‘Just Enough Operating System’

Mass-Customization

Today we discussed the concept of Mass Customization, where corporations find that perfect median between adapting to individual customers and still being able to manufacture at a mass level to realize the majority of the economies of scale. Some of the products we decided as a class that worked best customized were things like shoes, cars and computers (in the case of Dell). A recent article discusses the possibility of customizing one’s actual Operating System. There has been some personal customization from individual users, but they have often been unauthorized by the vendor. Can this be done at the mass level? A better approach is to create an environment using only the Operating System resources needed for a particular application set. This is known as JeOS (just enough operating system). The question is: How do we move to a JeOS environment? Particularly with Linux, although it has been designed to be customized, very few organizations take advantage of this capability. Linux's modular architecture is ideal for creating JeOS because it can be easily stripped down and modified.

Advantages

Mass customization would be able to realize the benefits and efficiency of mass production, but with ability to tailor an OS for different use cases. And because this tailoring is achieved through the arrangement of standard Linux components, it is fully supported. Other advantages include:

• Reduce Maintenance Costs
• Strengthened Security
• Greater Agility – helps IT reduce time required to deploy new servers

The future:

All that's needed now are tools to make mass customization a reality. We didn't achieve mass customization of cars until Ford thought up the assembly line. We need the equivalent of the assembly line in the OS world: tools that provide rapid, fully supported mass market efficiency, reliability and consistency, while allowing for individuality.

My thoughts:

When is it right to customize at the mass level? This is something that we discussed in class, and is Linux at a good place to do so? I would argue that this is most definitely a profitable and logical progression. This differs from some of the examples that we talked in class, because this actually makes financial sense. Companies would definitely go for a cheaper, easier to learn and easier to maintain system if that is all that they require. Where in the Adidas or even cars example the mass customization was being done purely for ‘looking cool’ or superficial reasons, here we see a more functional approach behind it. This parallels more of the customization that we saw in Dell. Where different users require different things from their computer, taking that one step lower, different companies require different applications in their Operating System.

For the multiple advantages discussed above this will work. The pressure is now on Linux on developing a system where then can build a manufacturing process which can accommodate this production method. What if they are able to? A question was posed today, that can this be a competitive advantage, especially with such large competition in the Operating System market? It was also mentioned in class today that it could not in fact be a competitive advantage most of the time, but this may be one time where it could be possible. The fact is that the Linux Operating System from a design perspective has an advantage that it has not currently been exploiting. Once Linux is able to create a method of mass customizing its Operating Systems they will have a clear edge over Windows or any other system. The following questions arise: Does Windows have the capability of doing this? How long will it take for other Operating Systems to match this personal-approach? How permanent can Linux’s first mover advantage last?

Regardless of the answer to these questions, this is a prime case of a product that is ready to be mass customized, and Linux is at the front of the line to be the first to do it.