Linked by Thom Holwerda on Tue 27th Sep 2005 17:24 UTC, submitted by Linuxfanboy
"I do not consider Windows ready for the desktop. I found it difficult to use, buggy and lacking in security. I also found technical support lacking. While Windows captured a significant portion of the desktop market, the product is clearly not a good fit for consumers who do not understand the risks associated with logging on to the Internet. The costs of providing aftermarket products can run higher than the price paid for the hardware."
Thread beginning with comment 37287
To read all comments associated with this story, please click here.
To read all comments associated with this story, please click here.
Member since:
Linux, nor Windows, are ready for the Desktop as far as Joe user sees it.
A real Desktop OS would have the qualities as follows missing from both Linux Distros and Windows Distros:
1. Predictability (Uniform Interface, Grammar, Etc...)
2. Simple Effective Security (than by obscurity)
3. Common Sense Interface (5 min learning curve for 6yr old)
4. Simple, Free/Cheap, Extendability (i.e. Apps, but not viewed as files containing data that executes to peform a function)
5. Driverless operation (A common hardware interface per device type, so Network devices ALL work alike, wireless or not, video, sound, etc...)
6. Blind Automatic Fail-Safe functions for EVERYTHING.
7. Auto-Live-Repair/Patch. Problems fix themselves.
8. Human-Readable Error Messages would help.
9. The absense of a User-Visible file system.
--organized like the internet instead, with a portal
10. no advertisments.
11. no spyware
12. no virus software
13. No Software Companies (ANY OF THEM)
14. Common API (for apps, duh, that is what is means)
How can we do this? Simple, actually, VERY simple:
Hardware.
Yup, get rid of the software and the whole world is better. Each peice of hardware has a common interface for data transferral from the OnBoard Programming. The system only runs in one fixed mode (64/128-bit). The BIOS provides all kernel level functions, the onlfy software would be for an interface.
The programming should be abstracted and clean enough that a Freshmen High School Computer Science course should enough to turn out a new IS at will, as a starter homework project for the first week in school.
The only thing that is stored on unreliable storage (i.e. hard drives) is data that changes too frequently. Unmodded data can be automatically, by the BIOS, copied to a faster more-stable device.
The InterfaceSoftware resides on its own reliable storage. RAM is already split by the BIOS for 'kernel' space, app space, interface space, and flex-space, which is taken care of by the BIOS. Boundaries are also in charge of by the BIOS.
So does this sound impossible??
Ever see a PalmOS Device? It comes kinda close.
Ever see an Apple PC? Getting kinda closer, but not quite, that is much more thanks to control freaks than anything.
How long would it take to make this a reality, if the entire computing world stood behind such an objective, and someone with some damned brains for once ran it??
I'd say about six months, tops, twenty developers for the BIOS/kernel, twenty for the Interface API, twnty for the Sound API, Network API, Video API, Storage API, and a few other groups to take care of the additional generic groups of classes of device.
Then the hardware companies to make their hardware compatible with the scheme.
Who would be best to run something like this? Well, me of course, heh. But, no, seriously, a five year old red headed freckled little girl. Yuppers.
--The loon