Re: Why Linux OS Desktop is not popular?

From: Mats Liljegren
Date: Mon Jun 03 2013 - 11:44:05 EST

On Tue, May 28, 2013 at 11:42 AM, Amit C
<amit.linux.os.desktop@xxxxxxxxx> wrote:
> Hi All,
> Linux OS Desktop has not become very popular because of the following reasons:
> 1. It is not because its GUI is not good.
> 2. It is because if I buy a new gadget/device/hardware, I am not sure
> whether it will work with Linux Desktop.
> 3. The reason for number 2 is that the vendor introducing
> gadget/device/hardware does not write a device driver for Linux.
> 4. The reason for number 3 is that developing new device driver on
> linux is hard.
> 5. The way to solve number 4 is to move to data oriented device driver
> development where the framework for driver is already available for
> all kinds of devices (PCI, USB, Firewire, etc.) and the only thing a
> device driver developer has to do is to collect io/virtual addresses
> from the vendor/manual and feed into a table (structure of structures,
> etc.) and the framework will pick up this data and configure the new
> gadget/device/hardware instantly.
> I do not want to pay for Microsoft OS but can Linux community give me
> a usable Linux Desktop. I have both Windows and Linux on my laptop and
> I use Microsoft just to connect to Internet because the USB dongle
> that I use for connecting to Internet does not work on Linux. I am in
> India and here USB dongles connecting to Internet are very popular
> because you can use them anywhere.

The big difference between Windows and Linux is that Windows is
developed by a single company, while Linux is developed by a community
including several companies as well as hobbyists. You can become part
of this community by participating in the development. It is not as
easy to become part of developing Windows.

The way Linux progress is by people realizing they have a pain they
want to fix. The fix is done with one or several patches. If the
maintainers like the patches, they become part of Linux. This is how
these pains are fixed.

You start this by experimenting for yourself until you've found a
solution you like. Then you send out patch(es) labeled "RFC" (Request
For Comments) to LKML to see if people think your idea will work.
There is a long way to go then until it actually ends up in the kernel
since there is a lot of quality assurance needed to be done, but
eventually the pain gets solved.

It is very hard to discuss any solutions without those RFC patches,
since no one knows whether the idea will work or exactly what the idea
consists of. So it's unlikely you'll get much constructive response
until you've tested the solution yourself and sent out patches for
comments so others might test it. Just keep it as small as possible
while still proving the idea.

I look forward to your patches, so I can better understand what you
want to accomplish.

Mats Liljegren
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at