I congratulate you on your recognition of the advantages of Debian. I'm a bit of a zealot there myself after spending a year on tour with every other distro I could reasonably find.
Now, about your case.
Companies have a philosophy that they want to have support from an organization rather than an individual. This makes RedHat and SuSE very appealing. Similarly this probably had some weight when they were selecting IBM for their IT contractor. This is also what appears to make Debian very un-appealing. Debian doesn't have the store-front appearance of being a big presence that won't go away and it super easy to manage.
Personally I can't imagine a more permenent presence than Debian considering the number of mirrors that they have and the difficulty it will be to put them out of business. But this will require a paradyme shift from the company.
I think there is some supporting arguements for Debian that you might be able to find from Perens and also from the fact that many companies (Lindows, Knoppix, Libranet...) base their product on the core that is Debian. This says a lot for Debian.
I think a lot of the arguement comes down to these elements:
You can support Debian for the same cost that you do everything else and the support expertise is In-House.
Hopefully you and the company have a solid belief that neither of you wish a voluntary seperation
Even companies can go bankrupt overnight giving them the appearance that they too where hit by a beer truck.
You are embarking on a very progressive project here and their conservatism may be your biggest enemy.
HTML::Mason is definitely as fast or faster than PHP under any circumstances. Test it.
Why did Yahoo pick PHP over HTML::Mason? Because you can find a MySQL+PHP developer hanging out at every cybercafe or war-chalked street corner. Their commonality is appealing to big companies. This is exactly why MCSE and Java developers are more likely to get a development project than anyone else. Because they are easier to replace, not necessarily better.
Oh, I never claimed to be an expert. But I've used both.
I think you've jumped into a few really bad assumptions on this one.
First, you obviously missed the part about not taking a year off to hand code it.
Second, is MySQL and PHP always the best solution to a problem? I find it useful, but for someone who is trying to get performance out of a server, I seriously doubt that MySQL + PHP is going to be the right answer. I've found the work I have done on a HTML::Mason web server to be easily 10 times faster than it's PHP counterpart.
Third, I don't know much about the project, but I think the question was in regards to the RDMS and not the web pages themselves. I am going to assume that most of the work here is in the database structure.
If this is the case, then even a hand migration shouldn't be that difficult to do. All you have to do is write the scripts to create the tables based on their current definition and procede to dump/load the data. It would be well worth it to develop scripts to do all of this and test them out so that you can migrate the entire back end in one shot.
I don't have an answer to a easy print interface, but I can venture some guesses on what might work. OpenOffice or StarOffice has some kind of ODBC connectivity and maybe that can be capitalized with it's capabilities. Alternatively, and maybe lastly, perl has some printing capabilities that would make for somewhat simple reports but at very high speed.
That's fine with me. At least this way we don't have to cater to all the idiots who want to have everything at their fingertips, including root access. And don't you dare make them login!
Linux doesn't have to be popular to be the best. But then neither do any of the BSD's.
As a correlary how can compete against someone who has the millions to subsidize hardware to assist their software inclusion and help tip the hands of the decision makers?
You have to include the support for Linux when you consider these equations. While you can go to San Francisco, CA and find a lot of potential Linux Admins, how many might you expect to find on the streets of Sri Lanka or Haiti?
There will be a more truthful display of TCO when really poor nations consider the software. It may not be all about proprietary mistrust of The Man.
They did not attempt to do business with Microsoft. They only sold them their wares and walked away. That happens a lot, that's what Microsoft would prefer. Find me someone who has succeeded in a business partnership or any joint efforts who didn't get it in the end.
So, you posting on the merits of Debian vs Fedora is On Topic and non-inflammatory, but me posting on the merits of Mandrake vs Debian (and debunking your FUD) is not? This makes just as much sense as the rest of your arguments (which have no real facts to support them since your assumptions are invalid).
Not quite. But Debian is completely based on being a free distribution whereas the others are loosely based on being a free distribution. I cannot get RHEL for free. But I can get a less stable variant that's Fedora.
The point is the Debian is made to become the basis of other distributions. Knoppix, Libranet are working examples. K12-LTSP is another good example of what I mean as a customized distribution that is built upon another core (in this case, Fedora)
If other packages can be build on Mandrake, Fedora, then fine -- but I don't think that they will remain free in the broadest sense of the words
maybe not, but how many millions of viagra ads do you have to troll through to find a real email? And how would you handle something like a mailing list based contact who wants to contact you directly?
Look, I don't mean to be a wanker here, but all of you who are all gung-ho about SPF and happy-happy-happy about working with some whizz-bang IETF project to bring on board the likes of AOL and MSFT are... gullible and naive. It's amazing how really smart people can do some really dumb things.
No one in the history of business has managed to engage in any kind of relationship what-so-ever with Microsoft Corporation and not gotten screwed in the end. No exceptions exist to date.
How many times do we have to go through this in order to understand this fact?
Any time that you do any business with Microsoft Corporation, either with the intention of cooperative standards development or for money, you will always, always, always get the short end of the stick.
Microsoft is very simple. They are a parasitic entity that will not have any engagement with other entities unless there is the distinct advantage that they will walk away with more than you do.
Now that you've reverted SPF so some Classic-SPF that has already been identified to have loop-holes in the technology to allow spammers to hit you again, what are you going to do? Sell your next of kin to Bill Gates?
I'm not surprised at all by MSFT's actions. But I'm postively disgusted that anyone would be some stupid as to think for one minute that it might work out OK.
Everytime you work with windows you hurt Open Source Development
I'll sound a lot like Perens here, but only because I firmly believe that he is absolutely correct in many aspects.
A project like these should be directed towards the few remaining truely free operating system installations that exist. The two largest being Debian and Gentoo. The installations that charge for their services or software (SuSE, RedHat) should not be supported on these projects in the future.
Why?
Debian is showing itself to have a highly unique feature that none others can offer. Debian is being built upon by others to create a greater variety of User Installations then what can be offered by Fedora alone. What I mean by this is the existence of installations like:
Lindows (or whatever they renamed it)
Knoppix
Libranet (I think)
Are all based entirely upon the Debian architecture as the launch point of the distribution.
This unique property of Debian, over the RPM based installations of SuSE, Fedora, Mandrake, allows others to build specialized OS packages using a common baseline. Gentoo has similar properties, but I am not personally aware of anyone attempting to build additional distributions from the base architecture of Gentoo.
Furthermore, with the continued arguement that the Open Source and/or Free software has advanced to such a point that it is really a viable solution to so many software needs, that to continue Open Source efforts that are based on installations which are not entirely free, is potentially a less-than-optimum use of resources
We must be careful to avoid accidentally supporting a software distrubition which cannot be continued as a useful, practical, and free. While Fedora may be free, it is arguably not practically useful or stable for all applications. After all, it is recognized as the bleeding edge version of the rather expensive RedHat version.
If you put together a really great bang up software package which is entirely dependant upon a purchased software package of any origin you are following the precursor avenue of supporting the likes of Microsoft. Microsoft's distaste, for me, is the fantastically high fees applied to get continued support and use of the software.
If the only thing we ever do is publish all these really great docs on how to do stuff using software which comes in either bleeding edge or expense to purchase, we are doing little to really promote the concept of Open Source and Free Software.
However, everything presented in the howto is actually Free Software and should be written up on software installation architecture which is not a limitation (non-free) but compatable with the same software.
As for the merits of Debian versus Gentoo and other distributions, any discussion in that vein would probably be Off Topic and inflammatory.
I think it would be more interesting if each company was allowed a limited number of patents per year. Then they would have to prioritize the territory they want to claim.
Individuals would be given a similar limit. Given that companies typically own all the patents an employee might make in a year, this would apply to the company limit.
Any patent that the individual made during that time would be exclusively theirs and might be granted to them by the company for lesser patents. The idea being the the company would claim the best of the patents for themselves and then allow the individuals to apply for their own claims with some kind of free use license by the employer for the term of employment.
Might make for an interesting mix of patents and put a limit on how many patents are hitting the USPTO.
For those of us who patent things like the Clapper and intermittent windshield wipers, it would have zero impact.
You forget that even the most basic client today has more than enough horse power to run the graphics rendering and the RAM to do the same. It's actually difficult to find a cheap computer that can't manage the 64-128MB of RAM and the video necessary to support client side graphics as mentioned in this article
The only possible exception would be the high definition 3D graphics of gaming, but then who would want to attempt graphics display over a network in that environment anyways!
How the F!! can something like Java "swoosh" past C/C++?
I seriously doubt that your experiences are based on anything which does not include a heavy contribution of fine Ale.
If Java 1.5 swooshes then why not just write the entire OS in Java, or better yet, embed the Java Machine onto a chip! Oh wait... they tried that once.
yes, but the difference between you and the fuckwit testing the login passwords is you are not attempting to impersonate anyone you are not (ie, root). You are simply walking around the car on the readily accessable avenues of access (ie: ftp, http) which are intended for public access or general consumption as differs from private access (locked door) requiring a login username and password (as a key is to a ssh/telnet account)
Possible the best solution I can offer is the Linux Terminal Server Project. It's really quite functional and can be used transparently with Gnome, KDE or whatever.
The concept is a true thin client configuration with a networked server providing all the applications. It's been run over ethernet, fast-ethernet, and even wireless 802.11b networks. One of the great advantages with it is that you do not require updates/upgrades of client hardware/software since there isn't any. It can also be run diskless and/or on very small machine (VIA EPIA 533 is one that does well and takes 32W of power).
disklessworkstations.com is a company put together by the LTSP owners to sell the LTSP client appliances. They can provide more information and hardware/software/technical support if needed. Jim McQuillan is a very cool guy.
When will they start building chips that have no support for 32-bit software?
I mean, we don't really support 8-bit today and I'm not really sure if we even have support for 16-bit software in todays CPU's. So when are we going to rid our selves of the legacy throw backs?
I've tried gentoo several times on a variety of machines and have consistently had two results:
The installation takes several days from step one, but when it's done, it's in fairly functional condition. I give it a "B".
Updates change a lot of stuff. The process of resolving conflicts between new and existing config files has fucked every installation I've ever touched. When you have to attempt to review >150 configuration files every few days and the methods for doing so are extremely bad, you're bound to screw the installation. I give it a very solid "F".
It would do Gentoo a lot of good to change their method of review/update of the configuration files. It's not terminal friendly and highly prone to massive errors.
I would like to recommend examining the methodology of Debian for their config file resolution process. It's much harder to really screw things up.
And please don't give me this holistic philosophy about these are the tools and get over it. The approach for resolving files is horrific on a good day. It's a barrier. It's going to keep Gentoo in the category of "toy" distributions for a very long time.
If you can't provide some reasonable level of useability then you have a limited success distro. And please bear in mind, I'm comparing Gentoo to Slackware and Debian, not Suse and RedHat.
IMHO this is bad. If the development process can no longer isolate between what's stable and what's under development, then they probably are incapable of measuring the stability of the current release product.
But I wonder how Debian and Gentoo will handle this since they aren't the stereotypical Corporation who stabilizes the kernel prior to release. Debian at least, does a very slim job of customizing the Linux Kernal when compared to RedHat, Mandrake, and SuSE.
Does this imply that the LKML has decided to abandon their origins of Free and just hack code and let someone else actually worry about a finished workable product? Sounds like they are kind of blowing off their community.
Or is the community filling up with whinnie-assed whimps who don't know what the meaning of "make clean" means?
Did anyone actually visit these links before they started bitching about GPL and debating Law (a subject of which they know nothing about but can find many links?)
This posted software was issued by a disgruntled employee with virii infesting therein, or at least so claims some of the pages you've found..
As near as I can tell, linksys has a problem with GPL, but I think this article is kind of a vague attempt to discredit a company based on a whiff of GPL-fear without a respectable trail of evidence to support it.
You present an excellent summary of the previous posting putting it in relation to a socialist or individual term of freedom. However, I have to disagree with you on the notion that GPL can be considered a Socialist license.
While you have the freedom to be an asshole, you do not have the freedom to exercise your assholeness upon others.
You speak entirely in terms of Capitalism and not in the terms of Politicial structures. GPL may be considered not so much Socialist as Libertarian in political nature.
You are asking for a non-inheritable licensing structure. When I create something from a GPL license, the license applies a rule that my code shall inherit this same openness as the code base that I drew from (your source code). This means that while I can draw code from a community of free software, I am not allowed to lock up my derivative work behind closed doors such that it is not free in some/most/all aspects, depending on how I would like to spin it.
You are using a bizarre notion of freedom which is non-transferrable. I do not have the right to remain free and simultaeneously enslave others. The USA decided that in the 1870's
In a commercial sense, you are asking for that same right to be extended into the business community. You are asking for the right to use software from other resources without restrictions or limitations, making you the Free Customer. But then you want to apply a limitation upon your customers that prevent them from seeing anything about your code or using it or modifying it in some/any way. This makes your customers not as free in their exercise of the use of the product as you yourself have benefitted to get to where you are.
What makes the GPL so important is that it reduces software to a commodity which becomes so ubiquitous and easy to acquire that the economy must change it's course to other goods and activities. This is the fundamental cornerstone of so many companies that are using the likes of Debian to provide a Service rather than a Product.
To simply want to take someone elses free software that they have provided and use it for your own profit without extending that offer of free use it really isn't much different from strip mining resources from public land.
Will they offer static IP's or is that going to be extra?
Now that I think about it, I was in a meeting with Verizon where there discussed having IP addresses for their telephone network, but they were planning on putting everything into one massive subnet. If this is the same network then you might not want it. It might make going onto the internet really painful.
Slashdot Mirror
I congratulate you on your recognition of the advantages of Debian. I'm a bit of a zealot there myself after spending a year on tour with every other distro I could reasonably find.
Now, about your case.
Companies have a philosophy that they want to have support from an organization rather than an individual. This makes RedHat and SuSE very appealing. Similarly this probably had some weight when they were selecting IBM for their IT contractor. This is also what appears to make Debian very un-appealing. Debian doesn't have the store-front appearance of being a big presence that won't go away and it super easy to manage.
Personally I can't imagine a more permenent presence than Debian considering the number of mirrors that they have and the difficulty it will be to put them out of business. But this will require a paradyme shift from the company.
I think there is some supporting arguements for Debian that you might be able to find from Perens and also from the fact that many companies (Lindows, Knoppix, Libranet...) base their product on the core that is Debian. This says a lot for Debian.
I think a lot of the arguement comes down to these elements:
You are embarking on a very progressive project here and their conservatism may be your biggest enemy.
I wish you luck!
Wow.... An AC dripping with sarcasm
HTML::Mason is definitely as fast or faster than PHP under any circumstances. Test it.
Why did Yahoo pick PHP over HTML::Mason? Because you can find a MySQL+PHP developer hanging out at every cybercafe or war-chalked street corner. Their commonality is appealing to big companies. This is exactly why MCSE and Java developers are more likely to get a development project than anyone else. Because they are easier to replace, not necessarily better.
Oh, I never claimed to be an expert. But I've used both.
I think you've jumped into a few really bad assumptions on this one.
First, you obviously missed the part about not taking a year off to hand code it.
Second, is MySQL and PHP always the best solution to a problem? I find it useful, but for someone who is trying to get performance out of a server, I seriously doubt that MySQL + PHP is going to be the right answer. I've found the work I have done on a HTML::Mason web server to be easily 10 times faster than it's PHP counterpart.
Third, I don't know much about the project, but I think the question was in regards to the RDMS and not the web pages themselves. I am going to assume that most of the work here is in the database structure.
If this is the case, then even a hand migration shouldn't be that difficult to do. All you have to do is write the scripts to create the tables based on their current definition and procede to dump/load the data. It would be well worth it to develop scripts to do all of this and test them out so that you can migrate the entire back end in one shot.
I don't have an answer to a easy print interface, but I can venture some guesses on what might work. OpenOffice or StarOffice has some kind of ODBC connectivity and maybe that can be capitalized with it's capabilities. Alternatively, and maybe lastly, perl has some printing capabilities that would make for somewhat simple reports but at very high speed.
That's fine with me. At least this way we don't have to cater to all the idiots who want to have everything at their fingertips, including root access. And don't you dare make them login!
Linux doesn't have to be popular to be the best. But then neither do any of the BSD's.
As a correlary how can compete against someone who has the millions to subsidize hardware to assist their software inclusion and help tip the hands of the decision makers?
You have to include the support for Linux when you consider these equations. While you can go to San Francisco, CA and find a lot of potential Linux Admins, how many might you expect to find on the streets of Sri Lanka or Haiti?
There will be a more truthful display of TCO when really poor nations consider the software. It may not be all about proprietary mistrust of The Man.
They did not attempt to do business with Microsoft. They only sold them their wares and walked away. That happens a lot, that's what Microsoft would prefer. Find me someone who has succeeded in a business partnership or any joint efforts who didn't get it in the end.
Not quite. But Debian is completely based on being a free distribution whereas the others are loosely based on being a free distribution. I cannot get RHEL for free. But I can get a less stable variant that's Fedora.
The point is the Debian is made to become the basis of other distributions. Knoppix, Libranet are working examples. K12-LTSP is another good example of what I mean as a customized distribution that is built upon another core (in this case, Fedora)
If other packages can be build on Mandrake, Fedora, then fine -- but I don't think that they will remain free in the broadest sense of the words
postfix
maybe not, but how many millions of viagra ads do you have to troll through to find a real email? And how would you handle something like a mailing list based contact who wants to contact you directly?
Your rocket is in trouble
What do you think?
They don't care about a few millions of dollars a year in this crud. It's all a tax write off to them.
They prefer raping over fighting
Look, I don't mean to be a wanker here, but all of you who are all gung-ho about SPF and happy-happy-happy about working with some whizz-bang IETF project to bring on board the likes of AOL and MSFT are ... gullible and naive. It's amazing how really smart people can do some really dumb things.
How many times do we have to go through this in order to understand this fact?
Any time that you do any business with Microsoft Corporation, either with the intention of cooperative standards development or for money, you will always, always, always get the short end of the stick.
Microsoft is very simple. They are a parasitic entity that will not have any engagement with other entities unless there is the distinct advantage that they will walk away with more than you do.
Now that you've reverted SPF so some Classic-SPF that has already been identified to have loop-holes in the technology to allow spammers to hit you again, what are you going to do? Sell your next of kin to Bill Gates?
I'm not surprised at all by MSFT's actions. But I'm postively disgusted that anyone would be some stupid as to think for one minute that it might work out OK.
Everytime you work with windows you hurt Open Source Development
I'll sound a lot like Perens here, but only because I firmly believe that he is absolutely correct in many aspects.
A project like these should be directed towards the few remaining truely free operating system installations that exist. The two largest being Debian and Gentoo. The installations that charge for their services or software (SuSE, RedHat) should not be supported on these projects in the future.
Why?
Debian is showing itself to have a highly unique feature that none others can offer. Debian is being built upon by others to create a greater variety of User Installations then what can be offered by Fedora alone. What I mean by this is the existence of installations like:
- Lindows (or whatever they renamed it)
- Knoppix
- Libranet (I think)
Are all based entirely upon the Debian architecture as the launch point of the distribution.This unique property of Debian, over the RPM based installations of SuSE, Fedora, Mandrake, allows others to build specialized OS packages using a common baseline. Gentoo has similar properties, but I am not personally aware of anyone attempting to build additional distributions from the base architecture of Gentoo.
Furthermore, with the continued arguement that the Open Source and/or Free software has advanced to such a point that it is really a viable solution to so many software needs, that to continue Open Source efforts that are based on installations which are not entirely free, is potentially a less-than-optimum use of resources
We must be careful to avoid accidentally supporting a software distrubition which cannot be continued as a useful, practical, and free. While Fedora may be free, it is arguably not practically useful or stable for all applications. After all, it is recognized as the bleeding edge version of the rather expensive RedHat version.
If you put together a really great bang up software package which is entirely dependant upon a purchased software package of any origin you are following the precursor avenue of supporting the likes of Microsoft. Microsoft's distaste, for me, is the fantastically high fees applied to get continued support and use of the software.
If the only thing we ever do is publish all these really great docs on how to do stuff using software which comes in either bleeding edge or expense to purchase, we are doing little to really promote the concept of Open Source and Free Software.
However, everything presented in the howto is actually Free Software and should be written up on software installation architecture which is not a limitation (non-free) but compatable with the same software.
As for the merits of Debian versus Gentoo and other distributions, any discussion in that vein would probably be Off Topic and inflammatory.
I think it would be more interesting if each company was allowed a limited number of patents per year. Then they would have to prioritize the territory they want to claim.
Individuals would be given a similar limit. Given that companies typically own all the patents an employee might make in a year, this would apply to the company limit.
Any patent that the individual made during that time would be exclusively theirs and might be granted to them by the company for lesser patents. The idea being the the company would claim the best of the patents for themselves and then allow the individuals to apply for their own claims with some kind of free use license by the employer for the term of employment.
Might make for an interesting mix of patents and put a limit on how many patents are hitting the USPTO.
For those of us who patent things like the Clapper and intermittent windshield wipers, it would have zero impact.
You forget that even the most basic client today has more than enough horse power to run the graphics rendering and the RAM to do the same. It's actually difficult to find a cheap computer that can't manage the 64-128MB of RAM and the video necessary to support client side graphics as mentioned in this article
The only possible exception would be the high definition 3D graphics of gaming, but then who would want to attempt graphics display over a network in that environment anyways!
How the F!! can something like Java "swoosh" past C/C++?
I seriously doubt that your experiences are based on anything which does not include a heavy contribution of fine Ale.
If Java 1.5 swooshes then why not just write the entire OS in Java, or better yet, embed the Java Machine onto a chip! Oh wait... they tried that once.
yes, but the difference between you and the fuckwit testing the login passwords is you are not attempting to impersonate anyone you are not (ie, root). You are simply walking around the car on the readily accessable avenues of access (ie: ftp, http) which are intended for public access or general consumption as differs from private access (locked door) requiring a login username and password (as a key is to a ssh/telnet account)
Possible the best solution I can offer is the Linux Terminal Server Project. It's really quite functional and can be used transparently with Gnome, KDE or whatever.
The concept is a true thin client configuration with a networked server providing all the applications. It's been run over ethernet, fast-ethernet, and even wireless 802.11b networks. One of the great advantages with it is that you do not require updates/upgrades of client hardware/software since there isn't any. It can also be run diskless and/or on very small machine (VIA EPIA 533 is one that does well and takes 32W of power).
disklessworkstations.com is a company put together by the LTSP owners to sell the LTSP client appliances. They can provide more information and hardware/software/technical support if needed. Jim McQuillan is a very cool guy.
When will they start building chips that have no support for 32-bit software?
I mean, we don't really support 8-bit today and I'm not really sure if we even have support for 16-bit software in todays CPU's. So when are we going to rid our selves of the legacy throw backs?
You forgot the classic www.wwww.com
And why not make RCS required?
I've tried gentoo several times on a variety of machines and have consistently had two results:
- The installation takes several days from step one, but when it's done, it's in fairly functional condition. I give it a "B".
- Updates change a lot of stuff. The process of resolving conflicts between new and existing config files has fucked every installation I've ever touched. When you have to attempt to review >150 configuration files every few days and the methods for doing so are extremely bad, you're bound to screw the installation. I give it a very solid "F".
It would do Gentoo a lot of good to change their method of review/update of the configuration files. It's not terminal friendly and highly prone to massive errors.I would like to recommend examining the methodology of Debian for their config file resolution process. It's much harder to really screw things up.
And please don't give me this holistic philosophy about these are the tools and get over it. The approach for resolving files is horrific on a good day. It's a barrier. It's going to keep Gentoo in the category of "toy" distributions for a very long time.
If you can't provide some reasonable level of useability then you have a limited success distro. And please bear in mind, I'm comparing Gentoo to Slackware and Debian, not Suse and RedHat.
IMHO this is bad. If the development process can no longer isolate between what's stable and what's under development, then they probably are incapable of measuring the stability of the current release product.
But I wonder how Debian and Gentoo will handle this since they aren't the stereotypical Corporation who stabilizes the kernel prior to release. Debian at least, does a very slim job of customizing the Linux Kernal when compared to RedHat, Mandrake, and SuSE.
Does this imply that the LKML has decided to abandon their origins of Free and just hack code and let someone else actually worry about a finished workable product? Sounds like they are kind of blowing off their community.
Or is the community filling up with whinnie-assed whimps who don't know what the meaning of "make clean" means?
Did anyone actually visit these links before they started bitching about GPL and debating Law (a subject of which they know nothing about but can find many links?)
This posted software was issued by a disgruntled employee with virii infesting therein, or at least so claims some of the pages you've found..
As near as I can tell, linksys has a problem with GPL, but I think this article is kind of a vague attempt to discredit a company based on a whiff of GPL-fear without a respectable trail of evidence to support it.
You present an excellent summary of the previous posting putting it in relation to a socialist or individual term of freedom. However, I have to disagree with you on the notion that GPL can be considered a Socialist license.
While you have the freedom to be an asshole, you do not have the freedom to exercise your assholeness upon others.
You speak entirely in terms of Capitalism and not in the terms of Politicial structures. GPL may be considered not so much Socialist as Libertarian in political nature.
You are asking for a non-inheritable licensing structure. When I create something from a GPL license, the license applies a rule that my code shall inherit this same openness as the code base that I drew from (your source code). This means that while I can draw code from a community of free software, I am not allowed to lock up my derivative work behind closed doors such that it is not free in some/most/all aspects, depending on how I would like to spin it.
You are using a bizarre notion of freedom which is non-transferrable. I do not have the right to remain free and simultaeneously enslave others. The USA decided that in the 1870's
In a commercial sense, you are asking for that same right to be extended into the business community. You are asking for the right to use software from other resources without restrictions or limitations, making you the Free Customer. But then you want to apply a limitation upon your customers that prevent them from seeing anything about your code or using it or modifying it in some/any way. This makes your customers not as free in their exercise of the use of the product as you yourself have benefitted to get to where you are.
What makes the GPL so important is that it reduces software to a commodity which becomes so ubiquitous and easy to acquire that the economy must change it's course to other goods and activities. This is the fundamental cornerstone of so many companies that are using the likes of Debian to provide a Service rather than a Product.
To simply want to take someone elses free software that they have provided and use it for your own profit without extending that offer of free use it really isn't much different from strip mining resources from public land.
I'm still waiting for DSL to show up.
Will they offer static IP's or is that going to be extra?
Now that I think about it, I was in a meeting with Verizon where there discussed having IP addresses for their telephone network, but they were planning on putting everything into one massive subnet. If this is the same network then you might not want it. It might make going onto the internet really painful.