SCO is a founding member, and the whole thing was really their (Ransom Love's) idea to start with. I hate SCO, but I love SuSE, who's the only one of the founding members who has any real market presence, and the UnitedLinux standard seems to be mostly based on SuSE.
There's more to UnitedLinux then a name and prerelease code. Those are pretty meaningless by themselves, as you point out.
What the software members get is a guarantee that software they develope will work on the other partners distros. They also (hopefully) get more support from third party vendors like Oracle, who can now target UnitedLinux rather than having to target each distro individually. That adds value to the ditros put out by the software members. It also speeds up developement, since all the developers at the various distros are working together.
What the hardware members (HP, Intel, and AMD) get is the ability to influence the direction of the standard and make sure their stuff is supported. That adds value to the hardware they're selling, as they know there is software that will take advantage of it. It makes things easier on them, too, since they can work with UnitedLinux as a whole, not have to go to each distro seperately.
IBM, of course, is in both groups. While they don't put out a distro of their own, they do provide third party apps, do a fair amount of Linux developement work, and they've supported all the UnitedLinux distros on their hardware since long before there was a UnitedLinux.
If people want to influence the way business makes decisions, become the person making the descisions.
I think that's the key to affecting any kind of change. You can't change the system from the outside. The trick, of course, is getting inside with your morals intact...
Everytime that phrase comes to your mind I want you to take a deep breath and think whether you would say that to your non-technical mother/father/granny/whatever.
I do say that to my non-technical wife, to which she responds "Fuck you, I'm not doing that!". That prompts me to write a script which I link to an icon on her desktop with a simple name like "play_dvd" (which would be a simple one-liner if it weren't for one stupid Powerpuff Girls DVD where Title 1 is one of the extras with a character voice over).
Granted, this approach won't work for households that don't include a Unix geek, but it will have to suffice until I can get mplayer to compile with a GUI.
You're right, he didn't switch because of the licenses, because if he understood the licenses at all he would have stayed with Linux. He switched because he for some reason thinks that BSD can't become commercialized, which, as I have already pointed out, is completely backwards because of the licenses involved.
And before you respond with the typical 'it freezes and crashes wah wah', stop and think the cause of the crashes/freezes could very well be because you've been trashing the filesystem when you cycle power on it like that.
And my other options are... I'm waiting...
Any way you look at it, it's still a Windows problem:
a) Windows freezes/locks up much more often than Linux. Windows is approaching Linux stability with XP, but they haven't reached it yet.
b) On the rare occasions that Linux locks up, I can just ssh to it from another machine and either kill the process or reboot it correctly. Can you show me how to do that in Windows? If you can't, then I need a hard reset button.
c) Linux has filesystems that don't fragment. Windows has marketing droids that say their filesystem doesn't fragment, even though it does, quite badly (and yes, I do mean NTFS).
I have Linux machines that only get shut down when the power goes out. Guess what? The filesystems aren't trashed. But hey, it couldn't be a Windows problem, it must be something that happens to all computers when they hard-reset...
This is a genuine problem in buying a laptop (as I understand it) -- not only do they have to pick a distribution (Debian, RH, etc) but also the role the computer will be fulfilling.
What a stupid problem.
The solution is pretty stupid as well. Build a basic install for each distro for your master disk, which then gets cloned and installed just like all the other customizations you can have done when you order a Dell (yeah, cloned, you don't really think they pay people to sit around and install software, do you?). Then you ship the distro CDs with the machine and the user can add or remove what they want (I don't know about other distros, but it's pretty damned easy to add or remove packages in SuSE, and I very much doubt they're that far ahead of the pack).
Linux is Linux is Linux. The various distros have only minor differences, and those differences should only be a problem for the most brain-dead of Process Engineers. The only real problem they could expect is with drivers, in which case that's in Dell's court anyway.
IBM seems to have a good approach, though. They ship Red Hat by default, and if you want something else you have to pay extra. That seems like a perfectly reasonable approach to me.
That has to be the stupidest "switch to BSD" arguement I've ever heard.
Big business come in and take over just like they do the Internet and the small hobby person lose all rights.
RTFLicense. It is impossible for Big Business to come in and take over, removing the rights from the little guy, under the GPL. That's why so many people hate it. Under the BSD License, however, that's it's not only possible but expected that Big Business will gobble up the code and lock it away from the little guy. That's the entire basis for all the "BSD is more Free than GPL" arguements.
If that's your reason for switching to BSD, you're an idiot.
That's exactly what I was thinking. The only real news here is that the guy worked out a systematic approach to doing it, but the basic idea should be obvious to anyone with even the slightest clue of how a master key system works.
That many Security Managers don't know about it is a little disconcerting, but not really suprising given the lack of dedication so many people have for their jobs.
In most cases the security of a lock comes from the fact that most people trying to get past them are too stupid or impatient to approach the task intelligently. The genius criminal safe-cracker is a media myth.
There are a lot of uses for the TCPA chip, and DRM is just one of them. That doesn't make the TCPA chip a DRM device any more than the fact that you can use the internet to pirate music makes the internet a music piracy device.
Also, what part of "it runs Linux" did you not understand? DRM won't work unless it is enabled in the OS, and since Linux is Open Source you can remove or modify the DRM parts to your hearts conent (if anyone were dumb enough to include them in the first place). Alternatively, you could simply download one of the anti-DRM patches that would be available about a month before you could actually buy the thing.
And hey, that's great that you can stream music to your Zaurus, but what's stopping me from sniffing your passwords and 0wning your server? Wouldn't it be nice if you could encrypt those streams and still have enough CPU left over to play your mpegs?
I'd quibble about a few points - MS's Office franchise is (financially) secure? Linux suffers from "real security issues"? Nonsense.
The way I read it, FT is saying that those are arguements MS could make, but will have a battle making them stick. "It can argue" v. "It can be argued".
I agree with you about MS Office, though. At the moment it might look secure at the moment, but then an undermined castle wall looks secure right up to the point where the besieger burn out the supports, and those Star/OpenOffice guys are working fast.
Sorry, Konrad Zuse, a German, built the first working computer, the Z3. It was operational in 1941, the same year Turing proposed Colossus. By the time Eckert and Mauchly got around to doing anything of note, Zuse had already moved on to designing a programming language, writing a chess playing algorithm, and finished building the Z4... and Colossus was 3 years old.
The article states that the server market is $200bn and that Linux's share is only $8bn which looks like a very small percentage until you read the Linux machines "cut the cost of hardware purchases eightfold, says Geoff Penney"
I'm not sure how to feel about that.
On the one hand, this is a weapon for MS. "Linux is only $8G in a $200G industry. That's only 4%! Even Apple has a bigger marketshare than that! You don't want to be a loner, you want to be able to interoperate with everyone else...", they might say.
On the other hand, what is a $8G marketshare at Linux prices would be a $64G marketshare at MS prices, which makes it more like a 25% marketshare.
I'm not entirely sure that my math or logic are entirely correct, but one thing is clear: money is a really stupid way to measure usage.
Yeah. So much for Stallmans vision of free software being wanted because you could control it and share it with friends. That sort of stuff is important only to a small fraction of the worlds people clearly.
As the parent said, it's called Finacial Times for a reason. The people who read it basically care about one thing: money. If you want to communicate with them you need to speak their language: money.
There's a very good reason for not mentioning RMS-style Freedom to this audience; it gives them the willies. It's very difficult for a money-minded person to see how those ideals benefit them since all they are able to see is monetary worth.
It's also rather depressing how much the involvement of IBM means to people. IBM has done a hell of a lot of good work, but the "we didn't pay any attention 'till IBM did" line indicates supreme daftness to me - Linux hasn't changed that much. I guess it's just a case of sheeple following whoever is biggest.
When a very small company I was working at a while back needed to move all the important business files off of the bosses desktop and onto a dedicated fileserver I suggested adding a big harddrive to an old unused machine we had and throwing Linux on it. After explaining what Linux is and why we don't have to pay anyone for it if we don't want to, the question posed was "Who else is using it?" It seemed like a very strange question to me, but eventually I understood what it was about. Managers really don't like to take risks. What they were asking for is examples of where Linux has been successfully deployed. Data is the backbone of any company, and you don't trust your data to a platform that doesn't have an acceptable track record, and to have that you need some big names (like IBM) saying publicly "Yes, we use that, it works".
Unfortunately, this was when IBM was first talking about supporting Linux, so examples from companies anyone had heard of were scarce, to say the least. I managed to talk them into it anyway, and the winning arguement was money. Hardware and software costs to do it the way I suggested were $125 (new HDD and NIC), whereas software costs alone for an MS solution was around $1000, or $1500 for Novell, and both of those would have required more hardware expenses to create a viable platform, and would have had additional expenses for licenses ($50 and $70 respectively per seat, IIRC) whenever a new employee were hired (the company was growing roughly 30% per year at the time).
It so happened that the CEO had recently asked the CFO to marry him, so once these numbers were pointed out to her, the issue was decided in favor of Linux;-)
Also regarding that quote, the access control mechanisms available in Windows NT et al *are* far more comprehensive than those available in most Unixes (including ones like Solaris that have some ACL support.)
The difference being that Unix permissions actually act as advertised, and are fairly simple to comprehend. And no, MS access control is NOT more comprehensive, they just have multiple checkboxes for the same permissions. Unix has Read, Write, and Execute. Windows has Read, Write and Modify (what exactly is the difference there), and Read & Execute. Notice that these are listed in terms of equivalent function. Windows also has Full Control, which provides no real extra functionality, and seperate Allow and Deny checkboxes, which is pointless and stupid (If you actually care about security you deny all priveleges not specifically granted, you know, like Unix does). Then of course it has an access control list, which does exactly the same thing as Unix ACLs.
Where exactly is this extra functionality you speak of?
I've been thinking about bounties for OSS developement, and I think that's going to be the way of the future. I've got this idea for a website, also. Basically it'd be just an organized, user created (moderated, it would have to be) list of functionality people want added with an easy way to donate to the cause. The business model is that the bounty goes into an interest-bearing no-risk account of some sort, and the website lives off that interest.
Anyway, I as a user might want CMYK for GIMP (bad example, GIMP 2 will have CMYK, but we'll run with it anyway), and maybe I see that you've already posted a request for it with a bounty of $20. At that point I might decide "hey, GIMP CMYK is worth $20 to me" and add my $20 to the bounty. If there are 50 other people that think GIMP CMYK is worth $20, then the bounty is over $1000. I know a few people in the printing and advertising industry who wouldn't mind a free graphics app that does CMYK, and it might just be worth the cost of a couple of Photoshop licenses to them. In easy numbers lets say 2 copies of Photoshop are $1000, and we get 4 companies to do that, now the bounty is up to $5000, and lets just say for the sake of arguement that there's some developer out there who, at $5000, considers the project to be worth his time.
The beauty of the system I envision is that, unlike the current situation where we the occasional private individual offering a large bounty ($20k for X functionality in Y package), the bounty is held in escrow (not entirely sure that's the correct term, but it's close enough), so it will be there until someone completes the job. This creates another interesting effect, in that even for rare itches the bounty can become sizable with time. If you and I were the only people in the world who wanted GIMP to have CMYK support, but we each donated $20 a month to the bounty, it would take a long time for the bounty to get up to $5000, but it would get there.
I see this plan benefiting users in that they can help direct the developement of a project that is important to them with a small outlay of capital. The benefit to developers is fairly obvious, in that they can get paid for something they might be doing anyway. It also removes the "pledge drive" feel that many requests for donations have, since it's the users offering money rather than the developers asking for it, hopefully without creating a code ransom situation which seems to be the other way OSS developers are able to make money.
Anyway, there's my idea. Feel free to steal it, shoot it full of holes, or give me the kick in the pants I need to actually get it started, as you see fit.
Two sys admins, a couple of DBAs and an IT manager, working for a large manufacturing company. You're telling me they have the skills, the time and the inclination to knock up a quick competitor to Oracle financials over the weekend?
Your examples completely ignore the fundamental strength of open source: one company isn't doing all the work.
You're correct in saying that the company you describe wouldn't be able to recreate Oracle on their own (never mind the timeframe), but they don't need to. There are plenty of other organizations that need that functionality, and there are plenty of OSS developers with detailed knowledge of OSS database apps who would be willing to do the work.
As a thought experiment, which will hopefully illustrate my point; what if California offered just 1% of the $95M it was prepared to spend on Oracle software as bounties and/or contracts to create the functionality they needed using MySQL or PostgreSQL? How long do you think it would have taken for that functionality and scalability to be achieved?
Now lets expand that; where would MySQL and PostreSQL be if all the companies buying Oracle and DB2 were instead pouring 1% of that money into the developement communities of those two packages?
Yeah, it's not going to happen over the weekend, but with that kind of incentive I doubt it would take very long.
So the people who managed the deployment were free?
Their called TAs, and at the district school level they often are free.
How about the user training?
Again, it's a school. Training is their business, and I very much doubt anyone at the high school level has any use for the "advanced features" of MS Office. I've never had a use for them, and I've been in the Real World for some time now.
As for training the teachers, you must be joking. I have a few teachers in my family, and the level of computer training they recieve is readily apparent: none at all.
How about the lost productivity time as end users got used to the new app?
I haven't converted many people away from MS Office yet, but in general the only "lost productivity time" I've seen has been the time it takes to install.
How about the conversion problems on the few especially complex documents star office struggles with?
Where exactly are all these "especially complex documents" I keep hearing about? Are these the same ones that MS Office seems to choke on?
It's been said before, and here it is again... free software is only free if your time is.
And yet OpenOffice still manages to be much cheaper than MS Office.
well I'm not saying anything. But, maybe this guy is a little biased
Biased or not he knows jack shit about Linux or the Linux community, and is therefore completely unqualified to make any comparisons.
Anyway, here's my favorite quote from the link:
Any developer documentation as comprehensive, accessible and easy to use as the MSDN Library?
Last semester I took classes in VB.NET and C. In VB I frequently spent hours following little crosslink circles without gaining any useful information. I did all my C on Linux, though, and found that everything I needed to know was a quick 'man function' away.
More than one researcher has argued that the world is flat, but that's not true either. These critics clearly have no conception of how science actually works.
I find Flat Earthers quite amusing. A while back a friend of mine decided to join the Flat Earth Society. I was able to disprove every one of their arguements based on my (at the time) one semester of Mechanics, without even having to refer to my book, in an hour or so.
It's truely amazing, and more than a little frightening, what people will accept as "proof".
Prime factorization is NOT a provably hard problem. Mathematicians have yet to show that it is NP-complete. For all we know, it could be quite easy and we just haven't found the easy way to do it.
You're right. In fact, factoring primes is provably easy: One times the number. Oh, except One isn't a factor, by definition. Oops!
It is impossible to factor a prime number. I have seen prime numbers expressed in terms of factors (for instance, 7 could be expressed as 2^3-1) but you cannot factor a prime number.
Slashdot Mirror
Services for Unix apparently includes a lot of OSS stuff, like gcc for example.
SCO is a founding member, and the whole thing was really their (Ransom Love's) idea to start with. I hate SCO, but I love SuSE, who's the only one of the founding members who has any real market presence, and the UnitedLinux standard seems to be mostly based on SuSE.
There's more to UnitedLinux then a name and prerelease code. Those are pretty meaningless by themselves, as you point out.
What the software members get is a guarantee that software they develope will work on the other partners distros. They also (hopefully) get more support from third party vendors like Oracle, who can now target UnitedLinux rather than having to target each distro individually. That adds value to the ditros put out by the software members. It also speeds up developement, since all the developers at the various distros are working together.
What the hardware members (HP, Intel, and AMD) get is the ability to influence the direction of the standard and make sure their stuff is supported. That adds value to the hardware they're selling, as they know there is software that will take advantage of it. It makes things easier on them, too, since they can work with UnitedLinux as a whole, not have to go to each distro seperately.
IBM, of course, is in both groups. While they don't put out a distro of their own, they do provide third party apps, do a fair amount of Linux developement work, and they've supported all the UnitedLinux distros on their hardware since long before there was a UnitedLinux.
If people want to influence the way business makes decisions, become the person making the descisions.
I think that's the key to affecting any kind of change. You can't change the system from the outside. The trick, of course, is getting inside with your morals intact...
The current state of video on Linux sucks ass. Especially on RH7.2
That would be a perfectly reasonable statement if RH 7.2 were the current state of Linux.
But it isn't!
You might as well bitch about the current state of NTFS support on WindowsME.
Everytime that phrase comes to your mind I want you to take a deep breath and think whether you would say that to your non-technical mother/father/granny/whatever.
I do say that to my non-technical wife, to which she responds "Fuck you, I'm not doing that!". That prompts me to write a script which I link to an icon on her desktop with a simple name like "play_dvd" (which would be a simple one-liner if it weren't for one stupid Powerpuff Girls DVD where Title 1 is one of the extras with a character voice over).
Granted, this approach won't work for households that don't include a Unix geek, but it will have to suffice until I can get mplayer to compile with a GUI.
You're right, he didn't switch because of the licenses, because if he understood the licenses at all he would have stayed with Linux. He switched because he for some reason thinks that BSD can't become commercialized, which, as I have already pointed out, is completely backwards because of the licenses involved.
And before you respond with the typical 'it freezes and crashes wah wah', stop and think the cause of the crashes/freezes could very well be because you've been trashing the filesystem when you cycle power on it like that.
And my other options are... I'm waiting...
Any way you look at it, it's still a Windows problem:
a) Windows freezes/locks up much more often than Linux. Windows is approaching Linux stability with XP, but they haven't reached it yet.
b) On the rare occasions that Linux locks up, I can just ssh to it from another machine and either kill the process or reboot it correctly. Can you show me how to do that in Windows? If you can't, then I need a hard reset button.
c) Linux has filesystems that don't fragment. Windows has marketing droids that say their filesystem doesn't fragment, even though it does, quite badly (and yes, I do mean NTFS).
I have Linux machines that only get shut down when the power goes out. Guess what? The filesystems aren't trashed. But hey, it couldn't be a Windows problem, it must be something that happens to all computers when they hard-reset...
This is a genuine problem in buying a laptop (as I understand it) -- not only do they have to pick a distribution (Debian, RH, etc) but also the role the computer will be fulfilling.
What a stupid problem.
The solution is pretty stupid as well. Build a basic install for each distro for your master disk, which then gets cloned and installed just like all the other customizations you can have done when you order a Dell (yeah, cloned, you don't really think they pay people to sit around and install software, do you?). Then you ship the distro CDs with the machine and the user can add or remove what they want (I don't know about other distros, but it's pretty damned easy to add or remove packages in SuSE, and I very much doubt they're that far ahead of the pack).
Linux is Linux is Linux. The various distros have only minor differences, and those differences should only be a problem for the most brain-dead of Process Engineers. The only real problem they could expect is with drivers, in which case that's in Dell's court anyway.
IBM seems to have a good approach, though. They ship Red Hat by default, and if you want something else you have to pay extra. That seems like a perfectly reasonable approach to me.
That has to be the stupidest "switch to BSD" arguement I've ever heard.
Big business come in and take over just like they do the Internet and the small hobby person lose all rights.
RTFLicense. It is impossible for Big Business to come in and take over, removing the rights from the little guy, under the GPL. That's why so many people hate it. Under the BSD License, however, that's it's not only possible but expected that Big Business will gobble up the code and lock it away from the little guy. That's the entire basis for all the "BSD is more Free than GPL" arguements.
If that's your reason for switching to BSD, you're an idiot.
That's exactly what I was thinking. The only real news here is that the guy worked out a systematic approach to doing it, but the basic idea should be obvious to anyone with even the slightest clue of how a master key system works.
That many Security Managers don't know about it is a little disconcerting, but not really suprising given the lack of dedication so many people have for their jobs.
In most cases the security of a lock comes from the fact that most people trying to get past them are too stupid or impatient to approach the task intelligently. The genius criminal safe-cracker is a media myth.
There are a lot of uses for the TCPA chip, and DRM is just one of them. That doesn't make the TCPA chip a DRM device any more than the fact that you can use the internet to pirate music makes the internet a music piracy device.
Also, what part of "it runs Linux" did you not understand? DRM won't work unless it is enabled in the OS, and since Linux is Open Source you can remove or modify the DRM parts to your hearts conent (if anyone were dumb enough to include them in the first place). Alternatively, you could simply download one of the anti-DRM patches that would be available about a month before you could actually buy the thing.
And hey, that's great that you can stream music to your Zaurus, but what's stopping me from sniffing your passwords and 0wning your server? Wouldn't it be nice if you could encrypt those streams and still have enough CPU left over to play your mpegs?
I'd quibble about a few points - MS's Office franchise is (financially) secure? Linux suffers from "real security issues"? Nonsense.
The way I read it, FT is saying that those are arguements MS could make, but will have a battle making them stick. "It can argue" v. "It can be argued".
I agree with you about MS Office, though. At the moment it might look secure at the moment, but then an undermined castle wall looks secure right up to the point where the besieger burn out the supports, and those Star/OpenOffice guys are working fast.
Sorry, Konrad Zuse, a German, built the first working computer, the Z3. It was operational in 1941, the same year Turing proposed Colossus. By the time Eckert and Mauchly got around to doing anything of note, Zuse had already moved on to designing a programming language, writing a chess playing algorithm, and finished building the Z4... and Colossus was 3 years old.
Computer Science: Alan Turing, UK.
The only genius computing has ever had, forced into topping himself because the UK goverment was homophobic.
John Von Neuman?
Charles Babbage?
Ada Byron?
Blaise Pascal?
Konrad Zuse?
It seems your knowledge of the history of computing is rather limited.
The article states that the server market is $200bn and that Linux's share is only $8bn which looks like a very small percentage until you read the Linux machines "cut the cost of hardware purchases eightfold, says Geoff Penney"
I'm not sure how to feel about that.
On the one hand, this is a weapon for MS. "Linux is only $8G in a $200G industry. That's only 4%! Even Apple has a bigger marketshare than that! You don't want to be a loner, you want to be able to interoperate with everyone else...", they might say.
On the other hand, what is a $8G marketshare at Linux prices would be a $64G marketshare at MS prices, which makes it more like a 25% marketshare.
I'm not entirely sure that my math or logic are entirely correct, but one thing is clear: money is a really stupid way to measure usage.
Yeah. So much for Stallmans vision of free software being wanted because you could control it and share it with friends. That sort of stuff is important only to a small fraction of the worlds people clearly.
;-)
As the parent said, it's called Finacial Times for a reason. The people who read it basically care about one thing: money. If you want to communicate with them you need to speak their language: money.
There's a very good reason for not mentioning RMS-style Freedom to this audience; it gives them the willies. It's very difficult for a money-minded person to see how those ideals benefit them since all they are able to see is monetary worth.
It's also rather depressing how much the involvement of IBM means to people. IBM has done a hell of a lot of good work, but the "we didn't pay any attention 'till IBM did" line indicates supreme daftness to me - Linux hasn't changed that much. I guess it's just a case of sheeple following whoever is biggest.
When a very small company I was working at a while back needed to move all the important business files off of the bosses desktop and onto a dedicated fileserver I suggested adding a big harddrive to an old unused machine we had and throwing Linux on it. After explaining what Linux is and why we don't have to pay anyone for it if we don't want to, the question posed was "Who else is using it?" It seemed like a very strange question to me, but eventually I understood what it was about. Managers really don't like to take risks. What they were asking for is examples of where Linux has been successfully deployed. Data is the backbone of any company, and you don't trust your data to a platform that doesn't have an acceptable track record, and to have that you need some big names (like IBM) saying publicly "Yes, we use that, it works".
Unfortunately, this was when IBM was first talking about supporting Linux, so examples from companies anyone had heard of were scarce, to say the least. I managed to talk them into it anyway, and the winning arguement was money. Hardware and software costs to do it the way I suggested were $125 (new HDD and NIC), whereas software costs alone for an MS solution was around $1000, or $1500 for Novell, and both of those would have required more hardware expenses to create a viable platform, and would have had additional expenses for licenses ($50 and $70 respectively per seat, IIRC) whenever a new employee were hired (the company was growing roughly 30% per year at the time).
It so happened that the CEO had recently asked the CFO to marry him, so once these numbers were pointed out to her, the issue was decided in favor of Linux
Also regarding that quote, the access control mechanisms available in Windows NT et al *are* far more comprehensive than those available in most Unixes (including ones like Solaris that have some ACL support.)
The difference being that Unix permissions actually act as advertised, and are fairly simple to comprehend. And no, MS access control is NOT more comprehensive, they just have multiple checkboxes for the same permissions. Unix has Read, Write, and Execute. Windows has Read, Write and Modify (what exactly is the difference there), and Read & Execute. Notice that these are listed in terms of equivalent function. Windows also has Full Control, which provides no real extra functionality, and seperate Allow and Deny checkboxes, which is pointless and stupid (If you actually care about security you deny all priveleges not specifically granted, you know, like Unix does). Then of course it has an access control list, which does exactly the same thing as Unix ACLs.
Where exactly is this extra functionality you speak of?
I've been thinking about bounties for OSS developement, and I think that's going to be the way of the future. I've got this idea for a website, also. Basically it'd be just an organized, user created (moderated, it would have to be) list of functionality people want added with an easy way to donate to the cause. The business model is that the bounty goes into an interest-bearing no-risk account of some sort, and the website lives off that interest.
Anyway, I as a user might want CMYK for GIMP (bad example, GIMP 2 will have CMYK, but we'll run with it anyway), and maybe I see that you've already posted a request for it with a bounty of $20. At that point I might decide "hey, GIMP CMYK is worth $20 to me" and add my $20 to the bounty. If there are 50 other people that think GIMP CMYK is worth $20, then the bounty is over $1000. I know a few people in the printing and advertising industry who wouldn't mind a free graphics app that does CMYK, and it might just be worth the cost of a couple of Photoshop licenses to them. In easy numbers lets say 2 copies of Photoshop are $1000, and we get 4 companies to do that, now the bounty is up to $5000, and lets just say for the sake of arguement that there's some developer out there who, at $5000, considers the project to be worth his time.
The beauty of the system I envision is that, unlike the current situation where we the occasional private individual offering a large bounty ($20k for X functionality in Y package), the bounty is held in escrow (not entirely sure that's the correct term, but it's close enough), so it will be there until someone completes the job. This creates another interesting effect, in that even for rare itches the bounty can become sizable with time. If you and I were the only people in the world who wanted GIMP to have CMYK support, but we each donated $20 a month to the bounty, it would take a long time for the bounty to get up to $5000, but it would get there.
I see this plan benefiting users in that they can help direct the developement of a project that is important to them with a small outlay of capital. The benefit to developers is fairly obvious, in that they can get paid for something they might be doing anyway. It also removes the "pledge drive" feel that many requests for donations have, since it's the users offering money rather than the developers asking for it, hopefully without creating a code ransom situation which seems to be the other way OSS developers are able to make money.
Anyway, there's my idea. Feel free to steal it, shoot it full of holes, or give me the kick in the pants I need to actually get it started, as you see fit.
Two sys admins, a couple of DBAs and an IT manager, working for a large manufacturing company. You're telling me they have the skills, the time and the inclination to knock up a quick competitor to Oracle financials over the weekend?
Your examples completely ignore the fundamental strength of open source: one company isn't doing all the work.
You're correct in saying that the company you describe wouldn't be able to recreate Oracle on their own (never mind the timeframe), but they don't need to. There are plenty of other organizations that need that functionality, and there are plenty of OSS developers with detailed knowledge of OSS database apps who would be willing to do the work.
As a thought experiment, which will hopefully illustrate my point; what if California offered just 1% of the $95M it was prepared to spend on Oracle software as bounties and/or contracts to create the functionality they needed using MySQL or PostgreSQL? How long do you think it would have taken for that functionality and scalability to be achieved?
Now lets expand that; where would MySQL and PostreSQL be if all the companies buying Oracle and DB2 were instead pouring 1% of that money into the developement communities of those two packages?
Yeah, it's not going to happen over the weekend, but with that kind of incentive I doubt it would take very long.
Dude, it's a fucking school.
So the people who managed the deployment were free?
Their called TAs, and at the district school level they often are free.
How about the user training?
Again, it's a school. Training is their business, and I very much doubt anyone at the high school level has any use for the "advanced features" of MS Office. I've never had a use for them, and I've been in the Real World for some time now.
As for training the teachers, you must be joking. I have a few teachers in my family, and the level of computer training they recieve is readily apparent: none at all.
How about the lost productivity time as end users got used to the new app?
I haven't converted many people away from MS Office yet, but in general the only "lost productivity time" I've seen has been the time it takes to install.
How about the conversion problems on the few especially complex documents star office struggles with?
Where exactly are all these "especially complex documents" I keep hearing about? Are these the same ones that MS Office seems to choke on?
It's been said before, and here it is again... free software is only free if your time is.
And yet OpenOffice still manages to be much cheaper than MS Office.
well I'm not saying anything. But, maybe this guy is a little biased
Biased or not he knows jack shit about Linux or the Linux community, and is therefore completely unqualified to make any comparisons.
Anyway, here's my favorite quote from the link:
Any developer documentation as comprehensive, accessible and easy to use as the MSDN Library?
Last semester I took classes in VB.NET and C. In VB I frequently spent hours following little crosslink circles without gaining any useful information. I did all my C on Linux, though, and found that everything I needed to know was a quick 'man function' away.
most children are born from accidental copulations.
"Honestly, it was an accident! I tripped and it just sort of fell in there!"
Sorry, I couldn't resist. I think you meant to say "conception", not "copulation".
Enron. WorldCom.
Well, they certainly weren't averse to risk, were they?
More than one researcher has argued that the world is flat, but that's not true either. These critics clearly have no conception of how science actually works.
I find Flat Earthers quite amusing. A while back a friend of mine decided to join the Flat Earth Society. I was able to disprove every one of their arguements based on my (at the time) one semester of Mechanics, without even having to refer to my book, in an hour or so.
It's truely amazing, and more than a little frightening, what people will accept as "proof".
Prime factorization is NOT a provably hard problem. Mathematicians have yet to show that it is NP-complete. For all we know, it could be quite easy and we just haven't found the easy way to do it.
You're right. In fact, factoring primes is provably easy: One times the number. Oh, except One isn't a factor, by definition. Oops!
It is impossible to factor a prime number. I have seen prime numbers expressed in terms of factors (for instance, 7 could be expressed as 2^3-1) but you cannot factor a prime number.