Besides, these sorts of systems break down when evaluating a medicated subject. Beta blockers are highly effective at suppressing the involuntary physiological effects of fear. Ask any professional musician. So is alcohol, though unlike beta blockers it has the problematic effect of making you drunk.
Cellular tower equipment doesn't draw significantly less power when there are fewer active calls, so limiting calls to 911 would not extend the battery life by much. I know, I used to do power monitoring systems for the industry.
The problem for carriers is that many sites won't allow generators. Tower sites usually will, although cost is a problem and in California getting a discharge permit from the air quality board is a problem. Getting fuel hauled in can be a problem for some of the more remote sites. Office buildings, apartment buildings, and so on usually will not allow generators because of safety problems with exhaust and fire code compliance problems with fuel storage. Water towers have contamination concerns from fuel storage.
Most cellular carriers up until now have designed noncritical cell sites at tower locations for 2 hours battery time and usually have a transfer switch wired in place for external power from a trailer-mounted generator. Then they have generator trailers at a rate of about one trailer per 10 cell sites. When power losses aren't widespread and the roads are passable, that works OK except that you have techs spending their whole day driving around fueling up generators. Inside a building they will try to get emergency power from the building owner, and depending on the history of the site and the size of the coverage footprint they will increase battery capacity considerably.
There are practical limits to how much battery capacity can be installed, because batteries are big and heavy. And batteries by themselves are not a guarantee, because the HVAC is not ordinarily on an inverter, and on a hot day in a rooftop installation you'll see the site shut down due to overtemp before it shuts down due to the low voltage disconnect for the battery string.
On site generators aren't the perfect answer either. These sites are unattended. Generators don't always start, and unless a site has natural gas as one of its fuel sources (few do) you have a limited fuel supply, usually 3-7 days. With greater reliance on generators, and impassible roads, the fuel haulers may not be able to serve everyone that frequently.
My present employer places great emphasis on intellectual property, and I can't imagine a new hire (or anyone else) being penalized for bringing up the situation described with management. If anything, we'd give them a medal. And yes, we would discard code and rewrite it to resolve IP problems, if we found that the code was a copyright violation.
Other employers aren't like that, especially smaller employers producing products where the code won't be seen by many people, or people working on internal software that will never see the light of day.
Part of the reason small business compliance is lower is that large corporations get volume licenses, which lowers the cost dramatically and reduces the amount of tracking and paperwork required per seat. It also gets them away from the purportedly non-transferable OEM licenses that the BSA is disinclined to count.
The criticism of the BSA isn't rooted in pro-piracy beliefs. The issue is that the BSA uses scare tactics to convince small business owners to sign a contract they shouldn't. The terms of the contract are such that the target ends up paying to purchase software which they have already paid for and are using lawfully and within the provisions of the license agreement.
One of many ways the law enforcement and legal system, with its constitutional protections and other checks and balances, is pushed aside by large corporate interests who have their own laws (shrink wrap agreement), investigators (BSA, RIAA), and courts (mandatory arbitration).
I was one of the people involved in creating the Notability policy. Articles on subjects that are not notable pose numerous problems for Wikipedia.
The wiki process works poorly regarding topics that are not of lasting interest to large numbers of people. Eventualism on Wikipedia, the idea that a stub or poorly written article will ultimately improve, relies upon there being interested individuals who will show up and edit the article based on their interests and knowledge. Subjects that aren't really notable don't draw many editors, and the editors they do draw tend to be connected with the subject in some way: fans, critics, friends, students, and others who are likely to add material which is not neutral and can't be substantiated by solid sources. The test case that led to the creation of notability policy years ago was a self-described surrealist and artist, who had created numerous articles related to his activities that were impossible to refute as unfactual because the references upon which they were purportedly based were local newspapers and magazines that were not available on-line or through even the largest libraries. The problem gets worse with the passage of time, since blogs and minor print publications can disappear without a trace. Hence, the notability policy came to emphasize coverage in national or major regional publications.
Articles on subjects that aren't notable draw the lion's share of formal complaints to the Wikimedia Foundation. This is mainly a problem with biographies, but to a lesser extent occurs with schools, bands, minor political offices, corporations, publications, and so on. These complaints have to be handled responsibly and are a burdensome distraction from the business of the Foundation, requiring the involvement of the most capable volunteers as well as, in some cases, paid staff. Truly notable people and organizations are unlikely to complain about their articles due to the widespread press coverage they already receive.
One of the posters above notes that there are widespread articles on Wikipedia on minor porn stars which do not draw deletions as do minor webcomix. Wikipedia is nothing if not inconsistent in its application of guidelines and policies, so the presence of such counterexamples doesn't mean much except that there aren't as many Wikipedians willing to take the time to police that category of articles.
Indeed, deletion policy is and always has been applied inconsistently. Any deletion decision has a certain amount of randomness because of the nature of the process. That said, there is a strong bias towards keeping stuff. Precious few truly meritorious articles get deleted, and there is a great deal of junk that gets kept only because of luck and the predisposition to keep things.
What is unfortunate is that IMSLP appears likely to remain down for a period of months while its founder considers the offers from Gutenberg and others and decides what to do, leaving a valuable resource created by a large group of volunteers unavailable in the interim.
Too bad they didn't offer database dumps or other convenient means for mirrors to obtain the content.
One might conclude that the Service listened to a recording of the conversation. Many if not most tech support and customer service calls are recorded.
That's easy. Anyone with a modicum of talent and an undergraduate engineering degree from a U.S. college or university can find just as good a job as they would with a graduate degree.
With an engineering degree from a non-U.S. institution, that's not true, especially for jobs in the U.S. Hence, people from many other countries come to the United States to get a graduate degree so they can open the door that a U.S. undergrad degree would open.
The only jobs where a graduate engineering degree is required or will help are teaching positions and very high-level technical jobs which also require unusual talent and decades of experience.
Take a look at what http://www.clearwire.com/ is doing for an idea of pricing for the devices and service and for the kinds of adapters available. The technology they are using is fundamentally very similar to WiMax.
The municipal WiFi players don't have spectrum, but they do have mounting locations and (in some cases) backhaul. Not trivial assets when contemplating a deployment.
The fuckup at the root of all this was that he was pushing his luck with the cop. You have to pick your fights. The fight with the store was a valid one and fundamentally winnable, and he could have made hay out of the whole situation if he had cooperated with the cop's request for ID. His principled opposition to giving ID to the cop cost him the game because the ensuing bogus arrest took center stage and didn't allow him to focus his resources, his family support, and the media attention on the behavior of the store.
The core problem is that the client system was compromised and as a result all the SSL in the world won't help. Clueless users will always be with us. It will always be the case that if you send out enough emails with plausible links, someone will click on them.
Banks should make greater use of authentication tokens like those provided by SecurID, because they offer a secure platform that is much more difficult to compromise with fishing attacks.
Analysts are hopeful that the technology now able to identify terrorists with 95% accuracy by their word usage and sentence structure will be able to be repurposed to identify trolls and banned users with equal facility. Funding for the transition to peaceful uses is expected to come from CowboyNeal and the makers of UBBthreads.
Advertising on web sites is a transitional source of revenue, much like registrations for shareware was prior to the open source era. Sites like Wikipedia.org, with phenomenal traffic and a tiny budget, show that ads aren't necessary to support a top ten web site.
I believe that most advertising on web sites is accepted in the blind hope that it will someday be a significant revenue source.
This is the consumer equivalent of the age-old problem in the corporate world of printing something to the wrong printer, something that resulted in many a red face and more than a few leaks of confidential information. It is an information security problem -- how do you prevent a user from erroneously placing confidential information in an insecure space? The problem is the same whether the insecure space is a printer, an extranet site, or a directory structure shared by a file sharing program.
Systems designed for handling a mixture of documents of varying sensitivity deal with this by classifying the documents and then refusing to send secure documents to insecure resources. Sooner or later, one would hope that such systems will become practical for home users, although in a world where people still surf the web from their admin account we have a long way to go.
That P2P was involved is mere happenstance, since many other communications utilities (drive sharing in Windows Networking, for example) can be misconfigured the same way. You would think that AUSA Warma would be smart enough to realize that rather than treat this as a risk categorically unique to P2P.
The blog article is weak on a number of important points.
Open source is part of the landscape and Microsoft is adapting to it. That does not mean that Microsoft "loves" open source. Microsoft could probably be said to "love" any vertical-market application that adds value to the platforms that it offers. It matters not a jot whether said applications are open-source or closed-source.
OpenOffice poses more of a long-term threat to Microsoft than Linux, because Microsoft's revenue stream for Office is less subject to margin erosion than its operating system revenues. Firefox also poses a threat insofar as a common browser platform that is OS independent undermines control of the desktop. Windows and IE provide control of the market, while Exchange and Office provide the revenue. Competition on any of the four fronts is taken seriously.
The cost of running power lines to remote locations can be considerable. Like solar-powered traffic lights and signs, the advantage of this technology is that it provides a means for powering remote traffic controls without the need for extensive cable runs, and without the significant minimum charges from the power company.
The real comparison is to solar, which is an esablished technology. It does not work well in northerly areas where the day length is short in the winter, particularly the many such areas that have a high percentage of overcast days that time of year. But then again, anything at all like a speed bump fares poorly in areas where snowplows are used.
Meanwhile, in the real world, propane and kerosene powered freezers that use the ammonia absorption process. This sort of equipment is widely deployed worldwide in places where electricity is unavailable or unreliable, but where there is sufficient funding for their purchase. They end up being used by missionaries and by medical clinics operating in remote areas.
I have friends who have run these things for years and they are trouble free. The main problem with them is that their thermodynamic efficiency is poor, so they end up producing a great deal more heat than cold. Therefore the fuel costs are higher than the electricity costs would be for the more common phase-change process used in ordinary everyday appliances.
There is nothing special about propane as a heat source. Kerosene is also used though other fuels are plausible. The caloric output of the heat source has to be fairly steady, which makes a wood-fired system more problematic, but still more plausible than a vortex system.
In reality a phase-change compressor driven by bicycle pedals would be more practical, if human-powered refrigeration is the goal. The technology is established and the efficiency has the benefit of decades of development.
I wonder whether the courts will continue their strategy of balancing 1st amendment rights and copyright protection. Though the Grokster ruling was a big win for the RIAA and Hollywood, it left P2P intact as a legitimate technology, with the betamax-like reasoning that it has noninfringing uses.
When freenet becomes common enough, government and industry will have to resort to Old Fashioned Police work, trying to trick file sharers into trusting them, then exploiting that trust in an investigation. I have no doubt that we will see that for highly objectionable content, such as child porn and terrorist communications. It won't be worth it for infringement cases, though.
The real question is whether the courts will be bold enough to make the technology unlawful based on the widespread criminal uses that are sure to develop. Stay tuned.
I use the Creative Commons license for a wide variety of work that I create -- text and photos contributed to Wikipedia, sheet music, and Ogg audio recordings of classical music that I have made.
The first thing to be understood is that Creative Commons has several licenses that it encourages content creators to use. In addition to the noncommercial license the Dvorak has reviewed, there are licenses available that permit commercial use, with attribution, with varying permissions for derived works. I use the "attribution-share alike" license, which is similar in its goals to the GPL but differs in its details.
The second thing to understand is that the licenses the Creative Commons provides are international in scope. Fair use rights and the nature of copyright varies considerably from place to place, and fair use rights in the United States are stronger than those in many other places. The nature of copyright also varies, with some jurisdictions refusing to recognize public domain materials or having a much lower bar for minimum creative effort required to obtain copyright. This has created problems with the mutopia project, which requires international copyright clearance because it has mirrors worldwide. The Creative Commons "public domain license" addresses these issues.
Like any standardized license, the adoption of a standard license text minimizes the problems for aggregation and reuse that can occur when various entities try to develop their own open source licenses, as has happened with so many software projects. Musicians and writers would do well to recognize this and adopt the Creative Commons licenses rather than making their own.
Dvorak places great faith in being able to obtain permissions for copyrighted works he wishes to reuse. That's fairly straightforward for a blog article that's a year old, but is almost impossible for older material still in copyright. I've tried. Often the original copyright holder can't be located, or the copyright has passed to heirs or other successors who aren't interested in relicensing or who want unreasonable fees. I don't want that to happen to creative works that I create. Fifty years from now, if someone else is working on a sheet music project, I want them to be able to use the material I have created as a starting point, regardless of what my heirs or attorneys think.
Finally, the Creative Commons has addressed many of the shortcomings of the GFDL that the FSF has refused to grapple with, among them the requirement for including the 8-page long license text with every copy or derivative work (a problem for someone who wants to reuse a photo for a postcard, for example), and the overly restrictive DRM limitations.
Dvorak's article is short on facts and long on opinions that appear not to be informed by the needs of content creators who are out there today making open-source content using digital media of all kinds.
PDF is a poor answer for scores. In many cases, the PDF files become unnecessarily and unworkably large even for simple scores. As a rule, they do not scale well. They cannot be edited and reposted, and are therefore useless on a wiki or any other collaborative forum. In many cases, I find that PDF scores require fonts or printer capabilities that my laser printer does not have.
That is what they would like you to believe, and it's not true.
There is nothing in copyright law that allows someone to copyright an "edition." The layout of the notes on the page, however carefully done, is a mechanical process that does not meet the content creation standard required by law.
The editorial matter, the foreward, the performance notes, and the like can be copyrighted in legitmate fashion. And an arrangement, rather than an accurate portrayal of the existing work, can be protected by copyright.
There is no law against claiming copyright over public domain works, and it is done all the time with sheet music. Beethoven? Copyrighted. Bach? Copyrighted. They didn't even have copyright laws on the books when much of his music was published.
If sheet music were software and pianos were computers, nobody would stand for this crap. Hopefully, MusicXML, or something equivalent to it, will change that.
I'm skeptical, but hoping for the best with this one.
There is a clear need for a better way to share music notation. At wikipedia, there has never been a consensus so TeX generated by Lilypond or something similar is used. That works poorly, because it is hard to integrate with CGI, and without integration only users who have Lilypond themselves can contribute.
Same set of problems at composerplanet.com, though they are still getting their site together and haven't chosen a strategy. Looks like.PNG and.JPG images will be the de facto standard. Ick.
Lilypond is free, and runs on Linux, but is unlikely to become much of an interchange standard because the UI isn't accessible to the vast majority of musicians, who are as a rule not experts on writing something according to a context-free grammar. Besides, Lilypond is best for typesetting-quality layout, at which it does indeed excel.
Whatever the solution becomes, the ability to share scores with ease will touch off another wave of handwringing among sheet music publishers. I just yesterday received a book with all of Scott Joplin's piano music in it -- all written before 1915 -- and guess what? It says right on the front page that it is against the law to copy them! So far, most musicians don't know any better, but if MusicXML comes to pass, that may all change.
I've done it. It was hard, and we made some money for a while. I'm back to the corporate world for the moment.
You have to find a market that is presently underserved, and where they have some problem that can be solved with software that is worth spending, oh, at least $10,000 or more to solve. You can't build a business out of $500 software sales unless you have a lot of seed capital.
Generally, you need a sales rep that has some sort of background or insight into the market you have chosen. If you want to sell inventory management for shoe stores, you better get a sales rep that used to run a shoe store, or that sold to shoe stores, or had some other industry contact.
There will be dry periods. You can't count on self-funding a startup and paying everyone based only on some custom projects up front. If you can all work for half pay for a year, or if you have a couple people that will work for stock only, then you have better chances. Generally, you can't get investors to put money in early unless someone on your team has a personal track record. Otherwise you have to wait until you have an established revenue stream.
The other comments are germane. You need someone to do support. There will be a lot of on-site work. You will need someone with both software skills and people skills.
Most of the costs are salary and travel. The professional fees, the phone, the fax machine, paper for the copier -- these don't really add up. Not like a market salary for two developers and a sales rep, anyway.
I believe that it is tough right now because most businesses that have a custom software need have already replaced their DOS-based systems. Most are reluctant to change. The overall consolidation in so many businesses means that most of the small corporate clients are much larger than they were 10 years ago. These larger businesses are less willing to deal with smaller, newer software firms.
And there is increasing commoditization of software. Places are willing to have a clerk spend a day or two each month "doing the reports" in Excel. It's hard to sell against that, particularly with the small outfits.
All that said, the deals are still out there. And it beats sitting around the house even if it isn't particularly lucrative.
Slashdot Mirror
Besides, these sorts of systems break down when evaluating a medicated subject. Beta blockers are highly effective at suppressing the involuntary physiological effects of fear. Ask any professional musician. So is alcohol, though unlike beta blockers it has the problematic effect of making you drunk.
Cellular tower equipment doesn't draw significantly less power when there are fewer active calls, so limiting calls to 911 would not extend the battery life by much. I know, I used to do power monitoring systems for the industry.
The problem for carriers is that many sites won't allow generators. Tower sites usually will, although cost is a problem and in California getting a discharge permit from the air quality board is a problem. Getting fuel hauled in can be a problem for some of the more remote sites. Office buildings, apartment buildings, and so on usually will not allow generators because of safety problems with exhaust and fire code compliance problems with fuel storage. Water towers have contamination concerns from fuel storage.
Most cellular carriers up until now have designed noncritical cell sites at tower locations for 2 hours battery time and usually have a transfer switch wired in place for external power from a trailer-mounted generator. Then they have generator trailers at a rate of about one trailer per 10 cell sites. When power losses aren't widespread and the roads are passable, that works OK except that you have techs spending their whole day driving around fueling up generators. Inside a building they will try to get emergency power from the building owner, and depending on the history of the site and the size of the coverage footprint they will increase battery capacity considerably.
There are practical limits to how much battery capacity can be installed, because batteries are big and heavy. And batteries by themselves are not a guarantee, because the HVAC is not ordinarily on an inverter, and on a hot day in a rooftop installation you'll see the site shut down due to overtemp before it shuts down due to the low voltage disconnect for the battery string.
On site generators aren't the perfect answer either. These sites are unattended. Generators don't always start, and unless a site has natural gas as one of its fuel sources (few do) you have a limited fuel supply, usually 3-7 days. With greater reliance on generators, and impassible roads, the fuel haulers may not be able to serve everyone that frequently.
My present employer places great emphasis on intellectual property, and I can't imagine a new hire (or anyone else) being penalized for bringing up the situation described with management. If anything, we'd give them a medal. And yes, we would discard code and rewrite it to resolve IP problems, if we found that the code was a copyright violation.
Other employers aren't like that, especially smaller employers producing products where the code won't be seen by many people, or people working on internal software that will never see the light of day.
Part of the reason small business compliance is lower is that large corporations get volume licenses, which lowers the cost dramatically and reduces the amount of tracking and paperwork required per seat. It also gets them away from the purportedly non-transferable OEM licenses that the BSA is disinclined to count.
The criticism of the BSA isn't rooted in pro-piracy beliefs. The issue is that the BSA uses scare tactics to convince small business owners to sign a contract they shouldn't. The terms of the contract are such that the target ends up paying to purchase software which they have already paid for and are using lawfully and within the provisions of the license agreement.
One of many ways the law enforcement and legal system, with its constitutional protections and other checks and balances, is pushed aside by large corporate interests who have their own laws (shrink wrap agreement), investigators (BSA, RIAA), and courts (mandatory arbitration).
I was one of the people involved in creating the Notability policy. Articles on subjects that are not notable pose numerous problems for Wikipedia.
The wiki process works poorly regarding topics that are not of lasting interest to large numbers of people. Eventualism on Wikipedia, the idea that a stub or poorly written article will ultimately improve, relies upon there being interested individuals who will show up and edit the article based on their interests and knowledge. Subjects that aren't really notable don't draw many editors, and the editors they do draw tend to be connected with the subject in some way: fans, critics, friends, students, and others who are likely to add material which is not neutral and can't be substantiated by solid sources. The test case that led to the creation of notability policy years ago was a self-described surrealist and artist, who had created numerous articles related to his activities that were impossible to refute as unfactual because the references upon which they were purportedly based were local newspapers and magazines that were not available on-line or through even the largest libraries. The problem gets worse with the passage of time, since blogs and minor print publications can disappear without a trace. Hence, the notability policy came to emphasize coverage in national or major regional publications.
Articles on subjects that aren't notable draw the lion's share of formal complaints to the Wikimedia Foundation. This is mainly a problem with biographies, but to a lesser extent occurs with schools, bands, minor political offices, corporations, publications, and so on. These complaints have to be handled responsibly and are a burdensome distraction from the business of the Foundation, requiring the involvement of the most capable volunteers as well as, in some cases, paid staff. Truly notable people and organizations are unlikely to complain about their articles due to the widespread press coverage they already receive.
One of the posters above notes that there are widespread articles on Wikipedia on minor porn stars which do not draw deletions as do minor webcomix. Wikipedia is nothing if not inconsistent in its application of guidelines and policies, so the presence of such counterexamples doesn't mean much except that there aren't as many Wikipedians willing to take the time to police that category of articles.
Indeed, deletion policy is and always has been applied inconsistently. Any deletion decision has a certain amount of randomness because of the nature of the process. That said, there is a strong bias towards keeping stuff. Precious few truly meritorious articles get deleted, and there is a great deal of junk that gets kept only because of luck and the predisposition to keep things.
Too bad they didn't offer database dumps or other convenient means for mirrors to obtain the content.
One might conclude that the Service listened to a recording of the conversation. Many if not most tech support and customer service calls are recorded.
That's easy. Anyone with a modicum of talent and an undergraduate engineering degree from a U.S. college or university can find just as good a job as they would with a graduate degree.
With an engineering degree from a non-U.S. institution, that's not true, especially for jobs in the U.S. Hence, people from many other countries come to the United States to get a graduate degree so they can open the door that a U.S. undergrad degree would open.
The only jobs where a graduate engineering degree is required or will help are teaching positions and very high-level technical jobs which also require unusual talent and decades of experience.
Take a look at what http://www.clearwire.com/ is doing for an idea of pricing for the devices and service and for the kinds of adapters available. The technology they are using is fundamentally very similar to WiMax.
The municipal WiFi players don't have spectrum, but they do have mounting locations and (in some cases) backhaul. Not trivial assets when contemplating a deployment.
The fuckup at the root of all this was that he was pushing his luck with the cop. You have to pick your fights. The fight with the store was a valid one and fundamentally winnable, and he could have made hay out of the whole situation if he had cooperated with the cop's request for ID. His principled opposition to giving ID to the cop cost him the game because the ensuing bogus arrest took center stage and didn't allow him to focus his resources, his family support, and the media attention on the behavior of the store.
The core problem is that the client system was compromised and as a result all the SSL in the world won't help. Clueless users will always be with us. It will always be the case that if you send out enough emails with plausible links, someone will click on them.
Banks should make greater use of authentication tokens like those provided by SecurID, because they offer a secure platform that is much more difficult to compromise with fishing attacks.
Analysts are hopeful that the technology now able to identify terrorists with 95% accuracy by their word usage and sentence structure will be able to be repurposed to identify trolls and banned users with equal facility. Funding for the transition to peaceful uses is expected to come from CowboyNeal and the makers of UBBthreads.
Wikipedia.org, kernel.org, openoffice,org, microsoft.com
Advertising on web sites is a transitional source of revenue, much like registrations for shareware was prior to the open source era. Sites like Wikipedia.org, with phenomenal traffic and a tiny budget, show that ads aren't necessary to support a top ten web site.
I believe that most advertising on web sites is accepted in the blind hope that it will someday be a significant revenue source.
This is the consumer equivalent of the age-old problem in the corporate world of printing something to the wrong printer, something that resulted in many a red face and more than a few leaks of confidential information. It is an information security problem -- how do you prevent a user from erroneously placing confidential information in an insecure space? The problem is the same whether the insecure space is a printer, an extranet site, or a directory structure shared by a file sharing program.
Systems designed for handling a mixture of documents of varying sensitivity deal with this by classifying the documents and then refusing to send secure documents to insecure resources. Sooner or later, one would hope that such systems will become practical for home users, although in a world where people still surf the web from their admin account we have a long way to go.
That P2P was involved is mere happenstance, since many other communications utilities (drive sharing in Windows Networking, for example) can be misconfigured the same way. You would think that AUSA Warma would be smart enough to realize that rather than treat this as a risk categorically unique to P2P.
OpenOffice poses more of a long-term threat to Microsoft than Linux, because Microsoft's revenue stream for Office is less subject to margin erosion than its operating system revenues. Firefox also poses a threat insofar as a common browser platform that is OS independent undermines control of the desktop. Windows and IE provide control of the market, while Exchange and Office provide the revenue. Competition on any of the four fronts is taken seriously.
The real comparison is to solar, which is an esablished technology. It does not work well in northerly areas where the day length is short in the winter, particularly the many such areas that have a high percentage of overcast days that time of year. But then again, anything at all like a speed bump fares poorly in areas where snowplows are used.
I have friends who have run these things for years and they are trouble free. The main problem with them is that their thermodynamic efficiency is poor, so they end up producing a great deal more heat than cold. Therefore the fuel costs are higher than the electricity costs would be for the more common phase-change process used in ordinary everyday appliances.
There is nothing special about propane as a heat source. Kerosene is also used though other fuels are plausible. The caloric output of the heat source has to be fairly steady, which makes a wood-fired system more problematic, but still more plausible than a vortex system.
In reality a phase-change compressor driven by bicycle pedals would be more practical, if human-powered refrigeration is the goal. The technology is established and the efficiency has the benefit of decades of development.
When freenet becomes common enough, government and industry will have to resort to Old Fashioned Police work, trying to trick file sharers into trusting them, then exploiting that trust in an investigation. I have no doubt that we will see that for highly objectionable content, such as child porn and terrorist communications. It won't be worth it for infringement cases, though.
The real question is whether the courts will be bold enough to make the technology unlawful based on the widespread criminal uses that are sure to develop. Stay tuned.
The first thing to be understood is that Creative Commons has several licenses that it encourages content creators to use. In addition to the noncommercial license the Dvorak has reviewed, there are licenses available that permit commercial use, with attribution, with varying permissions for derived works. I use the "attribution-share alike" license, which is similar in its goals to the GPL but differs in its details.
The second thing to understand is that the licenses the Creative Commons provides are international in scope. Fair use rights and the nature of copyright varies considerably from place to place, and fair use rights in the United States are stronger than those in many other places. The nature of copyright also varies, with some jurisdictions refusing to recognize public domain materials or having a much lower bar for minimum creative effort required to obtain copyright. This has created problems with the mutopia project, which requires international copyright clearance because it has mirrors worldwide. The Creative Commons "public domain license" addresses these issues.
Like any standardized license, the adoption of a standard license text minimizes the problems for aggregation and reuse that can occur when various entities try to develop their own open source licenses, as has happened with so many software projects. Musicians and writers would do well to recognize this and adopt the Creative Commons licenses rather than making their own.
Dvorak places great faith in being able to obtain permissions for copyrighted works he wishes to reuse. That's fairly straightforward for a blog article that's a year old, but is almost impossible for older material still in copyright. I've tried. Often the original copyright holder can't be located, or the copyright has passed to heirs or other successors who aren't interested in relicensing or who want unreasonable fees. I don't want that to happen to creative works that I create. Fifty years from now, if someone else is working on a sheet music project, I want them to be able to use the material I have created as a starting point, regardless of what my heirs or attorneys think.
Finally, the Creative Commons has addressed many of the shortcomings of the GFDL that the FSF has refused to grapple with, among them the requirement for including the 8-page long license text with every copy or derivative work (a problem for someone who wants to reuse a photo for a postcard, for example), and the overly restrictive DRM limitations.
Dvorak's article is short on facts and long on opinions that appear not to be informed by the needs of content creators who are out there today making open-source content using digital media of all kinds.
For one thing, I've never received anything from Microsoft in a tar file.
PDF is a poor answer for scores. In many cases, the PDF files become unnecessarily and unworkably large even for simple scores. As a rule, they do not scale well. They cannot be edited and reposted, and are therefore useless on a wiki or any other collaborative forum. In many cases, I find that PDF scores require fonts or printer capabilities that my laser printer does not have.
There is nothing in copyright law that allows someone to copyright an "edition." The layout of the notes on the page, however carefully done, is a mechanical process that does not meet the content creation standard required by law.
The editorial matter, the foreward, the performance notes, and the like can be copyrighted in legitmate fashion. And an arrangement, rather than an accurate portrayal of the existing work, can be protected by copyright.
There is no law against claiming copyright over public domain works, and it is done all the time with sheet music. Beethoven? Copyrighted. Bach? Copyrighted. They didn't even have copyright laws on the books when much of his music was published.
If sheet music were software and pianos were computers, nobody would stand for this crap. Hopefully, MusicXML, or something equivalent to it, will change that.
There is a clear need for a better way to share music notation. At wikipedia, there has never been a consensus so TeX generated by Lilypond or something similar is used. That works poorly, because it is hard to integrate with CGI, and without integration only users who have Lilypond themselves can contribute.
Same set of problems at composerplanet.com, though they are still getting their site together and haven't chosen a strategy. Looks like .PNG and .JPG images will be the de facto standard. Ick.
Lilypond is free, and runs on Linux, but is unlikely to become much of an interchange standard because the UI isn't accessible to the vast majority of musicians, who are as a rule not experts on writing something according to a context-free grammar. Besides, Lilypond is best for typesetting-quality layout, at which it does indeed excel.
Whatever the solution becomes, the ability to share scores with ease will touch off another wave of handwringing among sheet music publishers. I just yesterday received a book with all of Scott Joplin's piano music in it -- all written before 1915 -- and guess what? It says right on the front page that it is against the law to copy them! So far, most musicians don't know any better, but if MusicXML comes to pass, that may all change.
You have to find a market that is presently underserved, and where they have some problem that can be solved with software that is worth spending, oh, at least $10,000 or more to solve. You can't build a business out of $500 software sales unless you have a lot of seed capital.
Generally, you need a sales rep that has some sort of background or insight into the market you have chosen. If you want to sell inventory management for shoe stores, you better get a sales rep that used to run a shoe store, or that sold to shoe stores, or had some other industry contact.
There will be dry periods. You can't count on self-funding a startup and paying everyone based only on some custom projects up front. If you can all work for half pay for a year, or if you have a couple people that will work for stock only, then you have better chances. Generally, you can't get investors to put money in early unless someone on your team has a personal track record. Otherwise you have to wait until you have an established revenue stream.
The other comments are germane. You need someone to do support. There will be a lot of on-site work. You will need someone with both software skills and people skills.
Most of the costs are salary and travel. The professional fees, the phone, the fax machine, paper for the copier -- these don't really add up. Not like a market salary for two developers and a sales rep, anyway.
I believe that it is tough right now because most businesses that have a custom software need have already replaced their DOS-based systems. Most are reluctant to change. The overall consolidation in so many businesses means that most of the small corporate clients are much larger than they were 10 years ago. These larger businesses are less willing to deal with smaller, newer software firms.
And there is increasing commoditization of software. Places are willing to have a clerk spend a day or two each month "doing the reports" in Excel. It's hard to sell against that, particularly with the small outfits.
All that said, the deals are still out there. And it beats sitting around the house even if it isn't particularly lucrative.