"Now then, over the years, exceptions have been found. Shouting "FIRE!" in a crowded room is not protected speech. Urging an assembled mob to violence is not protected speech. Openly plotting to kill your leader is not protected speech"
Or to use a more common example: on US TV offensive speech is replaced with a beep.
>but there is no way of researching and proffiting from something like MP3
Xiph/OGG seen to be doing good research without patents, although they might not be profiting as much as Frauenhofer.
>Or is there anythign trivial, obvious or allready state of the art in them?
Not in MP3, but some of the patents are very broad.
However I was referring to amendmend 11:
>Member States shall ensure that a computer-implemented >invention may be claimed only as a product, that is as a >programmed device, or as a technical production process.
If this passes _software_ players end encoders should be ok.
>at least educate your self about the question, what is HARD computer science...
But I don't think hard science should be patentable whether it is physics or computer science.
No, they have not removed anything.
Piia-Noora Kauppi and others have tablet some amendments. It is not at all certain that the amendments will get the necessary votes. If you were working on a letter to you member of the European Parlament, then don't stop.
>I think the pint is pretty much a standard for beer the world over.
It might be a standard, but it is not very well defined. I was in Portugal this summer. I was not too happy about until i found out that in many places you could buy a "small pint" or "big pint" of beer.
Many places ordering a pint will get you 0.5 liter
2. Something like the MIT rootnet. http://slashdot.org/article.pl?sid=03/08 /29/232022 8 By having the routing in the AP, it would participate in the roofnet without a server. Servers use more electricity and are noisy.
3. Security for open accesspoints. You might leave our accesspoints open to share it with others. But you might want some extra security: a) Block port 25 for others so they don't send spam and get you blacklistet. b) Some VPN/SSH tunnels for privacy. c)Traffic shaping. You would like to reserve some bandwidth for yourself.
They are not shutting MS out. MS is a software company and I am sure MS can make a bid for contracts on this new system if _they want_ to. They might even win a contract if they are competitive.
and see "bigiron" stuff like accessing USB devices from java, python, wireless accesspoints, image manipulation, mini-libc for memory-constrained systems, smalltalk, XML, bluetooth, and printer drivers..
Normal Lindows users don't apt-get. CNR is simper to use than synaptic, kpackage, dselect, apt-get etc.
Technically the answer is versions and dependencies. I experienced Lindows 3 preinstalled on the laptop (http://www.agol.dk/elgaard/lindows.html). The Lindows specific programs were all cleanly installed as Debian packages, which make upgrading with apt safe and simpler. However the installed packages were an interesting mix of Debian/KDE/OO versions. Many programs were backported. I was not able to come up with a sources.list that matched what was installed and what I wanted to install. This means you get a lot of conflicts. You can resolve those, but then you end up updating _a lot_ of packages, and loosing most of the extra Lindows packages. It still works, but there really isn't much point in doing it compared to installing a fresh Debian, unless you have a preinstalled Lindows computer without floppy or CD drive.
OpenLindows packages on the other hand matches the installed Lindows packages. I started trying out OpenLindows. I downloaded OO from Openlindows, put it on a CF card, met my girlfriend with the Lindows laptop on the way to a meeting, put the CF card in the laptop (it has a CF slot), apt-get install openoffice, and it was working.
The laptop now runs debian unstable (except for Netscape), but that is just because I could do it. I liked a lot of things in Lindows (not the root thing) and $49/year (per family i think) could be worth it as long as you are not locked in---You have a way out (apt-get) if you decide it is no longer worth it.
Yes. On my K7 1500+ laptop it seemed better and ACPI speedstepping works (although that is still 2.5.72). However it was snappy with 2.4 too.
But I am writing this on an old laptop with a 233 P. classic, using Ion and Dillo so the 64Mbyte RAM is plenty. I am recompiling the kernel and upgrading my Debian at the same time just to test it. It is just flying. I would call it drastic.
I is sort of if I was on the olympic athletics team and finishing last in the 100 meter and then claiming the gold medal because is was very improbable that anyone could run 100 meters in less than 13 seconds without cheating or being doped.
It should be easy to convince any non-runners of this. Have you ever run a 100 meter in 10 seconds? Do you know any who have?
In the Linux case, see for yourself: check out the traffic on the kernel mailing list. Is has been summarized on the kernel-traffic page.
http://kt.zork.net/kernel-traffic/archives.html (KT is "published" every few weeks, so its not something Zack Brown made up after the SCO case)
Try searching for say, SMP. You won't find one post from IBM saying: Hey, just use out UNIX code but don't tell SCO. Instead you will see some very smart people working very hard on improving the Linux kernel. That is how it happened.
BTW: SCO is mentioned on KT: http://kt.zork.net/kernel-traffic/kt20000911_ 84.ht ml#1
Right. In theory other distributions can do the same as apt on Debian. In practice when you ask an admin on a computer with one of the other distributions to just upgrade to the newest version of say Apache, Sane, or XFree, they will look for CD's and give you some mumbo jumbo and ask you to come back later.
The reason is that is easy to upgrade from one version of a distribution to the next version, because that has been tested well.
But debian is the only distribution that has been able to iron out the depency problems by many years of large-scale upgrading (and downgrading) between exotic combitatons of versions of packages.
I installed a computer with debian in 97/98 and it newer saw a debian install CD again. The same debian installation is still running although almost all hardware is replaced by now.
The first years with APT was a nightmare of dependency problems but now it just works.
I guess the companies likes the GPL or the Linux developement model. I can see a couple of reasons.
Suppose IBM spend a billion improving one of the free BSD kernels. Then Oracle could improve a bit more, put their DB, cluster file-system etc, on top to compete with DB2, and sell it all as closed source. I.e. Oracle would gain from IBM's investment, whereas IBM would get nothing from Oracle. Or Microsoft could use some of IBM's BSD code (they already use BSD code).
If IBM develops some important code, they can release it for Linux under GPL and also use it in AIX/DB2/WebSphere thereby maintaing an advantage over Solaris, Oracle etc.
There are arguments for BSD (or LGPL) as well, eg embedded systems. I am sure that companies like IBM and Oracle made an internal evaluation of all this that would be an interesting to read.
cryptography, was:They are already killing the l
on
Replacing SMTP?
·
· Score: 1
But I do get important mail from people I dont know.
Maybe we could improve spam filters by building some kind of probabilistic web-of-nospam-trust.
Or how about we all start to only accept email containing a timestamp and encrypted using our very long public keys. Obviously it will not stop all spam since the public key is public. But collecting say a million public keys and encrypting a spam mail with every key would cost the spammers more time and money. And whereas the 1/10 cent per email tactic has some bad side-effects, having all mails sent to you encrypted would increase your privacy.
> there is simply no point giving LD a horizontal resolution
But that does not mean that it is infinite. It's just undefined.
> doing so is a complete insult to the very idea of an analog signal.
Nyquist's theorem does make a comparison possible.
>The fact is that with an analog signal, with better technology the signal can be improved to any point you like by improving the signal to noise ratio.
Not really. The resolution (actually capacity accounting for different playing times) of DVD's and LD are limited by the same factors like the frequency of the laser.
>A sufficiently advanced analog device will always be better than a digital one, but far harder to design, and normally more expensive to record, so digital is preferred for its simplicity
Or I could prase it different. A digital device will be better than an analogue in the same price range.
For instance it might be possible to make vinyl records with the 96dB Signal/Noise ration of CD's, maybe even the 1000dB you can get with FLAC files on DVDR. But no one could afford the recorders or players.
>Now, when DVD gets more scanlines than NTSC video, we can reconsider this.
Payment, stolen cards etc:
When I am in the US no physical store has ever refused any of my danish credit cards because they could be stolen etc. Why would a web-store?
Warranties and liabilities: Same thing. I have bought a lot of stuff in other countries. No one has ever refused to sell me anything because I would be leaving the country the next day, thereby making support difficult. IANAL but if I buy something on a US web-page the company would have the same liabilties as if I had sent a courier to buy it a their store.
I do speak english, but I am sure that I could go into any US computer store with danish credit card, speaking hungarian using Monty Pythons phrase book and buying a computer without anyone worring about a possible language barrier.
However I have bought a lot of stuff from US web-stores (GPS, electronics, books, MC gear). I just wish stores would put their terms on the main page. I consider it impolite that they let you fill your shopping basket and then refuse to let you buy it.
This goes not only for shipping out of the US but also for example shipping to a different address than the credit card address. I occasionally online buy presents for friends in other countries and naturally want to avoid unnecessary shipping cost, shipping time, and possibly taxes.
My first buy from a US web store was a GPS from http://www.vitelectronics.com/ in 1996. I see they are still "shipping worldwide". In fact their site still has the same classic 90'es look:-)
I did upgrade Lindows (v3) using apt. (http://www.agol.dk/elgaard/lindows.html) It was a bit messy, mostly because of gcc 2.95/gcc3.2 issues in debian KDE. If I had found a Woody KDE3.0 source, I am sure it would not have been messy at all, but I went for KDE 3.1.2 in unstable.
But no personal data was lost, not even configuration data. All Lindows-specific software is installed in Debian packages, so they can be removed in an unmessy way, but not upgraded with apt.
Upgrading with apt from http://www.openlindows.com/ is easier, but less interesting:-)
And MS i raising the stake. Now it not just boring documents that does not look perfekt in OO. Soon you might have to ditch Britney Spears and go to http://www.opsound.org/ and http://openmusic.linuxtag.org/ instead.
It will be interesting to see how many of the established artist will buy into this in the long run.
Slashdot Mirror
How is it not really Linux in a sense that we know it? It certainly seemed like Linux to me.
It is a Debian with apt-get.
"Now then, over the years, exceptions have been found. Shouting "FIRE!" in a crowded room is not protected speech. Urging an assembled mob to violence is not protected speech. Openly plotting to kill your leader is not protected speech"
Or to use a more common example: on US TV offensive speech is replaced with a beep.
You are lucky.
I got 10200 the last 4 days.
Thats one every 35 seconds on average.
Or about half a Gigabyte a day.
>but there is no way of researching and proffiting from something like MP3
...
Xiph/OGG seen to be doing good research without patents, although they might not be profiting as much as Frauenhofer.
>Or is there anythign trivial, obvious or allready state of the art in them?
Not in MP3, but some of the patents are very broad.
However I was referring to amendmend 11:
>Member States shall ensure that a computer-implemented
>invention may be claimed only as a product, that is as a
>programmed device, or as a technical production process.
If this passes _software_ players end encoders should be ok.
>at least educate your self about the question, what is HARD computer science
But I don't think hard science should be patentable whether it is physics or computer science.
The amendment in english are here (MS Word format)
>In any case, in my view of the world, this
>directive is a step forward from the current
>situation.
Well, lets see if eg. the MP3 and MPEG patents for software encoders and players fall in Europe if this amendmend passes.
I'll still use Ogg, but it's a goot litmus test.
--
Niels
>I think the pint is pretty much a standard for beer the world over.
It might be a standard, but it is not very well defined. I was in Portugal this summer. I was not too happy about until i found out that in many places you could buy a "small pint" or "big pint" of beer.
Many places ordering a pint will get you 0.5 liter
>Besides, they ship just about the only laptops you can buy that come with Perl pre-installed. ;-)
The Lindows laptop has Perl too.
1. AirSnort, already working.
8 /29/232022 8
2. Something like the MIT rootnet.
http://slashdot.org/article.pl?sid=03/0
By having the routing in the AP, it would participate in the roofnet without a server. Servers use more electricity and are noisy.
3. Security for open accesspoints.
You might leave our accesspoints open to share it with others. But you might want some extra security:
a) Block port 25 for others so they don't send spam and get you blacklistet.
b) Some VPN/SSH tunnels for privacy.
c)Traffic shaping. You would like to reserve some bandwidth for yourself.
They are not shutting MS out.
MS is a software company and I am sure MS can make a bid for contracts on this new system if _they want_ to.
They might even win a contract if they are competitive.
Check out:
http://www.ibm.com/developerworks/linux/
and see "bigiron" stuff like accessing USB devices from java, python, wireless accesspoints, image manipulation, mini-libc for memory-constrained systems, smalltalk, XML, bluetooth, and printer drivers..
Normal Lindows users don't apt-get. CNR is simper to use than synaptic, kpackage, dselect, apt-get etc.
Technically the answer is versions and dependencies. I experienced Lindows 3 preinstalled on the laptop (http://www.agol.dk/elgaard/lindows.html). The Lindows specific programs were all cleanly installed as Debian packages, which make upgrading with apt safe and simpler. However the installed packages were an interesting mix of Debian/KDE/OO versions. Many programs were backported. I was not able to come up with a sources.list that matched what was installed and what I wanted to install. This means you get a lot of conflicts. You can resolve those, but then you end up updating _a lot_ of packages, and loosing most of the extra Lindows packages. It still works, but there really isn't much point in doing it compared to installing a fresh Debian, unless you have a preinstalled Lindows computer without floppy or CD drive.
OpenLindows packages on the other hand matches the installed Lindows packages. I started trying out OpenLindows. I downloaded OO from Openlindows, put it on a CF card, met my girlfriend with the Lindows laptop on the way to a meeting, put the CF card in the laptop (it has a CF slot), apt-get install openoffice, and it was working.
The laptop now runs debian unstable (except for Netscape), but that is just because I could do it. I liked a lot of things in Lindows (not the root thing) and $49/year (per family i think) could be worth it as long as you are not locked in---You have a way out (apt-get) if you decide it is no longer worth it.
Yes.
On my K7 1500+ laptop it seemed better and ACPI speedstepping works (although that is still 2.5.72). However it was snappy with 2.4 too.
But I am writing this on an old laptop with a 233 P. classic, using Ion and Dillo so the 64Mbyte RAM is plenty. I am recompiling the kernel and upgrading my Debian at the same time just to test it.
It is just flying. I would call it drastic.
I guess this old laptop resembles a PDA nowadays.
I is sort of if I was on the olympic athletics team and finishing last in the 100 meter and then claiming the gold medal because is was very improbable that anyone could run 100 meters in less than 13 seconds without cheating or being doped.
_ 84.ht ml#1
It should be easy to convince any non-runners of this. Have you ever run a 100 meter in 10 seconds? Do you know any who have?
In the Linux case, see for yourself: check out the traffic on the kernel mailing list. Is has been summarized on the kernel-traffic page.
http://kt.zork.net/kernel-traffic/archives.html
(KT is "published" every few weeks, so its not something Zack Brown made up after the SCO case)
Try searching for say, SMP. You won't find one post from IBM saying: Hey, just use out UNIX code but don't tell SCO.
Instead you will see some very smart people working very hard on improving the Linux kernel. That is how it happened.
BTW: SCO is mentioned on KT:
http://kt.zork.net/kernel-traffic/kt20000911
Sorry,
s/distribution/linux distribution/
And gentoo haven't even existed "for many years" - yet.
Right.
In theory other distributions can do the same as apt on Debian.
In practice when you ask an admin on a computer with one of the other distributions to just upgrade to the newest version of say Apache, Sane, or XFree, they will look for CD's and give you some mumbo jumbo and ask you to come back later.
The reason is that is easy to upgrade from one version of a distribution to the next version, because that has been tested well.
But debian is the only distribution that has been able to iron out the depency problems by many years of large-scale upgrading (and downgrading) between exotic combitatons of versions of packages.
I installed a computer with debian in 97/98 and it newer saw a debian install CD again. The same debian installation is still running although almost all hardware is replaced by now.
The first years with APT was a nightmare of dependency problems but now it just works.
Apparantly there are devices running Windows in Spain too, although I am not sure about the version.
:-)
http://www.agol.dk/elgaard/win_extramadura1.jpg
I thought they were running Linux in Extramadura
Yes, Sun, Oracle etccan implement the same ideas in their own products. But they cannot reuse the code if it is GPL.
If they just implement the same ideas, they do not have to distribute their modifications to IBM.
I guess the companies likes the GPL or the Linux developement model. I can see a couple of reasons.
Suppose IBM spend a billion improving one of the free BSD kernels. Then Oracle could improve a bit more, put their DB, cluster file-system etc, on top to compete with DB2, and sell it all as closed source. I.e. Oracle would gain from IBM's investment, whereas IBM would get nothing from Oracle. Or Microsoft could use some of IBM's BSD code (they already use BSD code).
If IBM develops some important code, they can release it for Linux under GPL and also use it in AIX/DB2/WebSphere thereby maintaing an advantage over Solaris, Oracle etc.
There are arguments for BSD (or LGPL) as well, eg embedded systems. I am sure that companies like IBM and Oracle made an internal evaluation of all this that would be an interesting to read.
But I do get important mail from people I dont know.
Maybe we could improve spam filters by building some kind of probabilistic web-of-nospam-trust.
Or how about we all start to only accept email containing a timestamp and encrypted using our very long public keys. Obviously it will not stop all spam since the public key is public.
But collecting say a million public keys and encrypting a spam mail with every key would cost the spammers more time and money.
And whereas the 1/10 cent per email tactic has some bad side-effects, having all mails sent to you encrypted would increase your privacy.
> there is simply no point giving LD a horizontal resolution
:-)
But that does not mean that it is infinite. It's just undefined.
> doing so is a complete insult to the very idea of an analog signal.
Nyquist's theorem does make a comparison possible.
>The fact is that with an analog signal, with better technology the signal can be improved to any point you like by improving the signal to noise ratio.
Not really. The resolution (actually capacity accounting for different playing times) of DVD's and LD are limited by the same factors like the frequency of the laser.
>A sufficiently advanced analog device will always be better than a digital one, but far harder to design, and normally more expensive to record, so digital is preferred for its simplicity
Or I could prase it different. A digital device will be better than an analogue in the same price range.
For instance it might be possible to make vinyl records with the 96dB Signal/Noise ration of CD's, maybe even the 1000dB you can get with FLAC files on DVDR.
But no one could afford the recorders or players.
>Now, when DVD gets more scanlines than NTSC video, we can reconsider this.
They could make DVD's the physical of LD's
>Laserdisc stores infinite horizontal resolution,
o mpari son/
k Like.ht ml
It is analogue. That does not make it infinite.
According to:
http://www.cs.tut.fi/~leopold/Ld/ResolutionC
laserdiscs are 560x360 which is worse than DVD's.
Also check the scanning tunnel microscope picures on that page.
http://www.cs.tut.fi/~leopold/Ld/HowLDsLoo
Payment, stolen cards etc:
:-)
When I am in the US no physical store has ever refused any of my danish credit cards because they could be stolen etc. Why would a web-store?
Warranties and liabilities: Same thing. I have bought a lot of stuff in other countries. No one has ever refused to sell me anything because I would be leaving the country the next day, thereby making support difficult.
IANAL but if I buy something on a US web-page the company would have the same liabilties as if I had sent a courier to buy it a their store.
I do speak english, but I am sure that I could go into any US computer store with danish credit card, speaking hungarian using Monty Pythons phrase book and buying a computer without anyone worring about a possible language barrier.
However I have bought a lot of stuff from US web-stores (GPS, electronics, books, MC gear).
I just wish stores would put their terms on the main page. I consider it impolite that they let you fill your shopping basket and then refuse to let you buy it.
This goes not only for shipping out of the US but also for example shipping to a different address than the credit card address. I occasionally online buy presents for friends in other countries and naturally want to avoid unnecessary shipping cost, shipping time, and possibly taxes.
My first buy from a US web store was a GPS from http://www.vitelectronics.com/ in 1996. I see they are still "shipping worldwide". In fact their site still has the same classic 90'es look
Actually not.
:-)
I did upgrade Lindows (v3) using apt. (http://www.agol.dk/elgaard/lindows.html)
It was a bit messy, mostly because of gcc 2.95/gcc3.2 issues in debian KDE. If I had found a Woody KDE3.0 source, I am sure it would not have been messy at all, but I went for KDE 3.1.2 in unstable.
But no personal data was lost, not even configuration data. All Lindows-specific software is installed in Debian packages, so they can be removed in an unmessy way, but not upgraded with apt.
Upgrading with apt from
http://www.openlindows.com/
is easier, but less interesting
And MS i raising the stake.
Now it not just boring documents that does not look perfekt in OO. Soon you might have to ditch
Britney Spears and go to
http://www.opsound.org/ and http://openmusic.linuxtag.org/ instead.
It will be interesting to see how many of the established artist will buy into this in the long run.