Re: Will PC manufacturers change the selection of windows ?
Posted: Wed May 15, 2013 10:06 am
Yo:
It is my opinion that an OS environment that is decently designed, can run critically needed applications (commonly used server packages, Microsoft Office, Photoshop, AutoCAD, etc), games, and has a proper user interface (both at the command line and graphical level) and has a significantly lower sale price (say US$20.00 - US$60.00) than MS Windows will be able to gain ground very quickly.
Price is a much bigger driving influence for adoption than I've seen it given credit for so far in this thread. A decent NT32 compatibility layer + a POSIX compatibility layer will probably yield very promising results if shipped at a low price, on top of a kernel which offers the promise of the ability to scale to much higher needs. For example, a kernel which supports NUMA, a generic binary-portable driver interface, and has an asynchronous native API (along with other features), will probably be very attractive to users of all types (casual, gamer, server-admin) if the userspace environment constructed around it is focused highly aggressively on ease of use, even down to the command-line environment, should the user choose not to run the GUI (server, etc).
Unique features, feature richness, and all of that have a huge role to play, but the ultimate determinant of uptake will probably end up being raw cost -- cost of procurement, cost of transition, cost of retraining. You can easily make the latter two costs negligible if you take the time to create a decent user interface, and some solid compatibility layers.
--Peace out,
gravaera
It is my opinion that an OS environment that is decently designed, can run critically needed applications (commonly used server packages, Microsoft Office, Photoshop, AutoCAD, etc), games, and has a proper user interface (both at the command line and graphical level) and has a significantly lower sale price (say US$20.00 - US$60.00) than MS Windows will be able to gain ground very quickly.
Price is a much bigger driving influence for adoption than I've seen it given credit for so far in this thread. A decent NT32 compatibility layer + a POSIX compatibility layer will probably yield very promising results if shipped at a low price, on top of a kernel which offers the promise of the ability to scale to much higher needs. For example, a kernel which supports NUMA, a generic binary-portable driver interface, and has an asynchronous native API (along with other features), will probably be very attractive to users of all types (casual, gamer, server-admin) if the userspace environment constructed around it is focused highly aggressively on ease of use, even down to the command-line environment, should the user choose not to run the GUI (server, etc).
Unique features, feature richness, and all of that have a huge role to play, but the ultimate determinant of uptake will probably end up being raw cost -- cost of procurement, cost of transition, cost of retraining. You can easily make the latter two costs negligible if you take the time to create a decent user interface, and some solid compatibility layers.
--Peace out,
gravaera