The commentator claims in part that spatial browsing is better because it encourages a shallow directory structure, which is clearly preferred over deep directory hierarchies for organizing information. He gives as a metaphor the contents of a drawer, which is easily visible to anyone who opens it. But he fails to consider the problems for people who have large numbers of files and documents that need organizing. Imposing shallow directory trees implies that there will either be large numbers of files in each directory, or that there will be a large number of subdirectories under each root and branch node. The appropriate metaphor then is not a few drawers in a desk to keep track of, but a garage with walls that are packed with the contents of shelves, boxes, jars, drawers, cabinets, and other containers. After a while, people forget where things are stored and resort to brute force searching to find things they know are there, but can't recall exactly where.
The solution isn't to impose a particular form of organization for storing and browsing files, but rather to provide superior tools for indexing and cataloging all entries so that they are easy to recall. What we need are browsers that allow us to browse by content attributes, rather than simply by file name or directory path.
Reducing eye strain and mental fatigue when reading from an electronic display can involve changing the size and style of text font used, the foreground and background colors, the average number of words displayed per line, and the average number of lines displayed per page. I've found the use of small portable devices such as a Palm PDA places limitations on the amount of flexibility I have adjusting one or more of the above at the same time. Viewing etexts on the high resolution display of a laptop or desktop computer using an appropriate reading program allows me to fully customize the display, but trades off some of the convenience factors.
On my Palm PDA I use the already mentioned Weasel Reader program. Nice because it's zTxt file format supports adding bookmarks and annotations to a copy of the etext for later referral. On a desktop or laptop, I use the PyGERS reader from the PyGE project. It also uses the same zTxt format as Weasel Reader, while allowing full control over fonts, colors, line length, and pagination while reading.
Since you're already familiar with wxPython, the PyGE project may interest you. Among the applications is an etext reader program which includes text-to-speech capabilities for both Windows platforms (using SAPI) and Unix/Linux platforms (using Festival). The Python modules for generating speech from text could be easily adapted for generating.wav files, which could then easily be converted to the compressed format of your choice.
The problem for some(many?) people will be in getting Sarge installed on a system where Debian has not already been installed. That is because the Sarge installer is unfinished. This means the most reliable way to install Debian Sarge is to first install Debian Woody, then use apt-get to do an update distribution command while hoping some unforseen dependancy problem doesn't break the install scripts during the upgrade.
I've performed the operation twice: from the previous version of Debian before Woody, and from Woody to Sarge. Each time an error prevented a completely automatic update from the old version to the new one. The fixes involved locating specific packages whose dependencies couldn't be met by the installer, and manually forcing certain packages to be installed in the correct order to satisfy those dependencies. For many people, having an install script stop working while giving only an obscure error message after its crash would lead to great initial confusion about what to do next.
If Debian would concentrate on finishing the Sarge installer and releasing it in a condition suitable for general use, they would reduce a lot of the pressure to release the entire Sarge distribution in a shorter timeframe. People who absolutely needed the functionality of software in Sarge and who were willing to put up with occasional bugs could then more easily satisfy those needs on their own.
It's hard to get what your point is when you consistently blur the real distinctions between the different types of intellectual property. You said "While pentium is a trademark, the graphic logo is copyrighted. Neither is free to modify and redistribute and thus a violation of 'free software' "
For one thing, the word "Pentium" and the Intel graphic logo are both trademarks for Intel. No software license, especially the GPL, gives anyone the right to modify or use someone else's trademark. Also, there is no restriction on redistribution of a trademark, so long as it is made clear who owns the trademark and there is no confusion elicited in viewers about what it is representing. Your example plainly does not show an example of a violation of the GPL.
Second, there is no single definition of the term "free software" as you use it. There are only specific licenses, each of which has its own conditions, which fall under the wide umbrella of free and/or open source software. Before you can discuss a violation, you have to specify exactly which license you believe is being violated.
Third, you ignore the plain difference between a binary file representing executable code (even if it runs only on an embedded processor within a device) and a binary file representing a static image. The binary executable represents a program, which during the course of its execution may effect the state of the computer system within which it runs. It will likely effect the state of the kernel which contains the driver which accesses the firmware. Without the natural expression of this firmware, its source code, it is difficult to really understand what those effects will be under all conditions. The static image file would have no indeterminate effects on the system. Contrary to your unsupported assertion otherwise, since the graphic image is fully described by the binary contents of the file, that file can be considered a definitive source format.
You need to learn about the various types of intellectual property recognized in law. Product names can be protected through trademarks, specific expressions of ideas can be protected through copyrights, and ideas themselves can only be protected through patents. Each type of intellectual property has its own set of laws and regulations which govern it. Issues involving the GPL fall under the category of copyrights.
In your comment you use the word information, which by itself has no legal meaning. Assuming you intended to ask about the expression of a program through its code, or an artistic expression of a corporate logo, then the answer is that the redistribution of such expressions can be restricted through copyrights. The owner of the copyright gets to decide conditions under which their work may be distributed to others. If anyone fails to honor those conditions, they have no legal right to continue distributing the copyrighted material. The GPL establishes a simple set of conditions under which software covered by a GPL license may be distributed, and everyone has the option of either abiding by those conditions or not incorporating software covered by the GPL in their products.
Debian's position is that including software components in the kernel without making available their source code violates the GPL requirements of other components in the kernel. Given this interpretation, their only available options are to stop distributing the GPL parts of the kernel, or stop distributing the non-GPL-compliant components. Not surprisingly, Debian has decided to strictly adhere to GPL requirements with regards to the kernel it distributes.
That Debian's decision even appears on anyone's radar is a testament to it's success in attracting a significant number of users who have chosen it in preference to many alternative distributions. People have to remember that proactive support of free software has always been the primary goal of Debian. The fact that reliance on software meeting relatively strict definitions of being free has also produced a distribution that many people have come to trust and rely on is appreciated, but popularity has never been the main impetus for Debian's developers. Some people seem to view Debian's current stance on binary-only firmware to be disrespectful of its users, who may be temporarily inconvenienced by lack of kernel drivers for some hardware. But people should respect the Debian developers and their wishes to produce the best system they possibly can using only truly free software. Debian cannot and should not be bound to placate people who want the quality and stability of Debian software, but who do not honor the primary reason for Debian's very existence.
There are plenty of other distributions to choose from for users who find Debian's actions unacceptable. People who insist on pragmatism over principles should probably choose one of those alternatives. Debian should be left alone to pursue their software ideal, especially since they are one of the few to place such a high value on software freedom. Rather than chide them for not following the pack and refusing to compromise their principles, they should be encouraged in their efforts demonstrate what can be accomplished using clearly free software.
The biggest hurdle to having well-formatted versions of the PG books is not the formatting standard, but simply the effort needed to enter in the formatting information. The way most PG books are produced now is by scanning in original works, running the image files through an optical character recognition program, proofreading of the generated text files by an army of distributed volunteers, and final assembly by a much smaller number of experienced editors.
It's during the OCR process that a lot of formatting information is lost, and that information can only be recovered by people manually viewing the original image and adding markup language to the text output. The current process works well because the most labor-intensive part, the visual proofreading and error correction, requires little in the way of specialized skills and leaves little room for inconsistencies due to interpretation or personal preferences. Adding extra responsibility for including markup language for style and formatting reasons would quickly reduce the pool of volunteers who were both qualified and interested in spending time contributing to the proofreading work needed to keep PG running.
According to a developer who actually worked on the Linux porting project at Autozone (no link provided out of respect to Groklaw), your scenario is wrong. The Autozone applications were directly ported to Linux, and were not simply OpenServer-compiled binaries running under a compatibility library for Linux. Under these conditions, SCO's assumptions are completely wrong and they have no viable case.
SCO isn't saying they couldn't make Linux work, but rather that Autozone couldn't make a clean port of their own applications from Unix to Linux. It seems that in addition to claiming (in the IBM case) that independently developed Linux code must belong in part to SCO if it was ever used with Unix, they are now saying that applications ported to Linux from Unix must have improperly used SCO property if the porting went well.
SCO management obviously has no idea how easy it is to port many applications between Unix and Linux. Since the only basis for SCO's suit is a suggestion that shared libraries were being improperly used, a simple directory listing from randomly chosen Autozone kiosk terminals should quickly show the basis to have no merit.
Except that SCO has shown a propensity to make companies with which it has existing business relationships the primary targets of its legal attacks. Rather than mitigating risk, buying a license from SCO increases exposure to further efforts by SCO to monetize their IP in the future, while voluntarily waiving a number of rights that could have been used in self defense. How is this a wise business decision?
Our value-add is really in the R&D in the technology.
A disconnect with customers happens when Microsoft confuses who should be the beneficiary of added value from R&D. Have the enormous sums spent annually on R&D resulted in superior performance in areas that are most important to customers: security, reliability, affordability, and flexibility? Or is it the case that R&D spending is concentrated on technologies to displace existing products and vendors from the marketplace (Internet Explorer vs. Netscape,.NET vs. Java, etc...) in favor of Microsoft, but without seriously addressing the needs of its customers?
SCO has cleverly designed a license which requires no proof of SCO IP in Linux. They are asking licensees to pay money for the right to not be sued by SCO for SCO IP that "is in" Linux. Whether any SCO IP actually exists is irrelevant since the license is nonspecific on the amount and type of SCO IP it covers. Even if eventually no SCO IP is found in Linux, it could be argued that licensees made their own judgements on why they needed to purchase a license despite knowing there was a possibility that the quantity of SCO IP to be found in Linux was actually zero. The only thing SCO technically has to deliver under the contract is to not sue its licensees.
There is nothing but profit for SCO from any corporations that purchase licenses since there is nothing that they have to deliver, and they have protected themselves by making no specific claims about IP they actually own. By agreeing to the license terms, you explicitly hold SCO harmless for any of their actions. It's easy money if anyone falls for the scheme.
Except when the purpose of the advertisement is to convince a consumer to use Brand X, rather then Brand Y which has virtually the same effect. In this case the advertising dollars result in no additional lives being saved. Or perhaps it convinces the consumer to demand Brand X from their doctor over Generic Z, even though Generic Z could save them money which could be used in other ways to improve the quality of their lives.
Realistically viewing the types of drug advertisements seen today, which do you think is the more likely scenario, yours or mine?
On your first point, I'll point you to this article that attempted to examine the effects of US-imposed drug price controls on overall global drug R&D spending. Note that the 15 largest pharmaceutical companies he studied were Pfizer, Merck & Co., AstraZeneca, Aventis, Bristol-Meyers Squibb, GlaxoWellcome, Pharmacia, Roche, Johnson & Johnson, American Home Products, Eli Lilly, SmithKline Beecham, Abbott Laboratories, Bayer, and Amgen. As is obvious from looking at the list and checking company websites, the pharmaceutical business is a global one. You can also note from the article (on page 23) that many European countries, including where a number of the top pharmaceutical companies are based, have drug price controls of one kind or another. It might be more accurate to say that U.S. consumers are the target market for the lion's share of new drugs.
The conclusion of the modeling study is that U.S. adoption of price controls equal to the average of industrialized European countries would result in lowered R&D spending overall, but that the decline could range in intensity from "a very small decline to a near complete cessation of R&D activity; the latter is, of course, an absurd conclusion". In other words, we probably shouldn't place much faith in any results from his model. The question of whether U.S. price controls would significantly cut R&D expenditures, or simply shift more of the burden for R&D to other global customers is beyond his model to answer.
On your second point, you gave the URL for the Pharmaceutical Researchers and Manufacturers of America site which contained a number of links. Not knowing exactly which article you wished to refer to, I'll simply choose the rebuttal article to this report by the Attorney General of the state of Minnesota, as they seem to clearly summarize the viewpoints of the pharmacy trade group and the opposing camp. Anyone interested in the subject should read both, hopefully with an open mind. If you would like to discuss any particular point, just let me know.
On the third point, I think your characterization of scientists is more of a caricature than a true picture. Scientists, whether publicly or privately funded, are people, who have normal human concerns and awareness of things around them. One thing basic research does is provide knowledge that makes the drug discovery phase more efficient. Discovery in finding the right compositions to pursue for new drugs is essential for making sure only the most promising approaches are followed up with more expensive development efforts and clinical trials. The pharmaceutical companies are often better suited for carrying on the later processes, but they heavily benefit from public research efforts that point them in the right direction. It's not just about total dollars spent, but also how much is saved by developing a proper scientific foundation for the search.
This also brings up the point that while the industry likes to brag about how much it costs to bring a new drug to market and how risky the prospects are for success, the truth is that the greatest development costs occur during the latest stages of product development (such as holding large clinical trials involving thousands of subjects). By the time decisions are made to hold such trials, the company must have already experienced enough success from earlier work to have a high degree of confidence in success. Most unsuccessful drugs are thus weeded out early, before huge expenditures are incurred in development. The better the basic science that is available, the more efficient this filtering process can be.
And whose fault is it that so many hundreds of millions of dollars are spent on ineffective, unneeded, or unprofitable drug lines that vitally needed medicines can be priced out of the reach of the people who need them? If you want to stop talking about rhetoric, let's discuss the actual breakdown of costs and examine ways that a workable balance of economic and social requirements can be met. That discussion is going on across the nation and around the globe today, and it does no good to try and dismiss one side of a very complex issue without considering some real issues.
One issue is the argument made by supporters of pharmaceutical companies that development costs of new drugs are enormous, and that high profit margins are needed to provide incentives for continued R&D. One should consider that high costs and guaranteed large profit margins means that companies have little incentive to reduce those costs. Similar to costs-plus contracts, the companies know that increased costs will be more than made up with increased prices and resulting revenues for successful products.
This brings up the issue of costs due to unsuccessful products. Instead of vague references to undocumented dollars and failed companies, I'll refer to figures in a paper found here. I recommend reading the report, not because I agree with any of its conclusions, but because it gives real numbers backed by referenced sources. Those numbers suggest that the often-stated costs to pharmaceutical companies to develop new drugs are overstated, and that risks are highly mitigated. If you have similarly documented figures backing your statements I would be happy to view them.
Finally, we could discuss how you know that patents are crucial to having effective drugs in the future. In another section of the report cited above, an NIH report reveals that publicly funded research provided a majority of the R&D studies behind the most commercially successful drugs in 1995. This does not include the long-term efforts of public agencies like NIH in researching treatment for diseases like cancer and AIDS. Using the FDA priority rating system for new drug approval applications from 1992-1999 it appears that only 22% of them represented important therapeutic gains, while the majority were for "me too" treatments.
There is of course much more than can be said here. Trying to separate out the facts from the opinions and misrepresentations is difficult, but the consequences of making the wrong choices will be reflected in their effects on the lives of real people.
Yeah. Gosh. supply. Demand. Different markets. Variable pricing strategies. Absolutely shocking. We simply must create laws so that the government can set fair prices for everyone.
Sarcasm aside, the difference with HIV drugs is that laws are being used to eliminate the ability of competitors to offer alternative supplies of life-saving drugs. Drug companies enjoy a government-enabled monopoly through internationally recognized patent rights, which means that the government is helping to keep profits artificially high for patent holders versus what they would be in a true free market. To be consistent with a true free market, drugs should be freely distributed across national boundaries at the best price people could negotiate for themselves. Then there would be no need for the government to set fair prices.
So are you only against government-mandated price regulation when it doesn't hurt corporate profits? Or would you support the proposal that artificial government edicts which restrain free trade and reduce peoples choices should be ended?
Inventions do not require that any or all of their components be unique or revolutionary in design. Combining existing devices in original ways to produce something new is allowed. Nor does the resulting functionality have to be something never seen before. An invention can also be designed to do things done by other inventions, maybe just a little better. To think otherwise would be like saying that no patents should ever be issued for any mechanical device incorporating levers, gears, bearings, shafts, etc., because these are all known technologies that anyone skilled in the art should be aware of. It's not the individual pieces that matter, but how they are used in combination.
On their face, the Patriot patents certainly seem to describe a class of processor with specific features and details sufficient to qualify for designation as an invention. The troubling aspect is that the claims of the patents can be construed to cover many other processors developed by other vendors which have little in common with the actual invention described.
Patent No. 6,598,148, appears to be a divisional patent from patent No. 5,982,231, filed in 1989. In summary, the invention describes a CPU using both stack and register-based instructions which interfaces directly to DRAM memory over 32 shared address and data lines. Memory accesses are minimized by using mainly stack-based instructions and accessing memory in fast page mode. The invention is designed to minimize cost, fitting just tens of thousands of transistors into a 44-pin package, while delivering approximately 20 MIPS performance. While none of the design elements by themselves are new or unique, the particular combination chosen seems to fit the definition of a real invention that was not in existence before 1989.
The features which seem to relate to its claims of variable speed are the use of an on-board ring oscillator to drive the system clock, and memory cycle timing which automatically adjusts to the loading seen from attached memory devices. Using an on-board ring oscillator means that the system clock can be driven by a device built with process parameters identical to other devices on the chip. If CPU transistors are a little slow on a batch of chips, the ring oscillator transistors will be similarly slow and the chips will continue to work at a slower but reliable speed.
It's just not clear to me how Patriot Scientific can claim that companies like Intel are infringing on their patent. Is it because Intel's products contain technology which also happens to appear in their invention, but which they themselves did not invent? None of the individual technologies used in the Patriot device originated with them, and it appears that only the specific combination of those technologies described in the patent application would have qualified it for approval. However, the patent claims are written in such a way as to seemingly describe a much broader range of CPU designs, many examples of which were certainly in existence before even the original patent application by Patriot was filed.
I've seen it stated that the claims are the only important part of a patent. That seems to be the big problem with this particular patent. What happens when a legitimate invention is granted a patent with overly broad claims? Some patents are written with claims that include a range of claims, some very specific and some very broad, on the assumption that it's better to aim for as much as possible and settle for what you can get. This particular patent seems to have only broad claims. Does that make it easier or harder to try and overturn?
I think it's considered trolling to continue making negative accusations when the basis for those accusations has been refuted. This is particularly true when you have failed to provide additional evidence to support your opinions after being given ample opportunity to do so. If what you are arguing is factual, the weight of evidence should prove your point. If what you are promoting is opinion, then simply repeating the same controversial statements repeatedly might easily be considered to be intentionally redundant, uninformative, and designed to be disruptive. That would be a form of trolling.
This is one of the advantages of the Debian package management system. A developer creating a.deb package includes information about dependencies on other packages, such as libraries. Then when someone performs an apt-get install packagename command, any required components not already installed are also downloaded and installed automatically. The Debian package management system thus handles the kinds of dependency problems you are complaining about.
Other Linux distributions may not have package management systems with the same level of automated dependency handling when installing new packages. Even if they do, there is no guarentee that packages from different distributions will contain exactly the same set of files or use identical versions of them. Each distribution is free to make modifications to upstream source files and configuration settings in order to better fit the software into that distribution, and distros like Debian and Redhat do make some changes in their released software. This is all a consequence of the distributed manner in which most components are created and maintained. It's also obvious that individuals working on their own time and following their own schedules will not have the same coordination as programmers working for a single corporation like Microsoft. You also have to keep in mind that a lot of OSS software is not written just to be run on Linux, but is designed to be buildable and usable on different architectures and OS's. In addition, major revisions sometimes appear in essential components that require the updating of nearly every other package in a distribution (as with changes in the glibc library and the gcc compiler), resulting in major differences between versions of the same distribution. All these factors make the creation of a few standardized packages that are completely consistent between all distributions highly unlikely.
I also think you paint too rosy a picture of the situation on Windows. While the core APIs implemented as dlls provided by Microsoft are relatively consistent across different Windows machines, the amount of library sharing that is practical between applications from different vendors is limited by the closed nature of libraries developed to perform more specific operations. As a rather simple-minded example, within my Windows windows\system folder I have over 1000 dll files. Within my Debian Linux/usr/lib directory if have slightly more than 500.so files, a number of which are simply different versions of the same library. My Linux system supports literally hundreds of installed applications, while the number of installed Windows applications is less than 20. While not a truly exact or accurate measure of shared library usage since libraries exist on both systems in other locations, it does give some evidence that each Windows application must carry along a lot more of its own library functionality instead of relying on what is generally available. This is a price that closed-source approaches pay versus open source. If you were really worried about ensuring that your Linux applications were buildable on all systems, you could go a similar route as the Windows examples and simply include the exact versions of libraries required by your software within your own software release.
Re:Let the conspiracy theories begin...
on
Make More Mistakes
·
· Score: 1
People won't pay for something that they can get for free.
I think the bottled water business is glad they didn't listen to advice like that. And despite the wide availability of music on the radio, and recently through easily accessible downloading technology, people continue to purchase music at inflated prices. Your blanket statement is simply wrong. People will pay for anything they want badly enough; and their reasons do not have to be entirely rational. It's all about the marketing.
Doesn't this mean that code developed and contributed by common developers could be dual-licensed by them? There is no requirement that ALL code sections be independently developed, only that the appropriate licensing requirements are met. As in all cases involving questions of copyright infractions, you should be able to document the provenance of the code in question before making serious accusations.
People seem to feel compelled to take one of two opposing sides in response to the article. Either they feel the CEO's comments reinforce the idea that Linux for the desktop is a failure, or they feel Linux needs to be defended against all criticism. There is another way to look at things, which is to admit that Linux is a work in progress; and that it should be evaluated based on how it is progressing towards meeting the goals of its users and not just on where it is at today.
Viewed in this way, I think it is reasonable to admit that Linux does not currently meet the expectations of a majority of consumers as a desktop operating system. This is not an inherent deficiency in Linux itself, but reflects an immaturity in software that is currently available for it. This does not mean it has failed! Just as Windows required years of development (anyone remember Windows 1.0?) to convince users and developers that it was a worthy environment, Linux will likely require a similar period before gaining widespread acceptance.
Part of the growing process, however, requires that people listen to complaints and criticisms so that real problems can be addressed and corrected. In the end, admitting that there are problems and that improvements can and should be made will only make the final result a better product for everyone.
Slashdot Mirror
The commentator claims in part that spatial browsing is better because it encourages a shallow directory structure, which is clearly preferred over deep directory hierarchies for organizing information. He gives as a metaphor the contents of a drawer, which is easily visible to anyone who opens it. But he fails to consider the problems for people who have large numbers of files and documents that need organizing. Imposing shallow directory trees implies that there will either be large numbers of files in each directory, or that there will be a large number of subdirectories under each root and branch node. The appropriate metaphor then is not a few drawers in a desk to keep track of, but a garage with walls that are packed with the contents of shelves, boxes, jars, drawers, cabinets, and other containers. After a while, people forget where things are stored and resort to brute force searching to find things they know are there, but can't recall exactly where.
The solution isn't to impose a particular form of organization for storing and browsing files, but rather to provide superior tools for indexing and cataloging all entries so that they are easy to recall. What we need are browsers that allow us to browse by content attributes, rather than simply by file name or directory path.
Reducing eye strain and mental fatigue when reading from an electronic display can involve changing the size and style of text font used, the foreground and background colors, the average number of words displayed per line, and the average number of lines displayed per page. I've found the use of small portable devices such as a Palm PDA places limitations on the amount of flexibility I have adjusting one or more of the above at the same time. Viewing etexts on the high resolution display of a laptop or desktop computer using an appropriate reading program allows me to fully customize the display, but trades off some of the convenience factors.
On my Palm PDA I use the already mentioned Weasel Reader program. Nice because it's zTxt file format supports adding bookmarks and annotations to a copy of the etext for later referral. On a desktop or laptop, I use the PyGERS reader from the PyGE project. It also uses the same zTxt format as Weasel Reader, while allowing full control over fonts, colors, line length, and pagination while reading.
Since you're already familiar with wxPython, the PyGE project may interest you. Among the applications is an etext reader program which includes text-to-speech capabilities for both Windows platforms (using SAPI) and Unix/Linux platforms (using Festival). The Python modules for generating speech from text could be easily adapted for generating .wav files, which could then easily be converted to the compressed format of your choice.
The problem for some(many?) people will be in getting Sarge installed on a system where Debian has not already been installed. That is because the Sarge installer is unfinished. This means the most reliable way to install Debian Sarge is to first install Debian Woody, then use apt-get to do an update distribution command while hoping some unforseen dependancy problem doesn't break the install scripts during the upgrade.
I've performed the operation twice: from the previous version of Debian before Woody, and from Woody to Sarge. Each time an error prevented a completely automatic update from the old version to the new one. The fixes involved locating specific packages whose dependencies couldn't be met by the installer, and manually forcing certain packages to be installed in the correct order to satisfy those dependencies. For many people, having an install script stop working while giving only an obscure error message after its crash would lead to great initial confusion about what to do next.
If Debian would concentrate on finishing the Sarge installer and releasing it in a condition suitable for general use, they would reduce a lot of the pressure to release the entire Sarge distribution in a shorter timeframe. People who absolutely needed the functionality of software in Sarge and who were willing to put up with occasional bugs could then more easily satisfy those needs on their own.
It's hard to get what your point is when you consistently blur the real distinctions between the different types of intellectual property. You said "While pentium is a trademark, the graphic logo is copyrighted. Neither is free to modify and redistribute and thus a violation of 'free software' "
For one thing, the word "Pentium" and the Intel graphic logo are both trademarks for Intel. No software license, especially the GPL, gives anyone the right to modify or use someone else's trademark. Also, there is no restriction on redistribution of a trademark, so long as it is made clear who owns the trademark and there is no confusion elicited in viewers about what it is representing. Your example plainly does not show an example of a violation of the GPL.
Second, there is no single definition of the term "free software" as you use it. There are only specific licenses, each of which has its own conditions, which fall under the wide umbrella of free and/or open source software. Before you can discuss a violation, you have to specify exactly which license you believe is being violated.
Third, you ignore the plain difference between a binary file representing executable code (even if it runs only on an embedded processor within a device) and a binary file representing a static image. The binary executable represents a program, which during the course of its execution may effect the state of the computer system within which it runs. It will likely effect the state of the kernel which contains the driver which accesses the firmware. Without the natural expression of this firmware, its source code, it is difficult to really understand what those effects will be under all conditions. The static image file would have no indeterminate effects on the system. Contrary to your unsupported assertion otherwise, since the graphic image is fully described by the binary contents of the file, that file can be considered a definitive source format.
You need to learn about the various types of intellectual property recognized in law. Product names can be protected through trademarks, specific expressions of ideas can be protected through copyrights, and ideas themselves can only be protected through patents. Each type of intellectual property has its own set of laws and regulations which govern it. Issues involving the GPL fall under the category of copyrights.
In your comment you use the word information, which by itself has no legal meaning. Assuming you intended to ask about the expression of a program through its code, or an artistic expression of a corporate logo, then the answer is that the redistribution of such expressions can be restricted through copyrights. The owner of the copyright gets to decide conditions under which their work may be distributed to others. If anyone fails to honor those conditions, they have no legal right to continue distributing the copyrighted material. The GPL establishes a simple set of conditions under which software covered by a GPL license may be distributed, and everyone has the option of either abiding by those conditions or not incorporating software covered by the GPL in their products.
Debian's position is that including software components in the kernel without making available their source code violates the GPL requirements of other components in the kernel. Given this interpretation, their only available options are to stop distributing the GPL parts of the kernel, or stop distributing the non-GPL-compliant components. Not surprisingly, Debian has decided to strictly adhere to GPL requirements with regards to the kernel it distributes.
That Debian's decision even appears on anyone's radar is a testament to it's success in attracting a significant number of users who have chosen it in preference to many alternative distributions. People have to remember that proactive support of free software has always been the primary goal of Debian. The fact that reliance on software meeting relatively strict definitions of being free has also produced a distribution that many people have come to trust and rely on is appreciated, but popularity has never been the main impetus for Debian's developers. Some people seem to view Debian's current stance on binary-only firmware to be disrespectful of its users, who may be temporarily inconvenienced by lack of kernel drivers for some hardware. But people should respect the Debian developers and their wishes to produce the best system they possibly can using only truly free software. Debian cannot and should not be bound to placate people who want the quality and stability of Debian software, but who do not honor the primary reason for Debian's very existence.
There are plenty of other distributions to choose from for users who find Debian's actions unacceptable. People who insist on pragmatism over principles should probably choose one of those alternatives. Debian should be left alone to pursue their software ideal, especially since they are one of the few to place such a high value on software freedom. Rather than chide them for not following the pack and refusing to compromise their principles, they should be encouraged in their efforts demonstrate what can be accomplished using clearly free software.
The biggest hurdle to having well-formatted versions of the PG books is not the formatting standard, but simply the effort needed to enter in the formatting information. The way most PG books are produced now is by scanning in original works, running the image files through an optical character recognition program, proofreading of the generated text files by an army of distributed volunteers, and final assembly by a much smaller number of experienced editors.
It's during the OCR process that a lot of formatting information is lost, and that information can only be recovered by people manually viewing the original image and adding markup language to the text output. The current process works well because the most labor-intensive part, the visual proofreading and error correction, requires little in the way of specialized skills and leaves little room for inconsistencies due to interpretation or personal preferences. Adding extra responsibility for including markup language for style and formatting reasons would quickly reduce the pool of volunteers who were both qualified and interested in spending time contributing to the proofreading work needed to keep PG running.
Here is one open source project for displaying and reading PG e-texts on a computer. It doesn't do everything you asked for, but does quite a bit.
According to a developer who actually worked on the Linux porting project at Autozone (no link provided out of respect to Groklaw), your scenario is wrong. The Autozone applications were directly ported to Linux, and were not simply OpenServer-compiled binaries running under a compatibility library for Linux. Under these conditions, SCO's assumptions are completely wrong and they have no viable case.
SCO isn't saying they couldn't make Linux work, but rather that Autozone couldn't make a clean port of their own applications from Unix to Linux. It seems that in addition to claiming (in the IBM case) that independently developed Linux code must belong in part to SCO if it was ever used with Unix, they are now saying that applications ported to Linux from Unix must have improperly used SCO property if the porting went well.
SCO management obviously has no idea how easy it is to port many applications between Unix and Linux. Since the only basis for SCO's suit is a suggestion that shared libraries were being improperly used, a simple directory listing from randomly chosen Autozone kiosk terminals should quickly show the basis to have no merit.
Except that SCO has shown a propensity to make companies with which it has existing business relationships the primary targets of its legal attacks. Rather than mitigating risk, buying a license from SCO increases exposure to further efforts by SCO to monetize their IP in the future, while voluntarily waiving a number of rights that could have been used in self defense. How is this a wise business decision?
Our value-add is really in the R&D in the technology.
A disconnect with customers happens when Microsoft confuses who should be the beneficiary of added value from R&D. Have the enormous sums spent annually on R&D resulted in superior performance in areas that are most important to customers: security, reliability, affordability, and flexibility? Or is it the case that R&D spending is concentrated on technologies to displace existing products and vendors from the marketplace (Internet Explorer vs. Netscape, .NET vs. Java, etc...) in favor of Microsoft, but without seriously addressing the needs of its customers?
SCO has cleverly designed a license which requires no proof of SCO IP in Linux. They are asking licensees to pay money for the right to not be sued by SCO for SCO IP that "is in" Linux. Whether any SCO IP actually exists is irrelevant since the license is nonspecific on the amount and type of SCO IP it covers. Even if eventually no SCO IP is found in Linux, it could be argued that licensees made their own judgements on why they needed to purchase a license despite knowing there was a possibility that the quantity of SCO IP to be found in Linux was actually zero. The only thing SCO technically has to deliver under the contract is to not sue its licensees.
There is nothing but profit for SCO from any corporations that purchase licenses since there is nothing that they have to deliver, and they have protected themselves by making no specific claims about IP they actually own. By agreeing to the license terms, you explicitly hold SCO harmless for any of their actions. It's easy money if anyone falls for the scheme.
Except when the purpose of the advertisement is to convince a consumer to use Brand X, rather then Brand Y which has virtually the same effect. In this case the advertising dollars result in no additional lives being saved. Or perhaps it convinces the consumer to demand Brand X from their doctor over Generic Z, even though Generic Z could save them money which could be used in other ways to improve the quality of their lives.
Realistically viewing the types of drug advertisements seen today, which do you think is the more likely scenario, yours or mine?
On your first point, I'll point you to this article that attempted to examine the effects of US-imposed drug price controls on overall global drug R&D spending. Note that the 15 largest pharmaceutical companies he studied were Pfizer, Merck & Co., AstraZeneca, Aventis, Bristol-Meyers Squibb, GlaxoWellcome, Pharmacia, Roche, Johnson & Johnson, American Home Products, Eli Lilly, SmithKline Beecham, Abbott Laboratories, Bayer, and Amgen. As is obvious from looking at the list and checking company websites, the pharmaceutical business is a global one. You can also note from the article (on page 23) that many European countries, including where a number of the top pharmaceutical companies are based, have drug price controls of one kind or another. It might be more accurate to say that U.S. consumers are the target market for the lion's share of new drugs.
The conclusion of the modeling study is that U.S. adoption of price controls equal to the average of industrialized European countries would result in lowered R&D spending overall, but that the decline could range in intensity from "a very small decline to a near complete cessation of R&D activity; the latter is, of course, an absurd conclusion". In other words, we probably shouldn't place much faith in any results from his model. The question of whether U.S. price controls would significantly cut R&D expenditures, or simply shift more of the burden for R&D to other global customers is beyond his model to answer.
On your second point, you gave the URL for the Pharmaceutical Researchers and Manufacturers of America site which contained a number of links. Not knowing exactly which article you wished to refer to, I'll simply choose the rebuttal article to this report by the Attorney General of the state of Minnesota, as they seem to clearly summarize the viewpoints of the pharmacy trade group and the opposing camp. Anyone interested in the subject should read both, hopefully with an open mind. If you would like to discuss any particular point, just let me know.
On the third point, I think your characterization of scientists is more of a caricature than a true picture. Scientists, whether publicly or privately funded, are people, who have normal human concerns and awareness of things around them. One thing basic research does is provide knowledge that makes the drug discovery phase more efficient. Discovery in finding the right compositions to pursue for new drugs is essential for making sure only the most promising approaches are followed up with more expensive development efforts and clinical trials. The pharmaceutical companies are often better suited for carrying on the later processes, but they heavily benefit from public research efforts that point them in the right direction. It's not just about total dollars spent, but also how much is saved by developing a proper scientific foundation for the search.
This also brings up the point that while the industry likes to brag about how much it costs to bring a new drug to market and how risky the prospects are for success, the truth is that the greatest development costs occur during the latest stages of product development (such as holding large clinical trials involving thousands of subjects). By the time decisions are made to hold such trials, the company must have already experienced enough success from earlier work to have a high degree of confidence in success. Most unsuccessful drugs are thus weeded out early, before huge expenditures are incurred in development. The better the basic science that is available, the more efficient this filtering process can be.
And whose fault is it that so many hundreds of millions of dollars are spent on ineffective, unneeded, or unprofitable drug lines that vitally needed medicines can be priced out of the reach of the people who need them? If you want to stop talking about rhetoric, let's discuss the actual breakdown of costs and examine ways that a workable balance of economic and social requirements can be met. That discussion is going on across the nation and around the globe today, and it does no good to try and dismiss one side of a very complex issue without considering some real issues.
One issue is the argument made by supporters of pharmaceutical companies that development costs of new drugs are enormous, and that high profit margins are needed to provide incentives for continued R&D. One should consider that high costs and guaranteed large profit margins means that companies have little incentive to reduce those costs. Similar to costs-plus contracts, the companies know that increased costs will be more than made up with increased prices and resulting revenues for successful products.
This brings up the issue of costs due to unsuccessful products. Instead of vague references to undocumented dollars and failed companies, I'll refer to figures in a paper found here. I recommend reading the report, not because I agree with any of its conclusions, but because it gives real numbers backed by referenced sources. Those numbers suggest that the often-stated costs to pharmaceutical companies to develop new drugs are overstated, and that risks are highly mitigated. If you have similarly documented figures backing your statements I would be happy to view them.
Finally, we could discuss how you know that patents are crucial to having effective drugs in the future. In another section of the report cited above, an NIH report reveals that publicly funded research provided a majority of the R&D studies behind the most commercially successful drugs in 1995. This does not include the long-term efforts of public agencies like NIH in researching treatment for diseases like cancer and AIDS. Using the FDA priority rating system for new drug approval applications from 1992-1999 it appears that only 22% of them represented important therapeutic gains, while the majority were for "me too" treatments.
There is of course much more than can be said here. Trying to separate out the facts from the opinions and misrepresentations is difficult, but the consequences of making the wrong choices will be reflected in their effects on the lives of real people.
Yeah. Gosh. supply. Demand. Different markets. Variable pricing strategies. Absolutely shocking. We simply must create laws so that the government can set fair prices for everyone.
Sarcasm aside, the difference with HIV drugs is that laws are being used to eliminate the ability of competitors to offer alternative supplies of life-saving drugs. Drug companies enjoy a government-enabled monopoly through internationally recognized patent rights, which means that the government is helping to keep profits artificially high for patent holders versus what they would be in a true free market. To be consistent with a true free market, drugs should be freely distributed across national boundaries at the best price people could negotiate for themselves. Then there would be no need for the government to set fair prices.
So are you only against government-mandated price regulation when it doesn't hurt corporate profits? Or would you support the proposal that artificial government edicts which restrain free trade and reduce peoples choices should be ended?
Inventions do not require that any or all of their components be unique or revolutionary in design. Combining existing devices in original ways to produce something new is allowed. Nor does the resulting functionality have to be something never seen before. An invention can also be designed to do things done by other inventions, maybe just a little better. To think otherwise would be like saying that no patents should ever be issued for any mechanical device incorporating levers, gears, bearings, shafts, etc., because these are all known technologies that anyone skilled in the art should be aware of. It's not the individual pieces that matter, but how they are used in combination.
On their face, the Patriot patents certainly seem to describe a class of processor with specific features and details sufficient to qualify for designation as an invention. The troubling aspect is that the claims of the patents can be construed to cover many other processors developed by other vendors which have little in common with the actual invention described.
Patent No. 6,598,148,
appears to be a divisional patent from patent No. 5,982,231, filed in 1989. In summary, the invention describes a CPU using both stack and register-based instructions which interfaces directly to DRAM memory over 32 shared address and data lines. Memory accesses are minimized by using mainly stack-based instructions and accessing memory in fast page mode. The invention is designed to minimize cost, fitting just tens of thousands of transistors into a 44-pin package, while delivering approximately 20 MIPS performance. While none of the design elements by themselves are new or unique, the particular combination chosen seems to fit the definition of a real invention that was not in existence before 1989.
The features which seem to relate to its claims of variable speed are the use of an on-board ring oscillator to drive the system clock, and memory cycle timing which automatically adjusts to the loading seen from attached memory devices. Using an on-board ring oscillator means that the system clock can be driven by a device built with process parameters identical to other devices on the chip. If CPU transistors are a little slow on a batch of chips, the ring oscillator transistors will be similarly slow and the chips will continue to work at a slower but reliable speed.
It's just not clear to me how Patriot Scientific can claim that companies like Intel are infringing on their patent. Is it because Intel's products contain technology which also happens to appear in their invention, but which they themselves did not invent? None of the individual technologies used in the Patriot device originated with them, and it appears that only the specific combination of those technologies described in the patent application would have qualified it for approval. However, the patent claims are written in such a way as to seemingly describe a much broader range of CPU designs, many examples of which were certainly in existence before even the original patent application by Patriot was filed.
I've seen it stated that the claims are the only important part of a patent. That seems to be the big problem with this particular patent. What happens when a legitimate invention is granted a patent with overly broad claims? Some patents are written with claims that include a range of claims, some very specific and some very broad, on the assumption that it's better to aim for as much as possible and settle for what you can get. This particular patent seems to have only broad claims. Does that make it easier or harder to try and overturn?
I think it's considered trolling to continue making negative accusations when the basis for those accusations has been refuted. This is particularly true when you have failed to provide additional evidence to support your opinions after being given ample opportunity to do so. If what you are arguing is factual, the weight of evidence should prove your point. If what you are promoting is opinion, then simply repeating the same controversial statements repeatedly might easily be considered to be intentionally redundant, uninformative, and designed to be disruptive. That would be a form of trolling.
This is one of the advantages of the Debian package management system. A developer creating a .deb package includes information about dependencies on other packages, such as libraries. Then when someone performs an apt-get install packagename command, any required components not already installed are also downloaded and installed automatically. The Debian package management system thus handles the kinds of dependency problems you are complaining about.
Other Linux distributions may not have package management systems with the same level of automated dependency handling when installing new packages. Even if they do, there is no guarentee that packages from different distributions will contain exactly the same set of files or use identical versions of them. Each distribution is free to make modifications to upstream source files and configuration settings in order to better fit the software into that distribution, and distros like Debian and Redhat do make some changes in their released software. This is all a consequence of the distributed manner in which most components are created and maintained. It's also obvious that individuals working on their own time and following their own schedules will not have the same coordination as programmers working for a single corporation like Microsoft. You also have to keep in mind that a lot of OSS software is not written just to be run on Linux, but is designed to be buildable and usable on different architectures and OS's. In addition, major revisions sometimes appear in essential components that require the updating of nearly every other package in a distribution (as with changes in the glibc library and the gcc compiler), resulting in major differences between versions of the same distribution. All these factors make the creation of a few standardized packages that are completely consistent between all distributions highly unlikely.
I also think you paint too rosy a picture of the situation on Windows. While the core APIs implemented as dlls provided by Microsoft are relatively consistent across different Windows machines, the amount of library sharing that is practical between applications from different vendors is limited by the closed nature of libraries developed to perform more specific operations. As a rather simple-minded example, within my Windows windows\system folder I have over 1000 dll files. Within my Debian Linux /usr/lib directory if have slightly more than 500 .so files, a number of which are simply different versions of the same library. My Linux system supports literally hundreds of installed applications, while the number of installed Windows applications is less than 20. While not a truly exact or accurate measure of shared library usage since libraries exist on both systems in other locations, it does give some evidence that each Windows application must carry along a lot more of its own library functionality instead of relying on what is generally available. This is a price that closed-source approaches pay versus open source. If you were really worried about ensuring that your Linux applications were buildable on all systems, you could go a similar route as the Windows examples and simply include the exact versions of libraries required by your software within your own software release.
I think the bottled water business is glad they didn't listen to advice like that. And despite the wide availability of music on the radio, and recently through easily accessible downloading technology, people continue to purchase music at inflated prices. Your blanket statement is simply wrong. People will pay for anything they want badly enough; and their reasons do not have to be entirely rational. It's all about the marketing.
Doesn't this mean that code developed and contributed by common developers could be dual-licensed by them? There is no requirement that ALL code sections be independently developed, only that the appropriate licensing requirements are met. As in all cases involving questions of copyright infractions, you should be able to document the provenance of the code in question before making serious accusations.
People seem to feel compelled to take one of two opposing sides in response to the article. Either they feel the CEO's comments reinforce the idea that Linux for the desktop is a failure, or they feel Linux needs to be defended against all criticism. There is another way to look at things, which is to admit that Linux is a work in progress; and that it should be evaluated based on how it is progressing towards meeting the goals of its users and not just on where it is at today.
Viewed in this way, I think it is reasonable to admit that Linux does not currently meet the expectations of a majority of consumers as a desktop operating system. This is not an inherent deficiency in Linux itself, but reflects an immaturity in software that is currently available for it. This does not mean it has failed! Just as Windows required years of development (anyone remember Windows 1.0?) to convince users and developers that it was a worthy environment, Linux will likely require a similar period before gaining widespread acceptance.
Part of the growing process, however, requires that people listen to complaints and criticisms so that real problems can be addressed and corrected. In the end, admitting that there are problems and that improvements can and should be made will only make the final result a better product for everyone.