On 11.06.25 21:22, William Peckham via Dng wrote:

Hi,


> About X. I want to see X.ORG <http://X.ORG> spun off by some team that
> has actual ethics and standards, and enough oomph to maintain the spin.
> No one person has that, and no team has stepped up to help in the
last > two decades.

I hope we can fill that gap. Our ethics are pretty simple and
straightforward, just like in the old days of the hacker subculture:

Let's come together, building some great piece of software. Just be kind
and respectful to each other and let's have fun together.

> Frankly, X.ORG <http://X.ORG> is so seriously complex and unmanageable
> at this point that few teams would WANT it!  I know I do not.

Depends on what exactly you're looking at. The Xserver code base indeed
isn't entirely trivial, but far from being unmanagable.

> There are features of Xwindows (remember when that was the software
> name?) that I would like to see maintained that are not part of Wayland
> but ARE grandfathered into X.ORG <http://X.ORG>. The remoting and
> networking features to allow remote clients to use X over network have
> no direct replacement.

That's exactly one of the core reasons why I'm moving forward on X11.
And there's more: compatibility with existing software. We're keeping
the protocol intact, in order to not breaking existing clients and
infrastructure. One of the major problems w/ Wayland is, you have to
start pretty much from scratch (just porting some toolkits just isn't
enough)

We might drop *some* old extensions (eg. xf86vidmode) *IFF* there aren't
practical users for it anymore. Same for certain vintage hardware.
But besides that, keeping compatibility is a major goal.

> But I have doubts about this being the right way to move forward.

Why ?

> Perhaps we need a new and different option.   Perhaps something that
> supports the features in a secure way on TOP of X.ORG <http://X.ORG> ,
> Wayland, or Xlibre.

Since many people seems to declare X11 as "insecure". The core of this
problem already had been solved back in 1996 - the Xsecurity extension.
This might not work well with some clients (which aren't expecting
certian calls to fail and just crashing). And the grade of isolation
might not be sufficient for container-like use case - that's what the
new Xnamespace extension is for.



--mtx

--
---
Enrico Weigelt, metux IT consult
Free software and Linux embedded engineering
info@??? -- +49-151-27565287