" The "Laws of Thermodynamics" are a description of what is observed to happen to gases under experimental conditions. There's no real evidence that they scale out, you know, and a fair bit to hint that they don't. And even if they do apply, we know that they talk about what happens in the long term. They don't apply short-term. "
Okay, lets think about this more closely,
The three laws of Thermodynamics.
1: Energy is conserved. 2: Entropy may only increase 3: Absolute Zero may not be attained.
Let's look at them in more detail.
1: Energy is conserved.
Originally discovered in the context of mechanics, seems to be true for gases, electomagnetism, nuclear reactions, quantum mechanics. It seems to have been confimed in pretty much every physics experiment.
2: Entropy may only increase
This can also be restated as 'heat may not of it's own accord pass from a cold object to a hotter object'. As for systems biasing to complexity they offset this by creating much more disorder as waste.
3: Absolute zero may not be attained.
It hasn't been, and the predictions for dealing with very low temperatures have been bourne out the predictions for the amount of other energy that must be converted into heat to make it work.
As for it being the kooks who come up with whacky ideas and achieve them, to misquote Carl Sagan 'they laughed at Einstien, but they also laughed at Bozo the Clown.'
" find a phone company who promises to NEVER disclose your number. a company who's numbers can not be randomly generated. "
Or alternatively register with the Telephone Protection Service and never receive a junk phone call again which is considerably easier. Guess it sucks to live in the UK where we have strong data protection laws for this kind of thing.
PS Can you name a telephone company that sells numbers that can't be randomly generated.
" Surely the correct solution is to patch cdparanois with the --force option, and then simply rip track 1 with --force, and then tracks 2-? No doubt the patch would be accepted into the main branch, too. "
Correct as in of most use to the open source community. Yes. Correct as in the shortest possible patch to demonstrate the futility of the protection. No
" You didn't realise that cdparanoia takes a parameter telling it which tracks to rip (e.g. "cdparanoia 2-" for track 2 to last track). This is explained in the first few lines of the manual. So you rummaged around the raw source code of cdparanoia and changed it before you looked at its manual. "
No, track 1 is marked as a data track but contains audio - if you rip cdparanoia 2- you will miss off the first track, and cdparanoia 1- fails because it refuses to rip a data track. I've read the cdparanoia manual reasonably thoroughly although I really wanted a --force option.
" You used diff without the -U option, and didn't even tell us which file you commented out the exit() in. Which file is it? What's the context? "
fair point - fixed
" You used a C++/C99 comment delimiter in what is a C89 source code. Hope you have have a lax compiler that defaults to C99 or ignores standards, because it'll choke on that. "
I know, I was aiming for the shortest patch I could.// is two bytes,/* */ is four.
[alternate reply - it's a my fork and I'll use it how I like]
" You made a web-page about it and posted it on Slashdot, no doubt causing much wailing and gnashing of teeth from people even less experienced at editing source code than you, when all they needed to do was run the cdparanoia command differently. "
Er - this is a rehash of point 1 but more insulting isn't it?
" You bought an Avril Lavigne CD! Dude, how could you? "
Guess I shouldn't mention the Britney collection then:-)
How strange - try eliminating the network card by doing a transfer over the localhost interface. If that's still really slow my only guess is your ssh has bizarre processor optimisations so you might want to recompile for your system.
When I moved to Linux from Scratch I noticed a 30% speed up in this type of thing.
That's very strange, my HP Omnibook 800CT runs redhat 7.2 and usually manages around 500kbytes / sec over ssh between linux machines.
A P200 I have to hand manages 720kbytes a second.
Do you get the same problems with non encrypted transfers as 50kbytes/sec is the kind of speeds I was seeing with a full/half duplex mismatch between my desktop machine and the switch.
My computer is mine and is shared with whoever I wish to share it with, my girlfriend, my cat, even some people on the internet.
It certainly is not shared with people who wrote the software for it any more than your computer is shared with me because you read my comment on slashdot.
What is in dispute is the right to own and use the stuff I purchased in a way that I want. And one of the restrictions of sale is that the former owner transfers me all those rights.
Er, your marketing department and business model sucks then,
try doing the following,
sell a printer that makes a profit sell cartridges that make a profit too
then your cartridges should be price competitive with the knockoffs [i.e. more expensive but not four times the price], the consumer gets a valid choice of cartridge quality and you don't have to pay the R&D expenses to develop the non copy chips.
www.ex-parrot.com/~pete/copycontrol.html
My account of my first copy control cd - I can copy it but I can't play it. Includes possibly the worlds shortest patch to break the copy control technology.
WARNING - this page might be a violation of the DMCA but I'm British so [currently] I don't care.
You can't work more than 48 hours per week, unless you have explicitly signed out from the legislation, however you can sign back whether your employer likes it or not.
You must have 11 hours uninterrupted between consecutive working days
You may not work more than six consecutive days out of seven *or* tweleve consecutive days out of fourteen [employers choice]
You are entitled to at least twenty minutes uniterrupted rest if the working day is six hours or longer
I believe, if you are at risk of being fired the best thing to do is inform your employer you no longer waive the right to a long week which makes it very difficult for them to fire you.
I'm reminded of a quote in the StudioSpares catalogue about speaker power rating. Something like
'These speakers really are 100W. To test we recommend the following amplifier, if you wish to test other manufacturers speakers we recommned a fire extinguisher'
JRR Tolkien died in 1973, having written what is regarded to be many to be one of the greatest works ever written, published in 1948.
Both these dates are before I was born.
Are you arguing that it is beneficial to the general public to grant full and exclusive control to his family until 2068, or probably sometime after my death unless I am especially lucky.
In summary, despite being one of the great works, completed a before I [or my parents] were born is unlikely to enter the public domain until the my own grand children's time - providing there is no copyright law extension passed in the next 60 years.
However, to respond to the final comment in your post, without copyright law, JRR Tolkiens family would not have the right to prevent my duplicating his work. Please explain why JRR Tolkiens *great grandchildren* preventing *my grandchildren* from duplicating the work will benefit the public domain and promote progress in the Arts and Sciences.
In particular, note that due to the money received from sales on The Lord of the Rings he would have essentially no incentive to publish the other Tolkien works since he is guaranteed an income for life from the earlier work.
Not so sure, I'm not a great C programmer, but I've patched my kernel twice with third party patches,
once to fix a bug in the DMA on my stupid laptop
once to add low latency for my machine with a DVD player.
I've also modified [a few lines of C] Oogle for my own personal use to add slight compression in the audio, and recompiling with processor optimisations smoothed the playback.
The last one is a benefit, if you have an application that you use frequently and an rpm based distribution, it's not too hard to do
rpm --rebuild foo.src.rpm rpm -Uvh foo.rpm
which will accelerate the required program.
More importantly the advantage is other people can change and add to the source, I use a RedHat kernel which they've patched from the stock kernel, I benefit from the fact they can do that even though it wasn't me that did it.
"Now, stop crying and tell me, IN WHAT WAY is SQLServer a bad piece of software and IN WHAT WAY is SQLServer f.e. badly licensed, compared to competitors like DB2 and Oracle ?"
Okay,
Bad Software:
Quite a few remote exploits that require machine reboots to patch.
Only runs on MS Windows and hence isn't very stable.
Badly Licensed :
By default you get n connection licenses, you have to pay more to get more. Oracle licenses with either a per user fee ($100 / user?) or an unlimited license ($10000 / processor?). I believe Oracle development licenses are free for a single user too.
I have to buy a Windows license to use my SQL Server license and that allows a third party to install and run software on my machine without my knowledge.
That said, I do think that SQL Server is one of the best pieces of software to come out of MS and if it ran on linux it might be quite compelling as an Oracle substitute as it's cheaper and almost as featureful.
I buy Software X I do not agree to the EULA
- I have rights to reverse engineer and backup the
software on the CD I modify the software to remove the EULA clickwrap. I then install this modified software.
At what point did I either break the law or agree to the EULA ?
Slashdot Mirror
However, it's still quite hard to exploit sendmail bound to 127.0.0.1 only and with an external firewall on port 25.
At least that's true for the last machine I installed. [redhat 9]
How do you tell if a trainer has an RFID embedded in the sole? It's not like you can see it.
"
The "Laws of Thermodynamics" are a description of what is observed to happen to gases under experimental conditions. There's no real evidence that they scale out, you know, and a fair bit to hint that they don't. And even if they do apply, we know that they talk about what happens in the long term. They don't apply short-term.
"
Okay, lets think about this more closely,
The three laws of Thermodynamics.
1: Energy is conserved.
2: Entropy may only increase
3: Absolute Zero may not be attained.
Let's look at them in more detail.
1: Energy is conserved.
Originally discovered in the context of mechanics, seems to be true for gases, electomagnetism, nuclear reactions, quantum mechanics. It seems to have been confimed in pretty much every physics experiment.
2: Entropy may only increase
This can also be restated as 'heat may not of it's own accord pass from a cold object to a hotter object'. As for systems biasing to complexity they offset this by creating much more disorder as waste.
3: Absolute zero may not be attained.
It hasn't been, and the predictions for dealing with very low temperatures have been bourne out the predictions for the amount of other energy that must be converted into heat to make it work.
As for it being the kooks who come up with whacky ideas and achieve them, to misquote Carl Sagan 'they laughed at Einstien, but they also laughed at Bozo the Clown.'
"
find a phone company who promises to NEVER disclose your number. a company who's numbers can not be randomly generated.
"
Or alternatively register with the Telephone Protection Service and never receive a junk phone call again which is considerably easier. Guess it sucks to live in the UK where we have strong data protection laws for this kind of thing.
PS Can you name a telephone company that sells numbers that can't be randomly generated.
"
Surely the correct solution is to patch cdparanois with the --force option, and then simply rip track 1 with --force, and then tracks 2-? No doubt the patch would be accepted into the main branch, too.
"
Correct as in of most use to the open source community. Yes.
Correct as in the shortest possible patch to demonstrate the futility of the protection. No
"
// is two bytes, /* */ is four.
:-)
You didn't realise that cdparanoia takes a parameter telling it which tracks to rip (e.g. "cdparanoia 2-" for track 2 to last track). This is explained in the first few lines of the manual. So you rummaged around the raw source code of cdparanoia and changed it before you looked at its manual.
"
No, track 1 is marked as a data track but contains audio - if you rip cdparanoia 2- you will miss off the first track, and cdparanoia 1- fails because it refuses to rip a data track. I've read the cdparanoia manual reasonably thoroughly although I really wanted a --force option.
"
You used diff without the -U option, and didn't even tell us which file you commented out the exit() in. Which file is it? What's the context?
"
fair point - fixed
"
You used a C++/C99 comment delimiter in what is a C89 source code. Hope you have have a lax compiler that defaults to C99 or ignores standards, because it'll choke on that.
"
I know, I was aiming for the shortest patch I could.
[alternate reply - it's a my fork and I'll use it how I like]
"
You made a web-page about it and posted it on Slashdot, no doubt causing much wailing and gnashing of teeth from people even less experienced at editing source code than you, when all they needed to do was run the cdparanoia command differently.
"
Er - this is a rehash of point 1 but more insulting isn't it?
"
You bought an Avril Lavigne CD! Dude, how could you?
"
Guess I shouldn't mention the Britney collection then
See also
http://www.ex-parrot.com/~pete/copycontrol.html
How I patched cdparanoia to copy Avril Lavigne in order to play it under linux.
How strange - try eliminating the network card by doing a transfer over the localhost interface. If that's still really slow my only guess is your ssh has bizarre processor optimisations so you might want to recompile for your system.
When I moved to Linux from Scratch I noticed a 30% speed up in this type of thing.
The laptop is an HP Omnibook 800CT
That's very strange, my HP Omnibook 800CT runs redhat 7.2 and usually manages around 500kbytes / sec over ssh between linux machines.
A P200 I have to hand manages 720kbytes a second.
Do you get the same problems with non encrypted transfers as 50kbytes/sec is the kind of speeds I was seeing with a full/half duplex mismatch between my desktop machine and the switch.
My computer is mine and is shared with whoever I wish to share it with, my girlfriend, my cat, even some people on the internet.
It certainly is not shared with people who wrote the software for it any more than your computer is shared with me because you read my comment on slashdot.
What is in dispute is the right to own and use the stuff I purchased in a way that I want. And one of the restrictions of sale is that the former owner transfers me all those rights.
Er, your marketing department and business model sucks then,
try doing the following,
sell a printer that makes a profit
sell cartridges that make a profit too
then your cartridges should be price competitive with the knockoffs [i.e. more expensive but not four times the price], the consumer gets a valid choice of cartridge quality and you don't have to pay the R&D expenses to develop the non copy chips.
www.ex-parrot.com/~pete/copycontrol.html
My account of my first copy control cd - I can copy it but I can't play it. Includes possibly the worlds shortest patch to break the copy control technology.
WARNING - this page might be a violation of the DMCA but I'm British so [currently] I don't care.
See the regulations at
http://www.hmso.gov.uk/si/si1998/19981833.htm
I forgot to mention the minimum of 20 days holiday per year too.
In the UK
You can't work more than 48 hours per week, unless you have explicitly signed out from the legislation, however you can sign back whether your employer likes it or not.
You must have 11 hours uninterrupted between consecutive working days
You may not work more than six consecutive days out of seven *or* tweleve consecutive days out of fourteen [employers choice]
You are entitled to at least twenty minutes uniterrupted rest if the working day is six hours or longer
I believe, if you are at risk of being fired the best thing to do is inform your employer you no longer waive the right to a long week which makes it very difficult for them to fire you.
How about a fantastical filme about hobbits called the two towers? I can't believe they thought they'd get away with that.
I'm reminded of a quote in the StudioSpares catalogue about speaker power rating. Something like
'These speakers really are 100W. To test we recommend the following amplifier, if you wish to test other manufacturers speakers we recommned a fire extinguisher'
JRR Tolkien died in 1973, having written what is regarded to be many to be one of the greatest works ever written, published in 1948.
Both these dates are before I was born.
Are you arguing that it is beneficial to the general public to grant full and exclusive control to his family until 2068, or probably sometime after my death unless I am
especially lucky.
In summary, despite being one of the great works, completed a before I [or my parents] were born is unlikely to enter the public domain until the my own grand children's time - providing there is no copyright law extension passed in the next 60 years.
However, to respond to the final comment in your post, without copyright law, JRR Tolkiens family would not have the right to prevent my duplicating his work. Please explain why JRR Tolkiens *great grandchildren* preventing *my grandchildren* from duplicating the work will benefit the public domain and promote progress in the Arts and Sciences.
In particular, note that due to the money received from sales on The Lord of the Rings he would have essentially no incentive to publish the other Tolkien works since he is guaranteed an income for life from the earlier work.
Not so sure, I'm not a great C programmer, but I've patched my kernel twice with third party patches,
once to fix a bug in the DMA on my stupid laptop
once to add low latency for my machine with a DVD player.
I've also modified [a few lines of C] Oogle for my own personal use to add slight compression in the audio, and recompiling with processor optimisations smoothed the playback.
The last one is a benefit, if you have an application that you use frequently and an rpm based distribution, it's not too hard to do
rpm --rebuild foo.src.rpm
rpm -Uvh foo.rpm
which will accelerate the required program.
More importantly the advantage is other people can change and add to the source, I use a RedHat kernel which they've patched from the stock kernel, I benefit from the fact they can do that even though it wasn't me that did it.
I've had two.
Once hard disk cable had fallen out, in the other my video card had melted and never worked again.
My cacheless celeron 300 (o/c 375) could software decode DVD at 1024x768 only very rarely dropping frames so I imagine this would just about do.
"
If Apple(who owns the copyright) says that you can't use it that way, then you can't use it that way! It's their decision.
"
Er, no.
Apple can prevent me from distributing the software or using an illegal copy of the software for which I have not paid.
In this case, simple copyright laws apply. You have not paid for the software therefore you can not use it.
"Now, stop crying and tell me, IN WHAT WAY is SQLServer a bad piece of software and IN WHAT WAY is SQLServer f.e. badly licensed, compared to competitors like DB2 and Oracle ?"
:
Okay,
Bad Software
Quite a few remote exploits that require machine reboots to patch.
Only runs on MS Windows and hence isn't very stable.
Badly Licensed :
By default you get n connection licenses, you have to pay more to get more. Oracle licenses with either a per user fee ($100 / user?) or an unlimited license ($10000 / processor?). I believe Oracle development licenses are free for a single user too.
I have to buy a Windows license to use my SQL Server license and that allows a third party to install and run software on my machine without my knowledge.
That said, I do think that SQL Server is one of the best pieces of software to come out of MS and if it ran on linux it might be quite compelling as an Oracle substitute as it's cheaper and almost as featureful.
"DRM gives someone else the final say over what your computer allows you to do "
"No it doesn't, except for with certain specific types of encrypted file. "
So, yes it does then.
My computer does what I say always. If I give it an instruction it obeys. What DRM will do is allow MS to override my decisions about my computer.
I don't see how this works for cases like this
I buy Software X
I do not agree to the EULA
- I have rights to reverse engineer and backup the
software on the CD
I modify the software to remove the EULA clickwrap.
I then install this modified software.
At what point did I either break the law or agree to the EULA ?
"
Well, if you aren't, then the GPL isn't binding either, since you aren't intrinsically "signing" anything when you use GPL'd code.
"
Correct. Ten points to the smart one at the back. Question 2. Why doesn't this matter?