I think Frank's outline of future plans is a good start but let me toss in 
my 2 cents.

Overall, I would like to see some focus shift from just developing drivers 
to ensuring the infrastructure remains solid and up to date.  We can beg 
hardware manufacturers all we want to release specs but some of them have 
simply made the decision not to.  It is important to revisit that issue from 
time to time but for those hardware manufacturers that will not release 
specs we should try and develop a solid, stable infrastructure that will at 
least encourage them to write their own DRI drivers.

The development of the DRI architecture and most of its drivers were done by 
Precision Insight/VA Linux.  Much of the architecting was done in a closed 
manner.  By that I mean it was done by PI/VA employees and the discussions 
were predominantly done internally.  The result is, I believe, a large 
amount of the DRI knowledge did not get dispersed to the community.  Things 
like why something was done one way and not another, why something wasn't 
done, what still needs to be done, where can things be optimized, what short 
cuts were taken, etc.  There exists some old DRI architecture documentation 
but it is out of date.  I think it would be a useful exercise to try and tap 
into the brains of some of the DRI architects to update the documentation 
that exists and outline future enhancements that could/should be done.

I know there has been some discussion, both internal at PI/VA and elsewhere, 
about developing multimedia/video DRI type infrastructure.  It is probably a 
good idea to work with some of the Linux Video guys to try and better 
coordinate work between the two projects.

In my opinion the ultimate goal should be to get Linux graphics/multimedia 
at a level comparible to other platforms.  That means everything from 
infrastructure to documentation to drivers and setup utilities.  We need to 
expand our focus beyond just (open source) drivers.

Finally, Frank, when you are drafting up your letter to hardware vendors 
about "why open source driver development is good for you" remember that by 
now most of them have heard all the reasons for open source.  Although you 
want to remind them of the reasons you want to take it to the next level.  
If you want them to release specs you need to convince them that the 
expertise exists in the open source community to make use of the specs and 
that they will get something in return for their troubles (i.e. a stable, 
working driver that won't reflect badly on their product).


David



_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp


_______________________________________________
Dri-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/dri-devel

Reply via email to