Ooohh.. you mean those silly programs that ran in DOS that you could download from all your local dial-in bulletin boards at 14400 baud? You know.. before Linux and the whole GPL thing caught on.. I heard rumors that some people are dumb enough to still use 'shareware' in that 'other' operating system.
Perhaps a better question is: Why are people wasting their time trying to make a few dollars off poorly written programs when instead they could contribute to Open Source projects and actually have some influence and get their name out? Worse yet are those Windows freeware people who think they'll be more '31337' if they keep their amazing h4x0r c0d3z as t0p s3cr37!!!
On a more positive note, that's a lot of potentially valuable programmers that could be converted from the 'dark side' (-:
"It's a machine, Scroter, it doesn't get pissed off. It doesn't get happy, it doesn't get sad, it doesn't laugh at your jokes... It just runs programs!!"
Seriously, folks... Invasion by robots that we created?! Considering that these are the same crack pots proposing unsupportable theories of global warming just so they can get research grants with our tax dollars? AI is called AI for a reason. It's artificial. It's not human. The best it can do is make decisions based on a knowledge base and a set of criteria with varying order of importance. In other words, it's deterministic at any given time. Sure, you can use random number generation, but if RNG become a significant enough factor to really change decisions from deterministic ones, then most decisions will not truly meet the criteria. Machines can't have a soul.
And if you really believe that human behavior is dictated solely by our brains, then we are already 'robots' but just fully biological ones. If that is the case, then what does it matter whether or not machines take over the world? If we're just machines, life and survival has no meaning to begin with. This also kinda fits in with the argument that 'if consciousness is just an illusion, then how can we make that statement?'
Man is the best computer we can put aboard a spacecraft... and the only one that can be mass produced with unskilled labor. -- Wernher von Braun
Yes, having a computer in the kitchen is great for things like: keeping your grocery list, displaying recipes, setting multiple timers for cooking times, keeping track of nutritional contents, etc. But who on earth would watch a movie?! If your hands are that free, you're not going to stay in the kitchen to watch on a tiny screen. Playing music or streamed radio? Maybe. But piece of crap computer can handle that. This thing is a useless gadget. And if you really want to be able to browse / e-mail in the kitchen or other locations with small terminals, don't you think you'd at least want a centralized login to each machine in the house? Oops. And no, WinCE won't cut it for that. Lameness..
So about the modifications to the standard Debian installer and the graphical configuration bits.. are they released under GPL? Then somebody in the Debian project needs to integrate them into the standard distribution, even if optional. Debian is great for newbies if you can just get them over the first hurdle.
Just out of idle curiousity, where do I go to download debian? I mean, should I download the potato, 2.2r5, testing, unstable... which? I have downloaded 2.2r5 but was wondering which I should get... thanks.
You don't 'download' testing/unstable. You add it to your package sources.
/etc/apt/sources.list
Add something like:
deb ftp://ftp.us.debian.org/debian testing main contrib non-free
deb ftp://non-us.debian.org/debian-non-US testing/non-US main contrib non-free
The second one is for crypto software. Duplicate but replace testing with unstable if you want to live on the edge. (-:
Imagine getting paid by the SEC to make up stuff about a non-existent company. Where do I sign up?
Probably your local unemployment office. But keep in mind the job description requires at least 2 years prior experience working for a former dotCom. (-:
Yes! it's perfect! Muahahah! We'll create a whole bunch of fake companies, put up flashy web pages with trendy earthtone color-schemes, advertise things people need like.. uh.. value-added B2B online toilet paper warehouses. Then we'll launch a bunch of fake IPO's that'll go wild because we're on the Internet and everybody knows the future of business is on the Internet! Yes. And then we'll take the money and run! That'll teach the fools to not believe all they read AND it'll stimulate the economy at the very same time!! Muahahah!
It's a bad idea to get rid of the old DOS prompt. There are still things that a command line does better than a GUI. There is no GUI good enough yet to utterly replace the usefulness of the command line.
Sorry dude. DOS is 100% dead. It died many many years ago. But you're right about command line interfaces being more useful (although DOS had a crappy one). Use a real operating system (Unix) and you'll have all the command line interfaces you could dream of. And if you're still using the Win9x series, give it up man. 16-bit real-mode is history.
The rest of the legacy stuff is the reason why the PC platform is so much more useful than anything out there.
Uh.. No. The PC is more useful because it's an open architecture. If for some reason you still need legacy ports or floppies, use an appropriate USB adaptor. Linux supports most all of them.
So.. other than XFree86 4.2, I've been using everything 'new' included in MDK 8.2 for about 1-2 months now. (using Debian testing/unstable). On the other hand, I'd have to give the Mandrake people a thumbs up for the increasingly rich-featured installer. Allowing newbies to set up crypto filesystems with no effort is a great idea.
Finally, an alternative! I for one, am sick of the hacked up kludge that is modern Intel/IBM PC hardware. I swear, every time I upgrade my system to the latest and greatest x86 processor and chipset, I run into more and more bugs, flaws, unexplainable quirks and incompatibilities, etc. in the hardware. Think about it. PC hardware has so many relics from the past--all of which must be kept supposedly to support legacy apps and hardware. When will designers wake up and realize.. "hey! we don't need to run DOS anymore!" It's time to cut all ties with "legacy IBM PC" functionality. That means it's time to ditch: floppy controllers, all remnants the ISA bus, PS/2 ports, parallel and serial ports, BIOS functionality intended for real mode operating systems, etc.
Or.. if the price is right, maybe I'll just buy one of these PPC boards. (-:
Does Katz watch every single movie that comes out?! No wonder he doesn't have time to write good editorials. (-:
Rules of Movie Previews:
If it looks dumb, it's sickeningly stupid.
If it sounds great, it'll put you to sleep.
If it looks hilarious, it's definitely not.
If it looks exciting, it has no plot.
If you can't quite figure out the preview, it might be a good movie.
Why would people pay $5/month per machine to keep software updated when a well-designed free solution already exists? "What's that?" you say. (Oh no.. another Debian zealot..) Just consider this a moment:
1.) As a sys admin, if you're smart and value your time, you'll be using mostly diskless workstations to begin with. No, I don't mean what Sun calls diskless workstations. I mean ordinary full-featured PC's with no hard disks. Or if they do have hard disks, it's only for a cache of the networked filesystems. So there, right off the start, you don't need this RHN silliness because by very nature, diskless workstations don't need updated, only the server does.
2.) So lets say you're a home user with only 3 machines, making it silly to set up diskless workstations. There's still a better way. First, you use Debian. One machine is your 'test box.' Use both the Debian testing and unstable sources. (unstable is rather misleading if you think that implies the kinda useless mess that is Mandrake's 'cooker' tree. Debian unstable is quite stable enough for all but heavy production systems) So anyways, once a week, on your test box, you do:
apt-get update; apt-get upgrade
This will upgrade *everything* installed on the machine as packages. Most likely everything is fine, but if you're worried, use the test box for a couple days to make sure everything is stable. All other machines should have a weekly cron job that uses your test box as a package source and performs the same operation.. but 2-3 days after the test-box upgrade, just in case something broke.
It's that simple. The RH people need to wake up and realize that they need to sell an actual solution if you're going to make money. Repackaging free software is a solution to a problem that doesn't exist.
Frankly, if you are technically minded, it probably doesn't matter what you major in within the Science or Engineering departments. It's more about just getting a degree. CS is a bit easier than CmpE or EE due to less math and no hard-core engineering classes. Whatever. Education doesn't stop after school. If you put your mind to it, you can do anything you like. Just get through school and move on.
The way I see it, Wine is best used as a tool for engineering special-purpose Windows software into legacy status. I'm talking about software that businesses depend on that is written in-house or by smaller software companies. This software typically doesn't have all the secret API "features" that only big proprietary companies in bed with M$ use. For replacing Photoshop, Office, multimedia editors, etc., we still need native Open Source applications.
How's this for an idea.. an Open Source co-op. Lets say 10,000 people need a replacement for Photoshop. If they pool their resources at say, $100 per user (about 1/6 the price of a Photoshop license), that's a million bucks. The co-op then pays perhaps 10 programmers to work full time for 2 years turning Gimp into the most beautifully written image editor ever. Yes, some form of contract would be needed. Yes, the co-op would have to wait until the money is pooled before hiring. But it could work. Or here's another wacky idea. Invest that $1mil and pay a couple programmers to work full time using the interest alone. Money is powerful. We (the OSS community), really need to take advantage of it.
Giving the software away which was build in 1 year by 10 people is throwing away a lot of money, unless you didn't pay these developers. How are YOU planning to pay 10 developers who work 8-10 hours a day for a whole year? With promises that supportcontracts will pay them in 2 years? Aren't there a lot of bad examples about companies who thought that way and went chapter 11?
..Because you're not giving away software and selling support contracts. You're selling contracts to produce the required software in the first place, but then releasing any non-private code back to the community. There are dozens of ways to organize this but nobody has tried any of them to my knowledge. Many times, it's not a matter of writing a whole new program, but extending an existing Open Source package to meet somebody's needs so they don't have to waste money on a proprietary one. Yes, support contracts are also possible, but they're an accessory, not the focus of the business.
It is about money. In fact it's about innovating so you make more money yourself and put the proprietary guys out of business. Change happens, deal with it.
The biggest flaw I see in what you are saying is that one company / person cannot pay a developer to build them an application and then turn it loose for the world. The problem here is that a single sold copy would be too expensive. You must sell thousands of copies at a lower cost to make up for the time and over-head. So who do you charge?
I see your point, but I don't think it necessarily applies here. First off, because Open Source software is already as far along as it is, I don't believe there are many cases in which entire applications must be written from the ground up. It's more a matter of say.. contracting a programmer to add a dozen features to an existing Open Source package so that it meets the clients needs. In this case, it is quite probable that the cost of paying a couple programmers for a single job will be far less than spending money on commercial licenses of a completely different program. Plus, you eliminate the problem of software becoming obsolete / having to pay for new versions, etc. It's cheaper both in the short run and the long run. With a system of 'bounties' for features, even home users can "vote" by putting funds in escrow until the desired feature(s) are met and the programmer completing them is rewarded. I guess the example that sticks out clearest is office suites. It's not cheap to pay M$ for 100 licenses for every workstation PLUS the licenses for Windows itself. So you also kill two birds with one stone by eliminating a barrier to converting to Linux / other free software.
The second possibility is that of highly custom software that's going to have to be produced under contract to begin with. In this case, pieces that are private can remain within a company and pieces that extend the functionality of various supporting Open Source tools / libraries / etc., can be re-released to the public. I forget the statistic, but by far, most programming is actually in-house. In many cases, companies need roughly the same software, but with some relatively minor customizations for each client. In this case, the pooling of contracts can provide the income to complete development while still keeping any non-custom source code open to the public after the work is complete.
The third case is one in which an entire, complex software package must be developed from the ground up. Granted, this is the most difficult case. On the other hand, by dividing the software into highly modular features, some sort of bidding system can again be utilized to construct the desired software in order of importance. Or one could mix and match all different ways of paying for development.
Perhaps tax breaks, equipment and residence should be provider for open source developers.
Although I'm not big on the idea of government involvement, when you consider that otherwise our tax dollars will go to big corporations to produce needed software, it really wouldn't be such a bad idea. If we're paying taxes, we might as well benefit from how the money is spent. In a certain way, you could even view many types of software as part of national infrastructure. Hm...
Because Open Source development and the surrounding community is not compatible with traditional business models. And yet everyone refuses to open their eyes to this, innovate, and move forward. Thus far, there have been two approaches:
1.) Distributions - supposedly 'value-added' compilations of free software with various technical support options as the real selling point. (Well, tech support if you really want to use the quirky, obfuscated mess that is most commercial Linux distributions.) Now if these same companies had just helped to enhance the Debian project (with say.. a newbie-friendly installer / hardware detection / GUI control panel / etc.) and then supported that instead, maybe it would have worked. But instead, the commercial distro people waste enormous amounts of time and energy redo'ing work that other people have already done a good job on.
2.) Proprietary Software for Linux - these are the people who believe that Open Source doesn't work for all types of software, so instead of fully supporting the community, they've created proprietary products to "fill in the gaps" in the landscape of free software. Problem is, this isn't economically efficient for them OR the customers. And it certainly doesn't further the cause of Open Source software. People who use proprietary office suites or games are throwing money into a black hole. Sure, it may temporarily suit a need or desire, but that money doesn't go towards producing software that will live on, continually improving as an open code base.
So here's the part where innovation comes in. People are willing to pay for software that they need as long as it actually fits the bill. Because of this, there is no reason why these users cannot pay Open Source developers to write software they need, but which will also be free to the community. Why would you pay for something that will be free? Because if you don't, it will never exist.. or at least it won't exist by the time you need it. So would you, as a user of software, rather sink money into proprietary software whose code will never see the light of day? Or would you rather get better quality software for the same price and support Open Source at the same time? The key, of course, is how to *organize* this exchange between users and developers in some sort of contract form.
Unfortunately, I don't believe Eric Raymond ever really touched on this when he wrote The Magic Cauldron. Of the Open Source business models he listed, the closest match is perhaps found in section 9.3 "Give away recipes, Open a restaurant." But it does have a powerful point, one that needs to be re-thought and expanded. Software can be, in every sense of the word, a service. Programmers are paid for the labor of writing code that does what people need. That's it. Forget selling things. Forget delusions of 'value-added versions.' And forget distributions plastered with bright logos and 'commercial looking' shrinkwrap. It's a waste. The true value is the functionality of the code itself. Contract for it to be written to specification, then stick it up on an FTP server for the world to see. Or sell your services, promising that you will transform any current and future Open Source software into a turnkey solution. Because that's what customers want: a solution. It's that simple.
There are many ways to go about this and it's time to starting trying them. People, stop sitting on your asses complaining and go start making a difference.
Somebody needs to ship this brilliant Irish guy over to California to meet the "Wireless Free" wackos. Certainly his incredible new device wouldn't release any harmful radiation. I mean.. that might break the second law of thermodynamics or something! Hell, this amazing machine absorbes all the deadly cell-phone radiation within a 100 mile radius and simultaneously renders aspartame harmless! How? Sorry, I can't tell you. It's a secret. But honestly.. it does work! Hypochondriacs everywhere can attest to this. Just ask 'em!
..we all know what 'really' happened.. Negotiations surely fell apart as soon as Cox announced he would leave. I mean, what else of value does RH do but pay our hero to work full time? Put out a quality distro?
If anyone has been looking for a good example of why the GPL's anti-proprietary protections make it superior to the BSD license, here's a great one. (from a community perspective at least..) Imagine if Linux and associated GNU software used the BSD license instead. AOL could buy RedHat and then release a proprietary kernel, libc, etc. with DRM integrated throughout, backdoors as desired, phone-home capability to reap marketing data, forced advertisements, and other horrible evils. With GPL, the worst they can do is include a proprietary version of Mozilla and perhaps a DRM kernel module, which both can be easily removed. So if AOL ships out GPL'ed software, you can be rest assured that it is the real thing or at least come with full source to document any potentially undesirable changes. With BSD, we'd be screwed.
Saying that the GPL is less free than BSD is like saying the US is less free without slavery.
This kinda crap is never going to stop 'piracy.' It's only going to piss off the average consumer who paid for his cable line or satellite or movie disc and wants to do whatever he/she pleases with it. Think of it this way: at the same time all this copy control nonsense is being rolled out, so is HDTV. To take advantage of HDTV, you have new TVs that are using some sort of LCD or Plasma display. So while you may not have access to a plaintext digital stream, the analog signal being fed to these devices is matrix-addressed and quite pure (unlike that of CRTs). All someone has to do is connect the highest quality ADC's they can find directly to the DAC output from the integrated HDTV decoder and voila.. nearly perfect digital copy. (avoiding the crudeness of pointing a camera at a screen..) They might lose accuracy in the least significant bit, but that's not going to stop true pirates from selling fraudulent copies.
So when you get down to it, once again, this is not about copyright, this is about controlling consumers / reducing privacy to make more money.
If the government wants to see some progress made in nation-wide computer security, they ought to not waste money punishing big dumb companies, but instead fund the geeks over at the NSA to work on Open Source security-related projects, much as they did with Linux and ACL's. Otherwise, I fail to see the courts could be objective. Accidents happen. Would companies get a quota of security holes per year?
Slashdot Mirror
Ooohh.. you mean those silly programs that ran in DOS that you could download from all your local dial-in bulletin boards at 14400 baud? You know.. before Linux and the whole GPL thing caught on.. I heard rumors that some people are dumb enough to still use 'shareware' in that 'other' operating system.
Perhaps a better question is: Why are people wasting their time trying to make a few dollars off poorly written programs when instead they could contribute to Open Source projects and actually have some influence and get their name out? Worse yet are those Windows freeware people who think they'll be more '31337' if they keep their amazing h4x0r c0d3z as t0p s3cr37!!!
On a more positive note, that's a lot of potentially valuable programmers that could be converted from the 'dark side' (-:
"It's a machine, Scroter, it doesn't get pissed off. It doesn't get happy, it doesn't get sad, it doesn't laugh at your jokes... It just runs programs!!"
... and the only one that can be mass produced with unskilled labor. -- Wernher von Braun
Seriously, folks... Invasion by robots that we created?! Considering that these are the same crack pots proposing unsupportable theories of global warming just so they can get research grants with our tax dollars? AI is called AI for a reason. It's artificial. It's not human. The best it can do is make decisions based on a knowledge base and a set of criteria with varying order of importance. In other words, it's deterministic at any given time. Sure, you can use random number generation, but if RNG become a significant enough factor to really change decisions from deterministic ones, then most decisions will not truly meet the criteria. Machines can't have a soul.
And if you really believe that human behavior is dictated solely by our brains, then we are already 'robots' but just fully biological ones. If that is the case, then what does it matter whether or not machines take over the world? If we're just machines, life and survival has no meaning to begin with. This also kinda fits in with the argument that 'if consciousness is just an illusion, then how can we make that statement?'
Man is the best computer we can put aboard a spacecraft
Yes, having a computer in the kitchen is great for things like: keeping your grocery list, displaying recipes, setting multiple timers for cooking times, keeping track of nutritional contents, etc. But who on earth would watch a movie?! If your hands are that free, you're not going to stay in the kitchen to watch on a tiny screen. Playing music or streamed radio? Maybe. But piece of crap computer can handle that. This thing is a useless gadget. And if you really want to be able to browse / e-mail in the kitchen or other locations with small terminals, don't you think you'd at least want a centralized login to each machine in the house? Oops. And no, WinCE won't cut it for that. Lameness..
So about the modifications to the standard Debian installer and the graphical configuration bits.. are they released under GPL? Then somebody in the Debian project needs to integrate them into the standard distribution, even if optional. Debian is great for newbies if you can just get them over the first hurdle.
Just out of idle curiousity, where do I go to download debian? I mean, should I download the potato, 2.2r5, testing, unstable... which? I have downloaded 2.2r5 but was wondering which I should get... thanks.
You don't 'download' testing/unstable. You add it to your package sources.
/etc/apt/sources.list
Add something like:
deb ftp://ftp.us.debian.org/debian testing main contrib non-free
deb ftp://non-us.debian.org/debian-non-US testing/non-US main contrib non-free
The second one is for crypto software. Duplicate but replace testing with unstable if you want to live on the edge. (-:
..trying to stop Napster by uploading renamed MP3's of Homer Simpson repeatedly saying D'oh!
Imagine getting paid by the SEC to make up stuff about a non-existent company. Where do I sign up?
Probably your local unemployment office. But keep in mind the job description requires at least 2 years prior experience working for a former dotCom. (-:
Yes! it's perfect! Muahahah! We'll create a whole bunch of fake companies, put up flashy web pages with trendy earthtone color-schemes, advertise things people need like.. uh.. value-added B2B online toilet paper warehouses. Then we'll launch a bunch of fake IPO's that'll go wild because we're on the Internet and everybody knows the future of business is on the Internet! Yes. And then we'll take the money and run! That'll teach the fools to not believe all they read AND it'll stimulate the economy at the very same time!! Muahahah!
Oh wait..
Ok, so where's the source so that this won't be just another proprietary standard that nobody uses?
It's a bad idea to get rid of the old DOS prompt. There are still things that a command line does better than a GUI. There is no GUI good enough yet to utterly replace the usefulness of the command line.
Sorry dude. DOS is 100% dead. It died many many years ago. But you're right about command line interfaces being more useful (although DOS had a crappy one). Use a real operating system (Unix) and you'll have all the command line interfaces you could dream of. And if you're still using the Win9x series, give it up man. 16-bit real-mode is history.
The rest of the legacy stuff is the reason why the PC platform is so much more useful than anything out there.
Uh.. No. The PC is more useful because it's an open architecture. If for some reason you still need legacy ports or floppies, use an appropriate USB adaptor. Linux supports most all of them.
So.. other than XFree86 4.2, I've been using everything 'new' included in MDK 8.2 for about 1-2 months now. (using Debian testing/unstable). On the other hand, I'd have to give the Mandrake people a thumbs up for the increasingly rich-featured installer. Allowing newbies to set up crypto filesystems with no effort is a great idea.
Finally, an alternative! I for one, am sick of the hacked up kludge that is modern Intel/IBM PC hardware. I swear, every time I upgrade my system to the latest and greatest x86 processor and chipset, I run into more and more bugs, flaws, unexplainable quirks and incompatibilities, etc. in the hardware. Think about it. PC hardware has so many relics from the past--all of which must be kept supposedly to support legacy apps and hardware. When will designers wake up and realize.. "hey! we don't need to run DOS anymore!" It's time to cut all ties with "legacy IBM PC" functionality. That means it's time to ditch: floppy controllers, all remnants the ISA bus, PS/2 ports, parallel and serial ports, BIOS functionality intended for real mode operating systems, etc.
Or.. if the price is right, maybe I'll just buy one of these PPC boards. (-:
Does Katz watch every single movie that comes out?! No wonder he doesn't have time to write good editorials. (-:
Rules of Movie Previews:
If it looks dumb, it's sickeningly stupid.
If it sounds great, it'll put you to sleep.
If it looks hilarious, it's definitely not.
If it looks exciting, it has no plot.
If you can't quite figure out the preview, it might be a good movie.
Why would people pay $5/month per machine to keep software updated when a well-designed free solution already exists? "What's that?" you say. (Oh no.. another Debian zealot..) Just consider this a moment:
1.) As a sys admin, if you're smart and value your time, you'll be using mostly diskless workstations to begin with. No, I don't mean what Sun calls diskless workstations. I mean ordinary full-featured PC's with no hard disks. Or if they do have hard disks, it's only for a cache of the networked filesystems. So there, right off the start, you don't need this RHN silliness because by very nature, diskless workstations don't need updated, only the server does.
2.) So lets say you're a home user with only 3 machines, making it silly to set up diskless workstations. There's still a better way. First, you use Debian. One machine is your 'test box.' Use both the Debian testing and unstable sources. (unstable is rather misleading if you think that implies the kinda useless mess that is Mandrake's 'cooker' tree. Debian unstable is quite stable enough for all but heavy production systems) So anyways, once a week, on your test box, you do:
apt-get update; apt-get upgrade
This will upgrade *everything* installed on the machine as packages. Most likely everything is fine, but if you're worried, use the test box for a couple days to make sure everything is stable. All other machines should have a weekly cron job that uses your test box as a package source and performs the same operation.. but 2-3 days after the test-box upgrade, just in case something broke.
It's that simple. The RH people need to wake up and realize that they need to sell an actual solution if you're going to make money. Repackaging free software is a solution to a problem that doesn't exist.
Frankly, if you are technically minded, it probably doesn't matter what you major in within the Science or Engineering departments. It's more about just getting a degree. CS is a bit easier than CmpE or EE due to less math and no hard-core engineering classes. Whatever. Education doesn't stop after school. If you put your mind to it, you can do anything you like. Just get through school and move on.
The way I see it, Wine is best used as a tool for engineering special-purpose Windows software into legacy status. I'm talking about software that businesses depend on that is written in-house or by smaller software companies. This software typically doesn't have all the secret API "features" that only big proprietary companies in bed with M$ use. For replacing Photoshop, Office, multimedia editors, etc., we still need native Open Source applications.
How's this for an idea.. an Open Source co-op. Lets say 10,000 people need a replacement for Photoshop. If they pool their resources at say, $100 per user (about 1/6 the price of a Photoshop license), that's a million bucks. The co-op then pays perhaps 10 programmers to work full time for 2 years turning Gimp into the most beautifully written image editor ever. Yes, some form of contract would be needed. Yes, the co-op would have to wait until the money is pooled before hiring. But it could work. Or here's another wacky idea. Invest that $1mil and pay a couple programmers to work full time using the interest alone. Money is powerful. We (the OSS community), really need to take advantage of it.
I have two words to summarize: bottom feeders
Contribute to the community or get out of the game.
Giving the software away which was build in 1 year by 10 people is throwing away a lot of money, unless you didn't pay these developers. How are YOU planning to pay 10 developers who work 8-10 hours a day for a whole year? With promises that supportcontracts will pay them in 2 years? Aren't there a lot of bad examples about companies who thought that way and went chapter 11?
..Because you're not giving away software and selling support contracts. You're selling contracts to produce the required software in the first place, but then releasing any non-private code back to the community. There are dozens of ways to organize this but nobody has tried any of them to my knowledge. Many times, it's not a matter of writing a whole new program, but extending an existing Open Source package to meet somebody's needs so they don't have to waste money on a proprietary one. Yes, support contracts are also possible, but they're an accessory, not the focus of the business.
It is about money. In fact it's about innovating so you make more money yourself and put the proprietary guys out of business. Change happens, deal with it.
The biggest flaw I see in what you are saying is that one company / person cannot pay a developer to build them an application and then turn it loose for the world. The problem here is that a single sold copy would be too expensive. You must sell thousands of copies at a lower cost to make up for the time and over-head. So who do you charge?
I see your point, but I don't think it necessarily applies here. First off, because Open Source software is already as far along as it is, I don't believe there are many cases in which entire applications must be written from the ground up. It's more a matter of say.. contracting a programmer to add a dozen features to an existing Open Source package so that it meets the clients needs. In this case, it is quite probable that the cost of paying a couple programmers for a single job will be far less than spending money on commercial licenses of a completely different program. Plus, you eliminate the problem of software becoming obsolete / having to pay for new versions, etc. It's cheaper both in the short run and the long run. With a system of 'bounties' for features, even home users can "vote" by putting funds in escrow until the desired feature(s) are met and the programmer completing them is rewarded. I guess the example that sticks out clearest is office suites. It's not cheap to pay M$ for 100 licenses for every workstation PLUS the licenses for Windows itself. So you also kill two birds with one stone by eliminating a barrier to converting to Linux / other free software.
The second possibility is that of highly custom software that's going to have to be produced under contract to begin with. In this case, pieces that are private can remain within a company and pieces that extend the functionality of various supporting Open Source tools / libraries / etc., can be re-released to the public. I forget the statistic, but by far, most programming is actually in-house. In many cases, companies need roughly the same software, but with some relatively minor customizations for each client. In this case, the pooling of contracts can provide the income to complete development while still keeping any non-custom source code open to the public after the work is complete.
The third case is one in which an entire, complex software package must be developed from the ground up. Granted, this is the most difficult case. On the other hand, by dividing the software into highly modular features, some sort of bidding system can again be utilized to construct the desired software in order of importance. Or one could mix and match all different ways of paying for development.
Perhaps tax breaks, equipment and residence should be provider for open source developers.
Although I'm not big on the idea of government involvement, when you consider that otherwise our tax dollars will go to big corporations to produce needed software, it really wouldn't be such a bad idea. If we're paying taxes, we might as well benefit from how the money is spent. In a certain way, you could even view many types of software as part of national infrastructure. Hm...
Because Open Source development and the surrounding community is not compatible with traditional business models. And yet everyone refuses to open their eyes to this, innovate, and move forward. Thus far, there have been two approaches:
1.) Distributions - supposedly 'value-added' compilations of free software with various technical support options as the real selling point. (Well, tech support if you really want to use the quirky, obfuscated mess that is most commercial Linux distributions.) Now if these same companies had just helped to enhance the Debian project (with say.. a newbie-friendly installer / hardware detection / GUI control panel / etc.) and then supported that instead, maybe it would have worked. But instead, the commercial distro people waste enormous amounts of time and energy redo'ing work that other people have already done a good job on.
2.) Proprietary Software for Linux - these are the people who believe that Open Source doesn't work for all types of software, so instead of fully supporting the community, they've created proprietary products to "fill in the gaps" in the landscape of free software. Problem is, this isn't economically efficient for them OR the customers. And it certainly doesn't further the cause of Open Source software. People who use proprietary office suites or games are throwing money into a black hole. Sure, it may temporarily suit a need or desire, but that money doesn't go towards producing software that will live on, continually improving as an open code base.
So here's the part where innovation comes in. People are willing to pay for software that they need as long as it actually fits the bill. Because of this, there is no reason why these users cannot pay Open Source developers to write software they need, but which will also be free to the community. Why would you pay for something that will be free? Because if you don't, it will never exist.. or at least it won't exist by the time you need it. So would you, as a user of software, rather sink money into proprietary software whose code will never see the light of day? Or would you rather get better quality software for the same price and support Open Source at the same time? The key, of course, is how to *organize* this exchange between users and developers in some sort of contract form.
Unfortunately, I don't believe Eric Raymond ever really touched on this when he wrote The Magic Cauldron. Of the Open Source business models he listed, the closest match is perhaps found in section 9.3 "Give away recipes, Open a restaurant." But it does have a powerful point, one that needs to be re-thought and expanded. Software can be, in every sense of the word, a service. Programmers are paid for the labor of writing code that does what people need. That's it. Forget selling things. Forget delusions of 'value-added versions.' And forget distributions plastered with bright logos and 'commercial looking' shrinkwrap. It's a waste. The true value is the functionality of the code itself. Contract for it to be written to specification, then stick it up on an FTP server for the world to see. Or sell your services, promising that you will transform any current and future Open Source software into a turnkey solution. Because that's what customers want: a solution. It's that simple.
There are many ways to go about this and it's time to starting trying them. People, stop sitting on your asses complaining and go start making a difference.
Somebody needs to ship this brilliant Irish guy over to California to meet the "Wireless Free" wackos. Certainly his incredible new device wouldn't release any harmful radiation. I mean.. that might break the second law of thermodynamics or something! Hell, this amazing machine absorbes all the deadly cell-phone radiation within a 100 mile radius and simultaneously renders aspartame harmless! How? Sorry, I can't tell you. It's a secret. But honestly.. it does work! Hypochondriacs everywhere can attest to this. Just ask 'em!
..we all know what 'really' happened.. Negotiations surely fell apart as soon as Cox announced he would leave. I mean, what else of value does RH do but pay our hero to work full time? Put out a quality distro?
Use Debian, for it sucks not
If anyone has been looking for a good example of why the GPL's anti-proprietary protections make it superior to the BSD license, here's a great one. (from a community perspective at least..) Imagine if Linux and associated GNU software used the BSD license instead. AOL could buy RedHat and then release a proprietary kernel, libc, etc. with DRM integrated throughout, backdoors as desired, phone-home capability to reap marketing data, forced advertisements, and other horrible evils. With GPL, the worst they can do is include a proprietary version of Mozilla and perhaps a DRM kernel module, which both can be easily removed. So if AOL ships out GPL'ed software, you can be rest assured that it is the real thing or at least come with full source to document any potentially undesirable changes. With BSD, we'd be screwed.
Saying that the GPL is less free than BSD is like saying the US is less free without slavery.
This kinda crap is never going to stop 'piracy.' It's only going to piss off the average consumer who paid for his cable line or satellite or movie disc and wants to do whatever he/she pleases with it. Think of it this way: at the same time all this copy control nonsense is being rolled out, so is HDTV. To take advantage of HDTV, you have new TVs that are using some sort of LCD or Plasma display. So while you may not have access to a plaintext digital stream, the analog signal being fed to these devices is matrix-addressed and quite pure (unlike that of CRTs). All someone has to do is connect the highest quality ADC's they can find directly to the DAC output from the integrated HDTV decoder and voila.. nearly perfect digital copy. (avoiding the crudeness of pointing a camera at a screen..) They might lose accuracy in the least significant bit, but that's not going to stop true pirates from selling fraudulent copies.
So when you get down to it, once again, this is not about copyright, this is about controlling consumers / reducing privacy to make more money.
If the government wants to see some progress made in nation-wide computer security, they ought to not waste money punishing big dumb companies, but instead fund the geeks over at the NSA to work on Open Source security-related projects, much as they did with Linux and ACL's. Otherwise, I fail to see the courts could be objective. Accidents happen. Would companies get a quota of security holes per year?