Actually I think you've missed the point of the article. I'll address your points one at a time.
When it comes to forking of a code base, the benefit of having one company provide a solution is known throughout the industry. The forking you mention of Microsoft products is to address different sectors of the market. It is a controlled division, and all of the forked streams are under the same central management. Thus constomers can build their infrastructure around these streams with the confidence that Microsoft will not abandon the platform they are based upon. This is not the case with OSS. FOr example, in the KDE vs. GNOME debate, the two streams are incompatible (to a large extent). Either one could be eliminated in favour of the other should the dynamic of the open source community change. Also, this is only one example. In the community atmosphere of OSS, you inevitably have differing opinions. This is good in some cases, but not all. With the centralized, shared source approach, upper management can make a decision and force the issue - something that's good for the company, the product, and the customers. In OSS, the differing opinions often lead to splintered development efforts travelling in opposite directions, confusing users and diluting already scarce developer resources.
The viral nature of the GPL is also an issue for most customers. The GPL only protects intellectual property belonging to the public domain - essentially ensuring that IP that is in the public domain cannot become privatized. However, IP is a fundamental building block of today's knowledge based ecomomy. Although the patent system and other legal issues surrounding IP need to be reviewed, the concept of creating something and making money from it is core to success in today's economy. The GPL removes that right from the creators of the product. SO, I would argue that the GPL does not encourage the creation of new IP, it discourages it, since by it's very nature, the creator can not in any way profit from it's creation.
This leads me to my next point, that the majority of software developed is developed for specific customers. However, the majority of revenue generated from software development is not. Without designing software that can appeal to a large base of users, there is no economy of scale possible. Companies like Microsoft, Oracle, and other large software development houses are able to create large, complex, pieces of software because they are able to generate enough revenue to pay for the necessary R&D. If you remove thier ability to leverage the creation of their software, and force each customer to pay the entire costs of the software they need, I think you will find the pace of software development in the industry overall will slow to a halt, and the result will be software created only for those orginizations large enough to afford custom-built software.
I think Microsoft's point in this article was that although there is a place for OSS, there is also a place for traditional closed source development. Their new initiative, "shared source" is an excellent way to leverage some of the ideas in the open source movement and apply them to closed source development. In the market today, companies understand that other companies need to make money, and are willing to pay for their software. Shared source is an excellent way for customers to have an influence in the development process, to understand how the software the buy works, and build upon it. I think it's a great step forward for a company that has been so secretive and controlling to date.
Slashdot Mirror
Actually I think you've missed the point of the article. I'll address your points one at a time.
When it comes to forking of a code base, the benefit of having one company provide a solution is known throughout the industry. The forking you mention of Microsoft products is to address different sectors of the market. It is a controlled division, and all of the forked streams are under the same central management. Thus constomers can build their infrastructure around these streams with the confidence that Microsoft will not abandon the platform they are based upon. This is not the case with OSS. FOr example, in the KDE vs. GNOME debate, the two streams are incompatible (to a large extent). Either one could be eliminated in favour of the other should the dynamic of the open source community change. Also, this is only one example. In the community atmosphere of OSS, you inevitably have differing opinions. This is good in some cases, but not all. With the centralized, shared source approach, upper management can make a decision and force the issue - something that's good for the company, the product, and the customers. In OSS, the differing opinions often lead to splintered development efforts travelling in opposite directions, confusing users and diluting already scarce developer resources.
The viral nature of the GPL is also an issue for most customers. The GPL only protects intellectual property belonging to the public domain - essentially ensuring that IP that is in the public domain cannot become privatized. However, IP is a fundamental building block of today's knowledge based ecomomy. Although the patent system and other legal issues surrounding IP need to be reviewed, the concept of creating something and making money from it is core to success in today's economy. The GPL removes that right from the creators of the product. SO, I would argue that the GPL does not encourage the creation of new IP, it discourages it, since by it's very nature, the creator can not in any way profit from it's creation.
This leads me to my next point, that the majority of software developed is developed for specific customers. However, the majority of revenue generated from software development is not. Without designing software that can appeal to a large base of users, there is no economy of scale possible. Companies like Microsoft, Oracle, and other large software development houses are able to create large, complex, pieces of software because they are able to generate enough revenue to pay for the necessary R&D. If you remove thier ability to leverage the creation of their software, and force each customer to pay the entire costs of the software they need, I think you will find the pace of software development in the industry overall will slow to a halt, and the result will be software created only for those orginizations large enough to afford custom-built software.
I think Microsoft's point in this article was that although there is a place for OSS, there is also a place for traditional closed source development. Their new initiative, "shared source" is an excellent way to leverage some of the ideas in the open source movement and apply them to closed source development. In the market today, companies understand that other companies need to make money, and are willing to pay for their software. Shared source is an excellent way for customers to have an influence in the development process, to understand how the software the buy works, and build upon it. I think it's a great step forward for a company that has been so secretive and controlling to date.