This is about the fact that current incandescents only convert about 10% of the incoming electricity into light and throw away the rest as heat.
And if you want that heat in winter, it isn't exactly wasted now is it?
Putting a heating element so high mightn't be the best idea, but if you have good insulation it can help a little without having to resort to high power heaters.
My partner and I can canoodle without offending anyone else.
Why on earth would you want to be hugging your business partner? Could make for an awkward time at the office.
Deliberately being obtuse I know, I just find it curious that people feel the need to use gender neutral pronouns in an age where most people don't give two hoots who others shag.
Stereoscopic '3d' is a poor illusion. I for one will not be content with 3d displays until they are actually 3d and have real physical depth, like this one.
Radioactive decay turns unstable atoms into stable ones over time, and does so of it's own accord. Emitting alpha, beta or gamma radiation.
Nuclear fission is the splitting of one large atom into two not equally sized but both still sizable atoms through instigation by a neutron. _Very_ rarely with extemely large particles nuclear fission can happen through radioactive decay (instead of emitting alpha,beta etc, a large chunk comes off)
They are not the same thing in general. I'd say any dictionary that does not differentiate between them has failed it's job at conveying what it means.
That radioactive decay only sends off tiny bits and makes it more stable of it's own accord, and that nuclear fission does large chunks after being provoked by a neutron creating more unstable atoms, is a significant difference.
You could use the same argument for javelins as aside from sport their only use in antiquity was to kill people and/or hunting (I guarantee you a pistol depending on calibre can be more effective than a javelin for hunting purposes)
If sport is not a good enough purpose for you for people to have items, then let us ban chess sets also, as it's only 'good' purpose is gaming and a person could bludgeon someone with a decent quality stainless playing piece.
If I switch out pickup A for pickup B and then go back to A, how to I make sure i'm going to get the same sound in the second 'A' as in the first... It's bad enough when you put on new strings...
Aside from any miniscule (sub mm) changes in it from slight mounting changes, it is the same guitar, so it should sound the same.
If I attach a 50mm lens to my dslr, then swap it out for a 300mm one, then back to a 50mm one, should I expect the 50mm to behave differently between the two times? no. The laws of physics do not change just to piss off musicians, the same thing will do the same thing.
I'm a bassist and I have a Musicman Stingray and a Sterling. Two basses with different pickup magnets and different bodies. Is it worth the modularity to avoid having to spend say another $1500 on a bass? If you're serious about music, you'll just get two instruments one with each sound.
The price of decent musical instruments these days is ridiculous compared to the material costs, my digital piano costs about 4k new (not that I paid that much for it) but should be nowhere near that, some people are remedying this by building it themselves. When you can effectively build it yourself, for cheaper, with better materials.... you know something is seriously wrong.
The guitar amp equivalent of the digital camera emulating the soft errors of film is the equivalent of having a dsp doing the faulty but perceptually nice properties of the amp. Which with processing power is as it is is cheap as chips these days... while I agree unique hardware is nice.. tubes for the sound is a very uneconnomical way to achieve that sound, just like film to achieve kodachrome effect is very uneconomical today.
Real problem is though, no matter how good/bad of an amp you have if you can't play it will still sound shit... same deal with photos. Nothing trumps knowledge, experience and skill.
Modern digitals have had dsp settings for colours etc that people have liked for quite some time.. to the point it's an app on peoples phones these days. Would not have been unique at all, but at least they would have had a pony in the race if they did make non-crappy cameras again.. I have a six-20 kodak A, a medium format (6*9cm of film per shot) camera from the early 1950's sitting under my desk, the build quality was exceptional back then.
Define what that means. Are you suggesting that not patenting inventions, or failing to enforce the patents is ethical? If copycats take advantage of your R&D such that you lose market share and have to lay off employees, that is beneficial to society?
Yes, because r&d costs wind up being shared between all for everyone's mutual benefit, and those that do the initial research still get the first mover advantage which provides enough inventive to take the first steps. As a society we all benefit from the faster implementation and adoption of newer/better technologies.
The same argument as yours could be used as to how open source would never be paid for by companies, after all who would pay people to make software their competition could use just the same?/
If that were true, then someone else's implementation of a patented device wouldn't infringe.
Bingo, it isn't meant to. As an example the wright brothers patented "A flying machine" do you honestly think the idea of a flying machine was held up for 15 years to them alone? no, only the specific implementation of a flying machine they made was. Build a plane in the same manner as them and you got sued, whereas say a helicopter wouldn't exactly infringe on said patent if it were still valid, or any other sort of airplane that did not specifically use the same kind of design.
You need a better article, the narod.ru one if basically full of non-issues except for a few.
All of the hardware support ones are pretty much non-issues these days.
inability to run familiar Windows software
People should not see programs as ends in themselves, they should see their tasks as problems and programs as tools in which to complete them, if you can't run 3d studio blender will do your 3d models/animation just fine and dandy.
>No games. Full stop.
Wouldn't it be more correct to say _very few_ games? saying that not a single game exists for linux is just a little bit idiotic.
No native solutions for really simple file sharing in the local network.
If my grandmother can click an icon labelled 'file sharing' and set up an nfs share with two clicks, without any instruction, I'd bet most other people can too.
Steep learning curve
Again with the grandmother example, it took her all of a minute to figure out the fox icon was internet, and then she was fine.... if this is considered 'steep' you have far larger issues. Should be noted someone gave her a windows laptop afterwards, and she finds windows utterly unusable because she is now used to linux as it's what she started on.
His TL;DR version is most telling.
No games, no familiar software, no MS Office, no Active Directory or its equivalent.
this tells me "OMG, it isn't an exact replica of windows, wahhhh!"
Linux is unstable and prone to regressions when things which used to work break inexplicably.
Regressions do happen, but are you honestly going to tell me that there is not a single windows driver out there that is frequently updated that has not?
Too many Linux'es with incompatible configuration systems and different software packaging.
Again, "I have too much choice, wahhh!"
Linking to articles is useless when said articles are full of shite.
A better obstacle for linux's adoption is peoples dislike of change (people like to stick to what they know). But really, why should any linux users actually care about that? we already use it. Obsessing about what other people would like is useless to whatever we are up to.
but they can restart them without restarting the kernel.
What use is restarting a driver when it has left the hardware in a non-recoverable state? the driver would continue to be just as useless even when starting up a new copy.
No, of course they can. But that should only happen due to a bug in the kernel,
This is my point, microkernels are not flawless, all you are doing is making the process far more complex because of the crazy amount of message passing, lending itself to _more_ bugs, then saying it is all good because bugs in end drivers don't kill the system only ones in the core kernel can... in short bugs in the core kernel and complex messaging systems can still kill it, you do not magically get a 100% stable system as the original poster implied.
A thousand bugs that will have no adverse effects is infinitely more desirable than a handful of bugs that will bring down the system.
but you can still have bugs that bring down hardware, thusly the e1000e driver example, even with a microkernel a bug of that nature is still unrecoverable.
The point is 'microkernels always recover' is false, you can have still have bugs that leave the hardware in an unrecoverable state.
and several examples of microkernels working very well in real life, and for decades, no less
How many of those were on embedded systems with a very specific hardware set? let me specify the exact hardware that is being run and I'll give you a linux machine that never* goes down.
* - never in this case being over ten years with consistent use, we have to draw limits somewhere
As mentioned earlier, you want microkernels running on all your hardware, do so en masse, then you will discover they aren't perfect.
And there's no reason hardware would be left in an unrecoverable state. Most hardware can be entirely powered-off without power-cycling the entire machine.
Tell that to the e1000e users which intel didn't even put a lock bit on the card in regards to firmware so a bug in the driver bricks the card with the only method of recovery being manually flashing eeprom chips on the board.
And all the work needed to figure out how to gracefully recover has already been done in Linux, FreeBSD, etc., as it's required for S3/Suspend to work.
It doesn't really count as 'recovering' when you are putting it into a _known_ state before purposefully shutting it down. While I agree many drivers are capable of reinitializing hardware from an unknown state when it's seen to be acting funky, this is far from all hardware, and far from all drivers too.
They scientifically fix the most important issues.
really? since I'd say the most important issue with having a stable kernel is complexity.
Sure the individual components of a microkernel can be simple enough, but you are offloading the complexity to the interactions of the modules, which become exponentially more complex.
In the end you wind up with a more complex, more bug prone (because of said complexity) kernel, that takes an order of magnitude more development time for far less results.
With as many servers as I'm responsible for, it's only a matter of days between some system, somewhere, crashing.
More bugs in un privileged space in exchange for an incredibly stable micro kernel, would be an awesomely positive trade off.
So change all your linux machines to run on a straight microkernel, you will have a similar sample size to what you have now, and tell me how your results go
The proof is in the puddinng so to speak, microkernels do not magically fix everything, if hardware is left in an unrecoverable state microkernels still fail to fix the issues.
Going to microkernels is not some magical panacea that will fix all reliability issues.
The BSD license allows people to use code for pretty much whatever purpose, provided that they don't claim to have written it. The GPL allows people to use code for whatever purpose -- provided they conform to the GPL ideology, license their code under the GPL, and don't use it in certain ways that Stallman et al. think are unacceptable.
You tell me which embodies the spirit of freedom more.
The only freedom the GPL restricts, is your ability to limit others freedom.
By the same token the constitution (in theory) limits the power of the branches of government, so it must be _so_ anti-freedom, I mean without it they'd be free to do whatever they wished right? (Including the ability to remove freedom from others) which is far more 'free'.
The same deal with laws in general, whats this? it's illegal to throw a molotov cocktail into a shop window? it's restricting peoples freedom to throw fire bombs in certain places, sure it's allowing the shop owner to be free to go about his business without crazy bullshit, but it is limiting the freedom of the rioter/idiot.
Oh, but you have, actually... Every time you've seen a Linux system lock-up, you're seeing the effects of a monolithic kernel.
Last time I had a linux system lock-up was 1999, it was faulty hardware - dmesg gave me many 'danger will robinson' messages and windows could not even boot on the machine. So no, i have not 'seen the disadvantages' in the present decade or the last.
I would further pose that adding extra layers of unnecessary complexity lead to _more_ bugs, so should be avoided.
Slashdot Mirror
Some places aren't cold enough frequently enough to justify expenditure on central heating, heat pumps and the like. Think australia etc.
After all legally there isn't a single way to actually use those devices
Some people make their own videos with dv cams/modern dslrs and the like. While far from common, playback of such movies is a legal use.
This is about the fact that current incandescents only convert about 10% of the incoming electricity into light and throw away the rest as heat.
And if you want that heat in winter, it isn't exactly wasted now is it?
Putting a heating element so high mightn't be the best idea, but if you have good insulation it can help a little without having to resort to high power heaters.
When did that ever happen before?
Scientology protests back in 08
My partner and I can canoodle without offending anyone else.
Why on earth would you want to be hugging your business partner? Could make for an awkward time at the office.
Deliberately being obtuse I know, I just find it curious that people feel the need to use gender neutral pronouns in an age where most people don't give two hoots who others shag.
Stereoscopic '3d' is a poor illusion. I for one will not be content with 3d displays until they are actually 3d and have real physical depth, like this one.
Radioactive decay turns unstable atoms into stable ones over time, and does so of it's own accord. Emitting alpha, beta or gamma radiation.
Nuclear fission is the splitting of one large atom into two not equally sized but both still sizable atoms through instigation by a neutron. _Very_ rarely with extemely large particles nuclear fission can happen through radioactive decay (instead of emitting alpha,beta etc, a large chunk comes off)
They are not the same thing in general. I'd say any dictionary that does not differentiate between them has failed it's job at conveying what it means.
That radioactive decay only sends off tiny bits and makes it more stable of it's own accord, and that nuclear fission does large chunks after being provoked by a neutron creating more unstable atoms, is a significant difference.
Sport?
You could use the same argument for javelins as aside from sport their only use in antiquity was to kill people and/or hunting (I guarantee you a pistol depending on calibre can be more effective than a javelin for hunting purposes)
If sport is not a good enough purpose for you for people to have items, then let us ban chess sets also, as it's only 'good' purpose is gaming and a person could bludgeon someone with a decent quality stainless playing piece.
If I switch out pickup A for pickup B and then go back to A, how to I make sure i'm going to get the same sound in the second 'A' as in the first... It's bad enough when you put on new strings...
Aside from any miniscule (sub mm) changes in it from slight mounting changes, it is the same guitar, so it should sound the same.
If I attach a 50mm lens to my dslr, then swap it out for a 300mm one, then back to a 50mm one, should I expect the 50mm to behave differently between the two times? no. The laws of physics do not change just to piss off musicians, the same thing will do the same thing.
I'm a bassist and I have a Musicman Stingray and a Sterling. Two basses with different pickup magnets and different bodies. Is it worth the modularity to avoid having to spend say another $1500 on a bass? If you're serious about music, you'll just get two instruments one with each sound.
The price of decent musical instruments these days is ridiculous compared to the material costs, my digital piano costs about 4k new (not that I paid that much for it) but should be nowhere near that, some people are remedying this by building it themselves. When you can effectively build it yourself, for cheaper, with better materials.... you know something is seriously wrong.
But word itself is merely a toy, people who care about typesetting will be suing latex like they always have.
Both word and openoffice/libreoffice are for office people with limited use cases.
and the best argument against democracy is it winds up two wolves and a sheep deciding what is for dinner.
The guitar amp equivalent of the digital camera emulating the soft errors of film is the equivalent of having a dsp doing the faulty but perceptually nice properties of the amp. Which with processing power is as it is is cheap as chips these days... while I agree unique hardware is nice.. tubes for the sound is a very uneconnomical way to achieve that sound, just like film to achieve kodachrome effect is very uneconomical today.
Real problem is though, no matter how good/bad of an amp you have if you can't play it will still sound shit... same deal with photos. Nothing trumps knowledge, experience and skill.
Modern digitals have had dsp settings for colours etc that people have liked for quite some time.. to the point it's an app on peoples phones these days. Would not have been unique at all, but at least they would have had a pony in the race if they did make non-crappy cameras again.. I have a six-20 kodak A, a medium format (6*9cm of film per shot) camera from the early 1950's sitting under my desk, the build quality was exceptional back then.
No, he probably will not have a normal life again. And you know how he could have avoided it? By releasing only documents showing wrongdoing
Who determines what is 'wrong'?
Define what that means. Are you suggesting that not patenting inventions, or failing to enforce the patents is ethical? If copycats take advantage of your R&D such that you lose market share and have to lay off employees, that is beneficial to society?
Yes, because r&d costs wind up being shared between all for everyone's mutual benefit, and those that do the initial research still get the first mover advantage which provides enough inventive to take the first steps. As a society we all benefit from the faster implementation and adoption of newer/better technologies.
The same argument as yours could be used as to how open source would never be paid for by companies, after all who would pay people to make software their competition could use just the same?/
If that were true, then someone else's implementation of a patented device wouldn't infringe.
Bingo, it isn't meant to. As an example the wright brothers patented "A flying machine" do you honestly think the idea of a flying machine was held up for 15 years to them alone? no, only the specific implementation of a flying machine they made was. Build a plane in the same manner as them and you got sued, whereas say a helicopter wouldn't exactly infringe on said patent if it were still valid, or any other sort of airplane that did not specifically use the same kind of design.
Without that I see no reason why they shouldn't sue if someone copies their API definitions.
Because given an overly trivial task, two engineers can come up with the same api?
If it is not 'creative' it does not deserve copyright. What should be considered creative or not is another issue entirely.
That was why head-to-head benchmarks of Linux and FreeBSD on SMP and file-system performance showed FreeBSD coming out ahead pretty consistently.
When overall performance of freebsd is lower, who gives a damn?
You need a better article, the narod.ru one if basically full of non-issues except for a few.
All of the hardware support ones are pretty much non-issues these days.
inability to run familiar Windows software
People should not see programs as ends in themselves, they should see their tasks as problems and programs as tools in which to complete them, if you can't run 3d studio blender will do your 3d models/animation just fine and dandy.
>No games. Full stop.
Wouldn't it be more correct to say _very few_ games? saying that not a single game exists for linux is just a little bit idiotic.
No native solutions for really simple file sharing in the local network.
If my grandmother can click an icon labelled 'file sharing' and set up an nfs share with two clicks, without any instruction, I'd bet most other people can too.
Steep learning curve
Again with the grandmother example, it took her all of a minute to figure out the fox icon was internet, and then she was fine.... if this is considered 'steep' you have far larger issues. Should be noted someone gave her a windows laptop afterwards, and she finds windows utterly unusable because she is now used to linux as it's what she started on.
His TL;DR version is most telling.
No games, no familiar software, no MS Office, no Active Directory or its equivalent.
this tells me "OMG, it isn't an exact replica of windows, wahhhh!"
Linux is unstable and prone to regressions when things which used to work break inexplicably.
Regressions do happen, but are you honestly going to tell me that there is not a single windows driver out there that is frequently updated that has not?
Too many Linux'es with incompatible configuration systems and different software packaging.
Again, "I have too much choice, wahhh!"
Linking to articles is useless when said articles are full of shite.
A better obstacle for linux's adoption is peoples dislike of change (people like to stick to what they know). But really, why should any linux users actually care about that? we already use it. Obsessing about what other people would like is useless to whatever we are up to.
but they can restart them without restarting the kernel.
What use is restarting a driver when it has left the hardware in a non-recoverable state? the driver would continue to be just as useless even when starting up a new copy.
No, of course they can. But that should only happen due to a bug in the kernel,
This is my point, microkernels are not flawless, all you are doing is making the process far more complex because of the crazy amount of message passing, lending itself to _more_ bugs, then saying it is all good because bugs in end drivers don't kill the system only ones in the core kernel can... in short bugs in the core kernel and complex messaging systems can still kill it, you do not magically get a 100% stable system as the original poster implied.
bug in question didn't bring down the linux kernel at all, just the device.
If you are insinuating microkernels cannot kernel panic (unrecoverable severe error causing halt of execution), experience says otherwise
A thousand bugs that will have no adverse effects is infinitely more desirable than a handful of bugs that will bring down the system.
but you can still have bugs that bring down hardware, thusly the e1000e driver example, even with a microkernel a bug of that nature is still unrecoverable.
The point is 'microkernels always recover' is false, you can have still have bugs that leave the hardware in an unrecoverable state.
and several examples of microkernels working very well in real life, and for decades, no less
How many of those were on embedded systems with a very specific hardware set? let me specify the exact hardware that is being run and I'll give you a linux machine that never* goes down.
* - never in this case being over ten years with consistent use, we have to draw limits somewhere
As mentioned earlier, you want microkernels running on all your hardware, do so en masse, then you will discover they aren't perfect.
And there's no reason hardware would be left in an unrecoverable state. Most hardware can be entirely powered-off without power-cycling the entire machine.
Tell that to the e1000e users which intel didn't even put a lock bit on the card in regards to firmware so a bug in the driver bricks the card with the only method of recovery being manually flashing eeprom chips on the board.
And all the work needed to figure out how to gracefully recover has already been done in Linux, FreeBSD, etc., as it's required for S3/Suspend to work.
It doesn't really count as 'recovering' when you are putting it into a _known_ state before purposefully shutting it down. While I agree many drivers are capable of reinitializing hardware from an unknown state when it's seen to be acting funky, this is far from all hardware, and far from all drivers too.
They scientifically fix the most important issues.
really? since I'd say the most important issue with having a stable kernel is complexity.
Sure the individual components of a microkernel can be simple enough, but you are offloading the complexity to the interactions of the modules, which become exponentially more complex.
In the end you wind up with a more complex, more bug prone (because of said complexity) kernel, that takes an order of magnitude more development time for far less results.
How does this solve a damn thing?
With as many servers as I'm responsible for, it's only a matter of days between some system, somewhere, crashing.
More bugs in un privileged space in exchange for an incredibly stable micro kernel, would be an awesomely positive trade off.
So change all your linux machines to run on a straight microkernel, you will have a similar sample size to what you have now, and tell me how your results go
The proof is in the puddinng so to speak, microkernels do not magically fix everything, if hardware is left in an unrecoverable state microkernels still fail to fix the issues.
Going to microkernels is not some magical panacea that will fix all reliability issues.
The BSD license allows people to use code for pretty much whatever purpose, provided that they don't claim to have written it. The GPL allows people to use code for whatever purpose -- provided they conform to the GPL ideology, license their code under the GPL, and don't use it in certain ways that Stallman et al. think are unacceptable. You tell me which embodies the spirit of freedom more.
The only freedom the GPL restricts, is your ability to limit others freedom.
By the same token the constitution (in theory) limits the power of the branches of government, so it must be _so_ anti-freedom, I mean without it they'd be free to do whatever they wished right? (Including the ability to remove freedom from others) which is far more 'free'.
The same deal with laws in general, whats this? it's illegal to throw a molotov cocktail into a shop window? it's restricting peoples freedom to throw fire bombs in certain places, sure it's allowing the shop owner to be free to go about his business without crazy bullshit, but it is limiting the freedom of the rioter/idiot.
Oh, but you have, actually... Every time you've seen a Linux system lock-up, you're seeing the effects of a monolithic kernel.
Last time I had a linux system lock-up was 1999, it was faulty hardware - dmesg gave me many 'danger will robinson' messages and windows could not even boot on the machine. So no, i have not 'seen the disadvantages' in the present decade or the last.
I would further pose that adding extra layers of unnecessary complexity lead to _more_ bugs, so should be avoided.