So by this definition, the Chinese citizens are ACTUALLY freer than US and other westernized countries, because its all about the welfare of the country as opposed to personal freedoms.
No, as I said they're different kind of freedoms, one is not necessarily more free than the other. Which one you consider more free is simply based on which kind of freedom you value most. Besides, that may be the theory behind communism, but just like with capitalism (and any economical model or ideology), the practice is often significantly different from the theory.
What would Chairman Stallman have to say about this? Actually, I think this is what he wants...he's a modern day communist hippy thinking he's going to change the world by taking away choice.
You take some, you give some. It's the same as always. And the end result may or may not be more personal freedom, depending on what you want to do.
I mean, let me get this straight: GNU Darwin is the version of Darwin that the GNU project doesn't recommend?
GNU Darwin is simply Darwin with all BSD-lincensed programs (such as curl) replaced with GPL ones (wget) and a lot of extra programs (only GPL licensed ones obviously). It's still not a "pure" GPL system, as the kernel, most kernel extensions and probably some libraries will always be APSL (otherwise it wouldn't be Darwin anymore).
I think it's simplest to say that the GPL is about the freedom of the code (i.e., once the code is Free, it has to stay Free) and as such the freedom of the community at large (since this means that over time, they will almost by definition get a larger choice), while the BSD license is more about the personal freedom of users/companies (to do whatever they want with the code). Imho, it's not about which one is more Free than the other, they simply focus on different (contradicting) freedoms: the freedom of the individual vs. the freedom of the community.
Apple already has a fork. The gcc that comes with Mac OS X has some 30+ optimizations (both PPC-specific and generic) which aren't in FSF gcc, because FSF refused the patches. Apple does try to keep that kind of stuff at a minimum though, since it means more maintenance work for them as they have to merge those patches with every new gcc version and update them if necessary.
In Europe, a US based company already has the patent for this "invention". I hope we can use this situation to convince more politicians of the dangers of software patents.
I completely agree with you. The FFII unfortunately didn't exist yet when this directive went through the European Parliament and there was (almost?) no organised lobbying or protest. Now the FFII does exist however and is actively working on preventing software patents from being legalised in Europe.
I personally went to the EP for 3 days last week and there really are an incredible amount of MEPs who did not have any idea at all about how bad the current proposal is and who became quite supportive of us after we explained the situation to them. Informing politicians is not the job of the media, but largely of their assistants. Informing the assistants is a job of lobbying groups and individuals.
If you don't contact any politicians or their assistants, you cannot expect them to hear your concerns (although it would be really nice, of course). They vote on about 60 dossiers per week in the EP, so they just don't have the time to go actively looking for information about each and every one of those. Each party has its own specialist for different subjects. The specialists from each faction then create a voting list together, which more often than not is followed blindly by the rest of the faction (since they don't know anything about the dossier).
This is obviously not an ideal situation, since those specialists often have an agenda to push. So informing as many politicians as possible about your point of view is indeed what is necessary. Some will refer you to their local specialist, but most certainly not all of them.
Except the point made by TackHead was that *all* ISPs should block outbound port 25.
That is no problem if you set the smtp servers of your ISP as smarthosts for your system (unless they are broken like the Verizon ones supposedly are, in which case you should switch ISP's as the grandparent said, since the current one is selling you a broken service)
Just porting carbonlib wouldn't cut it in most cases. Almost all Mac apps also use Quicktime. For sound, they sometimes use CoreAudio under Mac OS X. Most games use the HID Manager for input... So I'm not sure whether just carbonlib would help a lot.
MS Office is a Carbon app. There's no carbonlib for any *nix except Mac OS X out there. Porting it to Linux would mean either porting carbonlib (i.e. a cleaned up version of the legacy Mac OS api + extensions) to Linux, or rewriting it from the ground up (just like the Mac OS X version is almost a complete rewrite compared to the Windows version).
However, 10.2 has IPSec support, just no front-end.
If you have a look at the lower right part of this page (the "Works Well with Windows" part), you get your answerw: yes, it's included in Internet Connect and they've also added support for RSA's SecureID-based vpn's/
Is X11 still a stand alone application in Panther or is more integrated with the OS?
I've read or heard somewhere you'll be able to simply double-clock X11 apps to launch them, but I don't remember where anymore (so take this with a grain of salt).
The Apple Panther page says "support for popular Linux APIs". Any indication of what this means?
They will include more programs and libraries which are standard in most Linux distributions so porting of apps and users will be easier.
Is NetInfo still used as the centralized database for all OS resources or have they finally replaced it with LDAP?
I don't have the slightest idea, but I don't see any reason why they would do that (except if they didn't know what else to do and wanted some extra work).
I don't know about you guys, but protests aren't working, letters aren't working, e-mails aren't working. Voting is not working, propaganda isn't working. There's only 1 alternative after peacful protest; violent protest and our leaders are too dumb to realize that if they piss enough people off, they are dead meat literally.
Are you completely out of your mind? Actually, what apparently isn't working is the informing of the media, because the article at the Register is factually wrong. First of all, the European parliament voted (unanimously!) last Thursday (2003/06/16) against rushing the proposal through, so there is no vote today on it. Secondly, a lot of MEP's are *against* the software patents proposal. I don't think the author of the article bothered to contact any politician, she just enumerated a couple of popular preconceptions about politicians.
I do have contacted my (Flemish) MEPs and it turns out all Flemish parties but one are against software patents (only the liberals haven't chosen a side yet). Arlene McCarthy is starting to get strong opposition from within her own (socialist) faction... Although I by no means want to suggest the fight is over or won, the article at the Register is spreading a lot of uninformed FUD imho.
It does make some good points, but it's by no means accurate about how a lot of politicians think about this issue, unless only the Belgian factions are against and the representatives from all other European countries are pro, but I doubt that... Most of the times, these factions try to take a common stand.
PS: I'm not politically active (except by writing an email - in Dutch - to the Flemish MEPs on this issue) or tied to one or other party in any way.
I would be interested if you could find even one reference to back up this ideas that you patent "implementations" not "ideas".
Maybe application would be correcter/more accurate to describe what I meant. I wanted to say that a taditional patent (in Europe) is generally granted for the application of some idea (e.g. of certain specific chemicals to a certain virus or bacteria, to use your example), and not for the general idea itself (using chemicals to kill viruses or bacteria). That's what I meant with implementation: remove the abstraction, move to something concrete/specific.
Going from an algorithm to a software "implementation" doesn't remove the abstraction, since it's a plain translation from one language to another. So if an algorithm is abstract (which I think everyone agrees on), then so is its translation in software.
After all, why should an idea written in C be patentable in that case, and one in English or some mathematical notation not?
Patents generally are written in English (in America at least). You submit some English and a few diagrams and you get a patent. You would seldom submit C code.
Here, I simply wanted to show that there is no distinction between allowing patents on software (written in e.g. C) and on algorithms in general (written in e.g. English or a mathematical notation).
And a physical device is just exploiting the laws of physics that everyone should be free to use.
You cannot patent simply an effect of the laws of physics. You can patent a machine, which will obviously use the laws of physics, but another machine which uses the same effect will not be covered by the patent protecting that first machine. These are not patents on ideas, these are patents on machines or parts of machines.
The goal is to ensure those that find things that were unlikely to have been found anyway are rewarded. Thereby encouraging both their development, and the distribution of that knowledge.
But as a lot of studies have shown, it doesn't work that way with software. Read the introduction and the conclusions of the paper I linked to in my previous post if you want to know why.
The current system is broken. It needs fixing. Things that are obvious are patented, statements of the problem are patented.
Actually, the current system works quite well, thank you. You're confusing things with the situation in the US. Software patents are not allowed in Europe currently. The European patent office is granting them anyway (since more patents granted means more money for them), but they are not enforceable.
But if you eliminate all protection for complex algorithm development then you'll discourage their development. A small company will be unable to deploy a new technique and recoup its investment if large companies can simply reverse engineer the solution and then deploy it on a large scale.
As I said, there is no software patent protection in Europe currently. Nevertheless, the software industry is doing quite well here. In fact, we have mainly small companies here, even though Microsoft, IBM etc are quite well represented here.
And as I said in my previous post: a small company does not get any actual protection from big companies with patents. Read this text about how small companies are "protected" from IBM by having software patents, if you don't agree with the reasoning in my post. So in fact, with software patents, you actually save big companies some time: they don't even have to reverse engineer your code anymore.
Finally, spending time on reverse engineering isn't nearly as efficient as just doing something new from scratch. Extracting an algorithm from compiler-optimised machine code is not something you want to do a lot. And afterwards, you still have to program and debug it again. And once you've done that, you've lost a lot of time that the original coders could use to innovate their product. Once you finally get to market, the technology is probably already outdated... And if you'd somehow manage to directly use the disassembly so you don't have to implement it again, I wouldn't want to be the one having to maintain or improve that software.
Software being disquised as hardware.
Then the patent is on the hardware+software combo as a whole. If you use the same software for something else, the patent is not enforceable.
Research being financed exclusively by large companies that knew they could afford to bring the resulting products to market themselves.
On the contrary, if there are no software patents, a small company can develop a program without fears that its investment will be voided when it finally goes to market, because some big company may suddenly appear and tell them they violate such and such patent (even though it may be a very innovative program for the rest). With software patents, only a big company is guaranteed to be able to force a cross-licensing deal for the programs it writes.
The research that was developed being kept as trade secrets.
And there is the problem! Sorry, but party leaders should actually talk to the arbitary people.
And failing that, arbitrary people should talk to the party leaders/MEPS. Too many people think of politics as a "far from my bed show" (literal translation from Dutch expression, I'm not sure if it makes any sense in English:). They think that voting once every so many years is doing their political duty. For the rest, they often simply complain to each other about how bad the politicians are doing. Trying to do something about it? Hey, the politicians were elected to do that for them and get paid (quite well) for that, so why would other people have to help them doing their job?
That kind of attitude doesn't help. Voting is a multiple choice system: you vote for the people who are thinking mostly like you, but you'll never find someone who wants all the exact things that you want. So apart from voting, telling those politicians what you think about specific issues (and substantiating your thoughts with sound reasonings and facts) is the only way to get your point across. Of course, no one can force "you" (not directed at you personally, of course) to do that, but then "you" have little right of complaining imho.
I also don't think you can compare politics to running a business (at least not completely). A business has one goal: making profit. One important factor in that is having happy customers. However, IKEA only makes furniture. Politics is about several issues a week. There is no "politics customer service line" or a "politics shop" (although you'd doubt that from time to time:) where a politician can spend just one week a year to get in touch with the "regular people" that are more or less directly involved with all decisions (s)he has to make that year.
The issue is not whether a patent is for "software" or not. The same algorithm can be developed either in software or in hardware.
A specific algorithm that represents a truly novel solution to a problem should be patentable, no matter how it is implemented.
The way you write an algorithm down in software is already protected by copyright law. A machine (hardware implementation) is not protected by copyright law and generally requires a lot of investment to create, which is why the patent system was created. Patents were never intended to protect ideas (an algorithm is also an idea), but to protect implementations (and investments done to create that implementation.
However, software isn't an implementation, it's a notation in a language that ultimately can be understood by a computer (probably after some more translation by a compiler). If you allow patents on software, you allow patents on ideas. After all, why should an idea written in C be patentable in that case, and one in English or some mathematical notation not? It also renders copyright on software completely useless. What good does it do if you own the copyright on a program if it's also covered by 20 patents you don't own?
It's the same as if there would be patents on plot-elements in books... It's not because you're writing the umpteenth book about a serial killer that you're ripping off all authors that wrote about a serial killer before. If you're downright plagiarising someone, then that other person can already defend himself using the copyright protection he got for free.
The goal of patents is to promote innovation by protecting the investment of entrepreneurs. If you carry over this protection to plain ideas, you are actually discouraging other innovators. After all, innovators are bound to reuse ideas other people have had before when doing something new (using RMS' analogy one more time: just like Beethoven didn't invent music from scratch and yet was very original, there's no programmer who can reinvent informatics from scratch - no matter how good he is).
Additionally, several studies have shown that allowing software patents hampers advances in software development. The reason is that companies start investing part of their R&D budget in obtaining patents instead of innovating more. Of course, they want to recoup that. Since big software companies have more patents than small ones, they can usually force cross-licensing deals with those smaller ones or make them pay. So the end result is that:
Big companies still get access to all advances in software development, but spend less on innovation themselves (see this study)
Small companies can do less innovation, because they are forced to buy "patent protection" from large companies. They also have to invest money getting patents of themselves, so that they can negotiate lower licensing fees from big companies. That's all money which cannot be spend on innovating.
The winners are patent bureaus, big companies and patent lawyers. The losers are smaller companies and society as a whole, since less money is spent on innovation.
As someone alse already posted, this attempt at fast-tracking the proposal through parliament as failed miserably. The real news is that even McCarthy's own (socialist) faction is becoming more and more opposed to software patents. I'm in direct email contact with several assistants of the Flemish MEPS from the Greens and socialist faction (since those were the ones that responded to my initial mail (in Dutch) and they all are completely against software patents. In fact, almost all Flemish parties are against.
One of those assistants told me he's never seen such an enormous amount of public attention for a proposal in the two years that he has worked at the European parliament. He thinks there's actually a very good chance of preventing this proposal from getting approved. Really, it's easy to say "all politicians are alike" and "corporations own the politicians anyway" etc, but that's simply not true (note: I'm not a member of any political party nor politically active, except in cases like this). Yes Virginia, there still are a lot of people with a conscience in politics who want to do the best for society at large, they just need access to the right information. In cases like this, people like us can make the difference.
If a non-programmer or non-ip-lawyer reads a proposal like McCarthy's, I can perfectly imagine that it's not that difficult for that person to be convinced that she's indeed trying to protect the software development community at large. The background text of her proposal is really full of misleading and sometimes outright wrong statements to justify her goals.
For example, she cites one study which shows that software patents are beneficial to small and medium-sized companies. In the same footnote, she states that they also looked at several other studies, however, at least one of those concludes exactly the opposite. Nevertheless, the way it is put forth in her text, it seems as if all those studies show exactly the same results. There really are a lot of things like that...
Here in Europe, the Socialist and Conservative leadership were working together to try to push the software patents through on the fast track. It makes me think maybe they're becoming the same party.
FWIW, the Belgian (both the Flemish and the Walloon) socialists have been against this proposal and the fast-tracking from the start, so I think you're generalising a bit too much.
Thanks, I missed that during my quick look. I did think I had seen it somewhere, but I couldn't find it anymore. Anyway, as you say, those two definitions completely void the extra requirement of making a "technical contribution".
I don't know whether there is such a thing as a recall vote in the European parliament (I don't even know what a recall vote is in the US). Regarding the socialist/conservatives: there are a lot more political factions than just socialists and conservatives. There's at least socialists, Christians, greens, nationalists and liberals. Greens and socialists are generally progressive, the rest conservative, but it's certainly not always that simple.
From what I hear though, there is little to fear. The EU patents on software won't allow any patenting of business methods
That's what McCarthy wants you to believe, yes. However, her amendments say otherwise:
Accordingly, a computer-implemented
business method or other method in
which the only contribution to the state of
the art is non-technical cannot constitute
a patentable invention.
Now, whether or not something is "technical" is defined as whether or not something makes a "technical contribution". However, technical contribution is nowhere defined in the proposal! It only says:
The technical contribution shall be
assessed by considering the state of the art
and the scope of the patent claim
considered as a whole, which must
comprise technical features, irrespective
whether or not such features are
accompanied by non-technical features.
Now, that's really clear, isn't it? Surely, this non-definition won't be abused by anyone. Regarding your one-click example: the commission is not even sure itself whether or not it would be possible in the new proposal (see this FAQ, search for click).
Anyway, things are looking actually quite good currently. Have a look at this press release, most MEPS are finally seeing that the proposal is completely wrong. I'm in direct contact with several people working with the MEPS and these people are really eager to learn as much as possible about the dangers of software patents.
You do realize that computers have shared RAM, right? They can't access it at the same time.
No, but a processor does a lot of other things besides accessing RAM, especially on a register-rich architecture like the PowerPC. It's true that there will be contention for RAM and other resources, but not so much that it'll nullify most of your gains. After all, each processor still has its private L1 and L2 caches.
Also, since the system controller provides point-to-point connections from the processors to RAM (and to other I/O "devices"), there will never be contention for the bus. I have to be honest that I don't know whether the fact that they're using DDR RAM means that both processors can get something from RAM in the same bus cycle, or that it only allows you to get a double data rate in the context of one transaction. I assume it's the latter.
They also have shared hard drives, video memory, and pretty much everything else. This extra overhead pretty much kills any speed advantage you may get from 2 procs.
Unless your system is memory starved or if you are working on huge data sets, the hard drive isn't accessed all that much on modern *nix systems. Also, under Mac OS X an application normally doesn't access video memory directly. All windows are double buffered by default and only the window manager/server compose all these windows and flush it to the screen (with the windows treated as OpenGL textures).
Also, for many applications, even on dual processor machines, single processor performance is key. Very few applications (certainly very few desktop applications) are designed (or even can be designed) to take advantage of multiple processors.
First of all, that's exactly why we have multitasking OS'es, so that even though one program may not be able to saturate your processor(s), you can run many at the same time. Additionally, even if your processor isn't fully taxed, having two processors and a multitasking OS can improve responsiveness. After all, if XCode is compiling in the background, Mail getting new mail and checking it for spam, a couple of browser windows with animated gifs or Java applets are open, some terminals are showing some data (e.g. top), iTunes is playing some music etc, all these things essentially happen at the same time.
As long as no two processes want to do something at the same time, two processors indeed won't gain you anything. After that, it depends on when the delays become perceptible for the user (which may happen only every now and then or very often, depending on the speed of the cpu). Further, your claim that most consumer apps don't take advantage of multiple processors, is incorrect. Most applications for Mac OS X are multithreaded nowadays. The last column in the following partial top output are the number of threads that process has (sorry for the formatting, slashdot doesn't accept PRE tags):
The processes are sorted by cpu usage. As you can see, most programs have more than one thread (although since my system is pretty idle currently, most are simply blocked waiting for input). The kernel alone has already 26 of them... Some of them also spawn extra threads when they're asked to do something (like lookupd).
I do agree having a second processor generally won't speed up things like surfing, unless you're the type that continuously opens several sites at the same time, possibly with flash and java applets etc. However, the "snappiness" of a system can increase a lot...
That's not entirely true, the motherboard/system controller is proprietary as well. So you still need a motherboard manufacturer. There is at least one company doing this, but they're not exactly producing leading edge technology.
Slashdot Mirror
I think it's simplest to say that the GPL is about the freedom of the code (i.e., once the code is Free, it has to stay Free) and as such the freedom of the community at large (since this means that over time, they will almost by definition get a larger choice), while the BSD license is more about the personal freedom of users/companies (to do whatever they want with the code). Imho, it's not about which one is more Free than the other, they simply focus on different (contradicting) freedoms: the freedom of the individual vs. the freedom of the community.
Apple already has a fork. The gcc that comes with Mac OS X has some 30+ optimizations (both PPC-specific and generic) which aren't in FSF gcc, because FSF refused the patches. Apple does try to keep that kind of stuff at a minimum though, since it means more maintenance work for them as they have to merge those patches with every new gcc version and update them if necessary.
In Europe, a US based company already has the patent for this "invention". I hope we can use this situation to convince more politicians of the dangers of software patents.
I personally went to the EP for 3 days last week and there really are an incredible amount of MEPs who did not have any idea at all about how bad the current proposal is and who became quite supportive of us after we explained the situation to them. Informing politicians is not the job of the media, but largely of their assistants. Informing the assistants is a job of lobbying groups and individuals.
If you don't contact any politicians or their assistants, you cannot expect them to hear your concerns (although it would be really nice, of course). They vote on about 60 dossiers per week in the EP, so they just don't have the time to go actively looking for information about each and every one of those. Each party has its own specialist for different subjects. The specialists from each faction then create a voting list together, which more often than not is followed blindly by the rest of the faction (since they don't know anything about the dossier).
This is obviously not an ideal situation, since those specialists often have an agenda to push. So informing as many politicians as possible about your point of view is indeed what is necessary. Some will refer you to their local specialist, but most certainly not all of them.
Just porting carbonlib wouldn't cut it in most cases. Almost all Mac apps also use Quicktime. For sound, they sometimes use CoreAudio under Mac OS X. Most games use the HID Manager for input... So I'm not sure whether just carbonlib would help a lot.
MS Office is a Carbon app. There's no carbonlib for any *nix except Mac OS X out there. Porting it to Linux would mean either porting carbonlib (i.e. a cleaned up version of the legacy Mac OS api + extensions) to Linux, or rewriting it from the ground up (just like the Mac OS X version is almost a complete rewrite compared to the Windows version).
Shoot you down ;) They did use (almost) the same motherboard in the low-end G4 as what could be found in the Blue&White G3's.
I do have contacted my (Flemish) MEPs and it turns out all Flemish parties but one are against software patents (only the liberals haven't chosen a side yet). Arlene McCarthy is starting to get strong opposition from within her own (socialist) faction... Although I by no means want to suggest the fight is over or won, the article at the Register is spreading a lot of uninformed FUD imho.
It does make some good points, but it's by no means accurate about how a lot of politicians think about this issue, unless only the Belgian factions are against and the representatives from all other European countries are pro, but I doubt that... Most of the times, these factions try to take a common stand.
PS: I'm not politically active (except by writing an email - in Dutch - to the Flemish MEPs on this issue) or tied to one or other party in any way.
Going from an algorithm to a software "implementation" doesn't remove the abstraction, since it's a plain translation from one language to another. So if an algorithm is abstract (which I think everyone agrees on), then so is its translation in software.
Here, I simply wanted to show that there is no distinction between allowing patents on software (written in e.g. C) and on algorithms in general (written in e.g. English or a mathematical notation).You cannot patent simply an effect of the laws of physics. You can patent a machine, which will obviously use the laws of physics, but another machine which uses the same effect will not be covered by the patent protecting that first machine. These are not patents on ideas, these are patents on machines or parts of machines.
But as a lot of studies have shown, it doesn't work that way with software. Read the introduction and the conclusions of the paper I linked to in my previous post if you want to know why.
Actually, the current system works quite well, thank you. You're confusing things with the situation in the US. Software patents are not allowed in Europe currently. The European patent office is granting them anyway (since more patents granted means more money for them), but they are not enforceable.
As I said, there is no software patent protection in Europe currently. Nevertheless, the software industry is doing quite well here. In fact, we have mainly small companies here, even though Microsoft, IBM etc are quite well represented here.
And as I said in my previous post: a small company does not get any actual protection from big companies with patents. Read this text about how small companies are "protected" from IBM by having software patents, if you don't agree with the reasoning in my post. So in fact, with software patents, you actually save big companies some time: they don't even have to reverse engineer your code anymore.
Finally, spending time on reverse engineering isn't nearly as efficient as just doing something new from scratch. Extracting an algorithm from compiler-optimised machine code is not something you want to do a lot. And afterwards, you still have to program and debug it again. And once you've done that, you've lost a lot of time that the original coders could use to innovate their product. Once you finally get to market, the technology is probably already outdated... And if you'd somehow manage to directly use the disassembly so you don't have to implement it again, I wouldn't want to be the one having to maintain or improve that software.
Then the patent is on the hardware+software combo as a whole. If you use the same software for something else, the patent is not enforceable.
On the contrary, if there are no software patents, a small company can develop a program without fears that its investment will be voided when it finally goes to market, because some big company may suddenly appear and tell them they violate such and such patent (even though it may be a very innovative program for the rest). With software patents, only a big company is guaranteed to be able to force a cross-licensing deal for the programs it writes.
This is the only positive side I can see, to b
MEPS *are* elected.
That kind of attitude doesn't help. Voting is a multiple choice system: you vote for the people who are thinking mostly like you, but you'll never find someone who wants all the exact things that you want. So apart from voting, telling those politicians what you think about specific issues (and substantiating your thoughts with sound reasonings and facts) is the only way to get your point across. Of course, no one can force "you" (not directed at you personally, of course) to do that, but then "you" have little right of complaining imho.
I also don't think you can compare politics to running a business (at least not completely). A business has one goal: making profit. One important factor in that is having happy customers. However, IKEA only makes furniture. Politics is about several issues a week. There is no "politics customer service line" or a "politics shop" (although you'd doubt that from time to time :) where a politician can spend just one week a year to get in touch with the "regular people" that are more or less directly involved with all decisions (s)he has to make that year.
However, software isn't an implementation, it's a notation in a language that ultimately can be understood by a computer (probably after some more translation by a compiler). If you allow patents on software, you allow patents on ideas. After all, why should an idea written in C be patentable in that case, and one in English or some mathematical notation not? It also renders copyright on software completely useless. What good does it do if you own the copyright on a program if it's also covered by 20 patents you don't own?
It's the same as if there would be patents on plot-elements in books... It's not because you're writing the umpteenth book about a serial killer that you're ripping off all authors that wrote about a serial killer before. If you're downright plagiarising someone, then that other person can already defend himself using the copyright protection he got for free.
The goal of patents is to promote innovation by protecting the investment of entrepreneurs. If you carry over this protection to plain ideas, you are actually discouraging other innovators. After all, innovators are bound to reuse ideas other people have had before when doing something new (using RMS' analogy one more time: just like Beethoven didn't invent music from scratch and yet was very original, there's no programmer who can reinvent informatics from scratch - no matter how good he is).
Additionally, several studies have shown that allowing software patents hampers advances in software development. The reason is that companies start investing part of their R&D budget in obtaining patents instead of innovating more. Of course, they want to recoup that. Since big software companies have more patents than small ones, they can usually force cross-licensing deals with those smaller ones or make them pay. So the end result is that:
- Big companies still get access to all advances in software development, but spend less on innovation themselves (see this study)
- Small companies can do less innovation, because they are forced to buy "patent protection" from large companies. They also have to invest money getting patents of themselves, so that they can negotiate lower licensing fees from big companies. That's all money which cannot be spend on innovating.
The winners are patent bureaus, big companies and patent lawyers. The losers are smaller companies and society as a whole, since less money is spent on innovation.One of those assistants told me he's never seen such an enormous amount of public attention for a proposal in the two years that he has worked at the European parliament. He thinks there's actually a very good chance of preventing this proposal from getting approved. Really, it's easy to say "all politicians are alike" and "corporations own the politicians anyway" etc, but that's simply not true (note: I'm not a member of any political party nor politically active, except in cases like this). Yes Virginia, there still are a lot of people with a conscience in politics who want to do the best for society at large, they just need access to the right information. In cases like this, people like us can make the difference.
If a non-programmer or non-ip-lawyer reads a proposal like McCarthy's, I can perfectly imagine that it's not that difficult for that person to be convinced that she's indeed trying to protect the software development community at large. The background text of her proposal is really full of misleading and sometimes outright wrong statements to justify her goals.
For example, she cites one study which shows that software patents are beneficial to small and medium-sized companies. In the same footnote, she states that they also looked at several other studies, however, at least one of those concludes exactly the opposite. Nevertheless, the way it is put forth in her text, it seems as if all those studies show exactly the same results. There really are a lot of things like that...
Thanks, I missed that during my quick look. I did think I had seen it somewhere, but I couldn't find it anymore. Anyway, as you say, those two definitions completely void the extra requirement of making a "technical contribution".
I don't know whether there is such a thing as a recall vote in the European parliament (I don't even know what a recall vote is in the US). Regarding the socialist/conservatives: there are a lot more political factions than just socialists and conservatives. There's at least socialists, Christians, greens, nationalists and liberals. Greens and socialists are generally progressive, the rest conservative, but it's certainly not always that simple.
Anyway, things are looking actually quite good currently. Have a look at this press release, most MEPS are finally seeing that the proposal is completely wrong. I'm in direct contact with several people working with the MEPS and these people are really eager to learn as much as possible about the dangers of software patents.
Also, since the system controller provides point-to-point connections from the processors to RAM (and to other I/O "devices"), there will never be contention for the bus. I have to be honest that I don't know whether the fact that they're using DDR RAM means that both processors can get something from RAM in the same bus cycle, or that it only allows you to get a double data rate in the context of one transaction. I assume it's the latter.
Unless your system is memory starved or if you are working on huge data sets, the hard drive isn't accessed all that much on modern *nix systems. Also, under Mac OS X an application normally doesn't access video memory directly. All windows are double buffered by default and only the window manager/server compose all these windows and flush it to the screen (with the windows treated as OpenGL textures).As long as no two processes want to do something at the same time, two processors indeed won't gain you anything. After that, it depends on when the delays become perceptible for the user (which may happen only every now and then or very often, depending on the speed of the cpu). Further, your claim that most consumer apps don't take advantage of multiple processors, is incorrect. Most applications for Mac OS X are multithreaded nowadays. The last column in the following partial top output are the number of threads that process has (sorry for the formatting, slashdot doesn't accept PRE tags):
362 dnetc 68.0% 82:52:25 2
The processes are sorted by cpu usage. As you can see, most programs have more than one thread (although since my system is pretty idle currently, most are simply blocked waiting for input). The kernel alone has already 26 of them... Some of them also spawn extra threads when they're asked to do something (like lookupd).5950 top 12.2% 0:00.53 1
2767 Terminal 5.4% 2:13.61 8
2720 Safari 1.3% 17:17.35 9
0 kernel_tas 0.6% 29:48.20 26
2595 Window Man 0.6% 13:22.23 2
2709 Mail 0.0% 13:11.21 4
170 ATSServer 0.0% 9:36.61 2
2736 TruBlueEnv 0.0% 3:11.28 16
2839 Project Bu 0.0% 2:58.67 3
315 lookupd 0.0% 2:24.75 2
361 cupsd 0.0% 1:37.86 1
105 configd 0.0% 1:25.05 3
2711 UniversalA 0.0% 1:09.67 1
70 update 0.0% 1:08.17 1
2717 SecurityAg 0.0% 1:01.98 2
2705 Finder 0.0% 0:56.42 1
490 slpd 0.0% 0:52.49 8
5602 Xquartz 0.0% 0:46.08 4
339 ntpd 0.0% 0:38.68 1
2710 iCal 0.0% 0:29.20 2
307 netinfod 0.0% 0:28.61 1
2703 Dock 0.0% 0:21.06 2
444 sendmail 0.0% 0:20.68 1
483 AppleFileS 0.0% 0:19.75 2
486 httpd 0.0% 0:18.37 1
385 coreservic 0.0% 0:11.38 3
2732 System Pre 0.0% 0:11.19 1
375 autodiskmo 0.0% 0:09.75 13
2719 AppleSpell 0.0% 0:08.05 1
I do agree having a second processor generally won't speed up things like surfing, unless you're the type that continuously opens several sites at the same time, possibly with flash and java applets etc. However, the "snappiness" of a system can increase a lot...
That's not entirely true, the motherboard/system controller is proprietary as well. So you still need a motherboard manufacturer. There is at least one company doing this, but they're not exactly producing leading edge technology.