This is what I call a stream of consciousness, edited. Please don’t rip my head off, call me names, etc.
Microsoft is a product of the commodity hardware era.
That is, Microsoft is a company that couldn’t have existed previously, in an era where computers cost tens of thousands of dollars (or hundreds of thousands of dollars). When the computers themselves were expensive, the hardware was the item of value.
There was only one business model.
Companies producing them were not thinking about the software running on them as valuable (and even missed patenting things like windowed interfaces, pointing devices, and networking protocols), because they through the valuable thing would be the hardware forever.
Modern business models for personal computing.
In the modern era, we have three major business models: hardware, software, service.
Apple represents the hardware model: they’re making billions selling top-end hardware (phones, laptops, computers, tablets). They’re basically giving away their software. They’re trying to commodify the software — making it trivial and making it easier for thousands of companies and even individuals to produce competing versions of every possible application on their hardware.
Microsoft represents the software model: they’re making billions selling software licenses for software that runs on any kind of hardware. They’re actively trying to commodify hardware, making their software run on any kind of hardware from hundreds of manufacturers, and driving prices of home computers, netbooks, and tablets down to a couple of hundred dollars.
RedHat (and Linux vendors in general) represents the service model: they’re making millions selling service contracts for free software that runs on that same really cheap hardware that Microsoft’s model has been driving.
How might that affect service, software, and hardware?
It wouldn’t be any surprise if service companies didn’t care much about how user friendly the software is: after all, they only make money if you need support to use it.
It would also not be surprising if software companies didn’t care how confusing the hardware choices are — they want to make sure that there are lots of hardware companies, because that’s how you maintain the commodity status of the hardware.
And of course, it wouldn’t be a surprise if hardware companies made high end hardware, released their software updates freely or cheaply, treating the software like firmware designed to run only on their specific hardware models. These companies might have the lowest barriers to switching hardware architectures and creating new hardware form factors, since they’re providing a stand-alone platform (the software is like firmware) without much regard for disruption of software platforms.
Which market is easier to enter?
It seems to me like the most sustainable of these models, and more importantly, the most favorable to the consumer … is the software model. I prefer business models which encourage competition, and I think that hardware and service models do not.
On the one hand it’s easier to grow a software company than it is to grow a new service company. It’s very hard to compete with the big guys on service because of reach (you can’t provide world-wide service), and as a result, very few service companies make it past the mom-and-pop size to medium-sized businesses employing more than 10 people.
On the other hand, it’s easier to create a new software company than it is to create a new hardware company. It’s very hard to compete with the big guys on hardware because of costs (you can’t get hardware costs down without substantial volume).
You only have to look at the past few years to see that it takes a long time for a new hardware company to grow to a useful size, and that service companies tend to consolidate long before they become competitive.
But you’re not taking X into account
Of course, this is a fairly simple interpretation of the personal computer marketplace. I know that. Tell me something I don’t know below.