is greater or less than the voltage at the other input. A suitable external circuit is used to generate "guesses", and we can determine the two points between which the comparator changes state. This type of converter is slow since it takes at least as many guesses as bits in the output, but uses fewer components.
TYPE TWO, the fast type {sometimes called a "flash converter"} uses as many comparators as there are states in the digital output -- i.e. 2 ** [the number of bits]. The input signal is compared simultaneously to voltages corresponding to each step. A logic array is used to determine where in the chain the comparator outputs change state. There is no requirement for successive "guesses", the only time constraint being due to the propagation delay in the decision logic. However, for large numbers of bits, flash converters become unwieldy.
There is also a "half-flash" converter. This uses two flash A-to-Ds, a D-to-A and an op-amp. The first A-D digitises the signal to half the desired resolution, giving the higher-order bits of the result. The D-A converts this back to analogue; but there will be a difference, since the conversion is not exact. The op-amp amplifies this "error" signal and the second A-D digitises it, giving the lower-order bits of the result. Think of it as a number with a fraction: the first converter deals with the digits before the point and the second converter deals with the digits after the point. Half-flash converters depend more than most on circuit quality.
Any kind of analogue-to-digital converter requires a well-regulated power supply, tight tolerance resistances and above all, low noise components. All connections must be sound - dry solder joints and loose or dirty mechanical contacts introduce noise. However, it is just about feasible to construct an 8-bit, half-flash converter on copper strip breadboard. For video, you will require three such converters; one each for the red, green and blue signals {which you can often get from a SCART socket; but note that domestic VCRs often don't generate RGB outputs, so you may need some kind of PAL to RGB converter}. You will also need to extract the timing signals -- the LM1881 does all this in a single 8-pin package.
Audio has a narrow enough bandwidth to be digitised using a "guessing" type converter. The CD standard is to sample to 16 bits {originally 14 bits but left-alighned in a 16-bit word with the lowest order 2 bits set to zero}, 44100 times a second.
The receiving device must, by definition, have an analogue output. It's got a freakin' loudspeaker in it, for crying out loud! Maybe even two ..... all you need do is carefully cut it open, and solder some wires on to the terminals. Or, since a loudspeaker has an electromagnet in it, just place another coil around the outside of the headphone. A current will be induced in this, varying in perfect sympathy with the current flowing in the speaker's coil. Or of course you could train a microphone on the speaker itself.
Whatever method you use, any distortion it introduces will be irretrievable {and so steps should be taken to minimise it} but at least it won't get any worse with subsequent copying.
Copyrights and patents owned by a real person usually don't immediately expire when that person dies: they can be passed on to next-of-kin, unless they made a specific provision in their will. But when a company "dies" it may not necessarily have a "next-of-kin". Or it may take so long to find exactly who owns the rights in question that they expire in the meantime. Breach of copyright is a civil offence, and only the copyright holder can take you to court.
If you live on a piece of land for twelve years, and during that time nobody tries to evict you or charge you rent, you own it. The same doctrine probably could be held to cover the legal fiction of "intellectual property", especially as there is now precedent for compulsory purchase of IP.
Beside which, nVidia is not innovating much nowadays. It's a matter of time now before either their graphics cards are reverse-engineered, or a law is passed somewhere in the world making full disclosure mandatory.
AFAICT it is the fact that nVidia graphics cards contain some so-called "intellectual property" claimed to belong to SGI {as if ideas could ever belong to anyone} that is preventing nVidia to release a true open-source driver enabling them to be used to the fullest extent under the popular GNU/Linux operating system and others.
If SGI are bought out, the purchaser might be more keen to release the necessary information. Alternatively, if SGI are wound up, then the information might effectively revert to the public domain by default {since there will be no party in a position to assert a claim over it}.
{Of course, it's also possible that nVidia are using the egregiuous technique of "crippling" a "£200" graphics card by making a slight change to the firmware [so the driver for the £200 card won't work with it] and selling it for £30. If they can make a profit selling the card for £30, then why should they get away with charging £200 for it? An open-source driver would reveal this blatant deception and dog-in-the-mangerism for what it is.}
Does it still have dead babies in it?
The Nestlé company sells powdered baby milk in the third world, using advertising methods that would be illegal in the West to suggest that it is "better" for babies than the natural milk that they have been drinking from their mothers' own breasts for years without ill effect. The upshot of this is twofold. Mothers who don't know better are left out-of-pocket buying an unnecessary product, and babies are exposed to health hazards from the unclean water {not to mention they don't get the massive immune system boost that comes from drinking breast milk}.
It isn't strictly Nestlé's fault that the local drinking water is so polluted, but it's still downright irresponsible of them to market their products in this way.
I'm kinda interested! Pray tell me more .....
In 1998 I almost bought a MiniDisc recorder. But I changed my mind at the last minute; and instead, I bought enough bits to build a PC and one of the first ever ATAPI CD-R drives. Best decision I ever made.
You only need to sue if there has been a violation of civil law. What Sony have done violates criminal law on several counts -- it is deception, misuse of a computer, criminal damage and aiding and abetting the misuse of a computer. Don't call a lawyer, call the police!
I think there is a market for a small Linux distro which will load itself to RAM {thereby allowing the CD to be ejected}; then wait for a music CD, and copy the tracks to a USB-storage device or something similar. There is no need for an X server, so everything probably could be kept quite small. In fact, there probably is room on a 1GB device for the OS, the CD tracks themselves as .wav files and maybe even the .mp3s. I would imagine that in countries where mathematics cannot be patented, even a pre-compiled LAME is legal.
A simple analogue to digital converter for RGB signals can be made with a dozen 2901 quad comparators and some 74HC chips. This gives you 12 bits {4096 colours}. Sure, it's not much; but add a digital-to-analogue converter, an op-amp and the same circuit again, and you have a 24-bit {2**24 colours} circuit. You can build all this on breadboard. Stick in a 1881 sync separator, and you have a device that will capture the signal straight out of a SCART socket directly. You just need an I/O port wide enough to take it all. If you can still find a mobo with the old-style 16-bit expansion slots, and they can be overclocked to 11MHz instead of the usual 8, so much the easier for you. 32-bit expansion slots are by all standards a 'mare to interface to -- you'd almost think they didn't want us building our own homebrew appliances to plug into our own computers?!
If you are not constrained by the limitations of breadboard, then you can go for something much less messy. But I think it's important to get the point across that it's possible to build A-to-D and D-to-A from some really low-tech stuff -- well, not exactly bronze age, but certainly within the grasp of anyone who knows the way to their nearest Maplin store.
To the uninitiated, a GNU userland on top of an OpenSolaris kernel will feel just like a GNU userland on top of a Linux Kernel, or a FreeBSD kernel, or the HURD. Come to think of it, if it was possible to get a GNU userland running on top of the Windows kernel, you would be hard pushed to tell it from a GNU userland running on anything else.
In the end, it's really probably a good thing. The Open Source world is a genuine meritocracy: there is no way to survive except to be the fittest {nobody to bribe and nothing to bribe them with}, and so the competition is fierce. Yet there is no shame {and indeed, little impediment} in failure, either: up from the ashes of disaster, grow the roses of success. Open Source can only get better, according to an equation of the form where
Q = quality expressed on a scale 0 {no solution to problem}
k = some constant
t = time
Windows, by its very way of being, has instilled in people a certain culture that is at odds with the Open Source movement.
Windows brings the idea that the act of writing software is a service that must be paid for, by charging money for the privilege of installing said software. The Open Source movement believes that the act of writing software should not chargeable. After all, the programmer who wrote it was going to write it anyway irrespective whether or not you paid for it; so refusing to hand it over without payment is just playing dog-in-the-manger -- and contravening Freedom Two in the process. But Windows goes on further and directly contravenes Freedoms One and Three {both of which are contingent upon the user having access to the source code, which is not supplied with the software} and even goes some way towards infringing upon Freedom Zero by means of the -- mercifully unenforcible -- End User Licence Agreement.
If something costs money, goes the thinking, then it can be stolen. And so there has appeared a group of people who have obtained Windows and other software without paying for it. Microsoft calls these people "pirates" and "thieves"; in reality they have stolen nothing but are merely attempting unilaterally to assert Freedoms Zero and Two. This fuels an attitude amongst these people that they are "sticking it to The Man" somehow, when in fact they are just as dependent upon The Man as any paying customer. In fact, by far the vast majority of software used in the world consists of unpaid-for copies of Microsoft Windows and applications running upon it.
Back in The Days, when computer users had no choice but to be knowledgeable in the field, the Four Freedoms could to a large extent be taken for granted. Computer users were effectively a society who looked after their own. Everyone benefitted from everyone else's work, and collaboration was the only way to advance. The community shared the bond that comes from a survival situation. The incompatibilities between different kinds of computer meant that the only feasible way to share software was in the form of source code {and often, different implementations of the "same" programming language meant that even this needed significant alteration}.
Today, it is possible even for a person who knows nothing about the inner workings of a computer to use one. This situation has led to a large number of people who are ignorant of the existence of this thing called "source code", believing it to be of importance only to boffins in laboratories somewhere. This attitude is deplorable: it is like someone who lives in a city not caring about which plants are edible, or how to build a bivouac, dismissing such knowledge as of importance only to "survival nutters" and the Military. Even if you do not know these things yourself, it is important at least to have friends who know these things -- it might save your life one day.
Ignorance of the benefits of the Four Freedoms is exactly what fuels the acceptance of the way Windows, and other closed-source software, trample on them. Windows users typically have been taught not to care about Open Source software. Either they expect to have to pay for software and not get the source code, or they expect to be able to get it without paying {like they are doing something big and clever} and do not care about the source code. Windows users who go on to become developers typically perpetuate the Cycle of Abuse by releasing their software closed-source. {cf. children brought up in violent families who go on to inflict violence upon their own children}. In some cases, the abuse of users' Four Freedoms is malicious but in many cases, it may be attributed to ignorance caused by developers having no better example.
The kind of computer user who is savvy enough to understand the importance of the Four Freedoms in general, and source code in particular, typ
Sounds too much like VAT to me. Thatcher's government steadily increased VAT from 8% to 15% to 17.5% and broadened its scope, while cutting income tax for the rich. Blair is just carrying on the programme -- there are tories on his left FCOL!
You have to tax people where they earn it, because people do not spend everything they earn. Otherwise, how are you supposed to distribute the wealth of the nation equitably? Face it, the poor are not going to get any richer, unless the rich get a little poorer.
We also need a maximum wage. Nobody, absolutely nobody, needs more than £100 000 a year to live on -- I say we should tax the rich till the pips squeak. And a Basic Living Allowance, paid to everybody at a flat rate {cheapest to administrate}, which would replace all state benefits. Plus free veterinary care for cats and dogs -- it's a proven fact that people who keep pets are fitter {from the exercise, mental stimulation and company}, so it would be well worth a little money to save the NHS a lot of money.
What is this Broadcast Flag of which you speak? Is this something that indicates whether the content being broadcast is "programmes" or "adverts", so that the recorder knows when to insert chapter markers / stop recording? If so, I would love to see something like this. It would make every TV station just like the BBC!
Given that it's released under the BSD licence, which everyone on Slashdot knows is so much better and freer than the GPL, who exactly do you think is likely to stop you creating your own ISO images?
.... er, security-conscious, almost nobody is going to want to download it if you do create one .....
Although seeing as it's OpenBSD, which is famed for being para
I still don't see why we can't do away with the old stepwise linear regression for taxes {which approximates a polynomial, and was invented purely to make it feasible to do tax calculations without the aid of then-non-existent computers} and introduce a pure polynomial regression instead. For instance, your tax would be worked out according to which is easy for the Government's computers to do for every taxpayer, yet still possible for a human being to verify their own tax figure at home.
At first this probably would mean some people would be better off and others would be worse off. But this is bound to happen with any change to the tax regime. If the coefficients were chosen carefully, the effects could be minimised, and the long-term benefits {from not just disincentivising a particular kind of tax fraud, but making it physically impossible} would outweigh the short-term disadvantages.
Also, I probably would do away with Income Tax for employed people and introduce a Wages Tax instead {which would be payable by employers in respect of wages paid to their employees}. For the working classes, who receive their wages with the tax already taken out, there would be little visible change; except that the figure quoted in job advertisements would be the actual, "take home" figure, net of all taxes. This probably would make people feel less like they were being cheated {"It said £12000 a year so how come I'm only taking home £9600?"}
For instance, say I have two scripts share_broadband {which sets up a simple firewall and NAT through my broadband line} and share_modem {which does something similar, but uses my POTS modem} on my system. Now I create a symlink share_internet which points to share_broadband. {If my ADSL packs up, all I need to do is re-point one symlink and I can maintain some internet functionality.}
If I rename share_broadband, perhaps because I am making changes to it, I might prefer share_broadband to remain as a broken symlink {which will do nothing} than to point to the file being edited {which might be catastrophic}. That's right. Why should it be any other way? You might have removed that symlink for a reason. It is really none of the computer's business either way. A chrooted file system deliberately cannot see anything outside of the chroot. If it could, that would ruin the whole point of chroot! Look into bindmount if you really want to access stuff outside a chroot from within the chroot.
Those are hard links, which aren't the same thing as symbolic links.
A hard link is when you have two directory entries which refers to the same file. In effect you have one file with two or more names. GDBM files are created this way; you have foo.pag and foo.dir which are hard links to each other. Obviously both files must exist within the same physical device, since they both resolve to a starting cylinder/head/sector position.
A symbolic link is a directory entry which does not contain the usual cylinder/head/sector starting position. Instead it resolves to a pathname, which can refer to a file anywhere on any mounted file system {or even nowhere}. The pathname referred to must then be further resolved. This may take longer to resolve, but the file referred to need not be on the same physical device as the symlink itself.
Virtual folders - hmm. Is this a bit like what Konqueror does when you type audiocd:/ in the address bar with a music CD in the drive, and it shows you a bunch of directories apparently containing the tracks in .wav, .mp3 and .ogg formats {and possibly more, if additional codecs are installed} ?