RC3 Install Report

Rahul Sundaram metherid at gmail.com
Tue May 18 13:43:53 UTC 2010


On 05/18/2010 07:03 PM, Bill Davidsen wrote:
>
> And the idea that FOSS is always better than closed source software isn't based 
> on blind belief and dogma?

I don't see anyone making that claim here.  You are building up a strawman.

>  That having source is going to work better than 
> having design documents for the hardware and firmware being used?

Nvidia doesn't provide any "design" documentation either.   They
consider it a trade secret as well. 

> True, many older (3-4 years) ATI cards and laptop chipsets were supported before 
> mode setting came into frivers. Now, even with modesetting disabled, the 
> hardware only works in VASA mode. The developers are more interested in adding 
> 3D functions and wobbly windows than getting an older chipset to work usefully 
> the way it used to.
>   

Atleast within Fedora, we have consistently prioritized basic
functionality over any "wobbly windows".  Look,  Red Hat pays developers
to work on the Linux kernel and Xorg drivers in Fedora and to suggest
that they are more interested in wobbly windows over robust supportable
drivers.  is hilariously inaccurate

> As Antonio Olivares mentioned, just getting a kernel.org source to build with 
> recent drivers is an adventure, So the better support for FOSS is more in 
> principle than practice by his evaluation.
>   

Let's face the facts.  If you are going to talk about non technical end
users,  they are not interesting in building any kernel from source
either.  He was answered in the users list FYI

> My point was that users care about software working, not it's purity. People 
> wouldn't be looking for other drivers if the ones in the delivered kernel worked 
>   acceptably for them. That particularly means that suspend and hibernate work 
> on laptops, lest that wasn't obvious.
>   

Developers care about software working as well.  It is a not a question
of purity.  Proprietary drivers are plain unsupportable within Fedora. 
Suspend/resume works just fine for me.  If you have a problem with that,
a reference to a bug report would be useful but don't confuse
implementation issues with the fundamental approach. 

Rahul



More information about the test mailing list