Do we need a modularized X windows protocol?

Anthie

Anthie

We know that personal computers are assembled with independent parts which can be produced by different manufacturers. CPUs, memory, disks, DVD drivers, video cards, monitors, etc work fine together by connecting a public bus-line with standard interfaces in a computer. The X windows protocol is too big and complicated to maintain for an open source community. Can the protocol be split into 4-6 parts which communicate with standard interfaces and supply a workable GUI? In this way, a open source team can focus on one part of the X windows system. More development teams can join in progress of the X windows software development.
 
You raise a good point. I do think we need one, but the various vendors supporting X.Org and Wayland seem to be going in the opposite direction. I really think this is an area where Apple could help out the BSDs by open sourcing their graphics stack, it can't hurt to dream I guess.
 
retrogamer said:
I really think this is an area where Apple could help out the BSDs by open sourcing their graphics stack, it can't hurt to dream I guess.
Click to expand...
That would really help, yes. But they would need to be some central pilar of trust and keeper of the trunk. Imagine what would happen when the first "enhancements" creep in which will tie it to some poetterware. That would not be in our or their interest. Also the licence would need to keep it free, forbidding the switch to GPL. Sadly, today I see this as neccesary.
 
Anthie said:
The x-window protocol is too big and complicated to maintain for a open source community.
Click to expand...
Why do you think it's too big and complicated to maintain? The FreeBSD OS code is big and complicated and the FreeBSD Foundation doesn't seem to have a problem managing it. In a similar fashion the code for Xorg is maintained by the X.org Foundation. Why do you think they can't cope?

http://www.x.org/wiki/XorgFoundation/
 
The biggest problem of X has always been the insistence of the X developers writing their own hardware drivers for graphics hardware and not trust the hardware vendors to do a proper job of writing drivers for their own hardware (a very big LOL from me). This is thankfully now changing to the right direction with the KMS system and putting the drivers to a the proper abstraction level so that we may finally start to see a situation where a new graphics card could be made to work with X on FreeBSD/Linux/etc by installing a small binary blob from the vendor and be done with it. In and ideal world there would be no need for specialized drivers for any graphics hardware and every single graphics card would "talk" the same language with the operating system. I just don't see that happening with the commercial competition that is the reality between hardware vendors.
 
SirDice said:
Why do you think it's too big and complicated to maintain? The FreeBSD OS code is big and complicated and the FreeBSD Foundation doesn't seem to have a problem managing it. In a similar fashion the code for Xorg is maintained by the X.org Foundation. Why do you think they can't cope?
http://www.x.org/wiki/XorgFoundation/
Click to expand...

Agreed. I think that FreeBSD does a good job distributing packages and ports of Xorg for us; it meets my needs.
 
You must log in or register to reply here.
Back
Top