Canonical are evaluating different options for improving the display server in Ubuntu as it seeks to span multiple devices, Ubuntu Community Manager Jono Bacon has recently confirmed.
The news comes by way of response to questions raised on the status of Ubuntu’s adoption of alternative display server ‘Wayland’.
But before we jump to precisely what Jono has said – and what it might mean for us – let’s go over the basics of what a display server is and what Ubuntu currently uses…
Display Servers, Serving You
At present Ubuntu’s Unity interface runs atop X server, a sort of ‘middle-ma
X is efficient at doing what it does but it has a problem: it is built upon code that’s almost 30 years old.
A mature code-base brings with it legacy support for features and functions that are no longer needed on modern desktops, much less phones and tablets.
Intending to address this is Wayland – a relatively new display protocol that integrates features (like compositing) directly into the display server rather than adding them on as an afterthought.
Ubuntu founder Mark Shuttleworth expressed an interest in using Wayland as Ubuntu’s display server back in 2010.
Ubuntu’s needs from a display server are now different…
In the intervening years since Ubuntu’s initial interest in Wayland their needs have change considerably.
The project has has gone from being solely focused on the desktop market to one spanning multiple form-factors, like Phones and TVs. With each form factor boasting different screen-sizes hardware configuarations, input-devices, and more it’s efficiency, performance, and adaptability that are most prized.
Stressing that no decision has yet been made, Jono Bacon encompassed the dilemma as thus:
“What we don’t want to do is say: ‘Let’s move away from X which includes a tonne of functionality that we don’t need, that drags us down’ and move to something else that includes a tonne of functionality that we don’t need and drags us down.’
‘Things are still very much up in the air right now, nothing is finalized.’
Adding that ’the simple reality is that X doesn’t meet those needs, Wayland doesn’t meet those needs.’
Several weeks back, Bacon mooted that one possible solution on the table was the ’….likelihood …of building our own display server’ - but stressed that such a decision, and all of the complexities it would entail, was yet to be made.
‘Things are still very much up in the air right now,’ he emphasised, ’Nothing is finalized.’
The ‘What If?’
If Ubuntu don’t decide to go with either X or Wayland – and, again, no such decision has yet been taken – what options might we see them use?
Android is powered by an open-source display driver that is neither X or Wayland, but is capable of working work across mobile, tv, tablet and laptop interfaces. Could a fork of this be one solution?
Another option would be to create their own – either from scratch or based on an existing project. But the quagmire of complexity that would go with such an effort is vast – even for a company as seemingly well-resourced as Canonical.
The good news is that we don’t have long to wait before we find out.
Firstly, Bacon says that more information on the decision making process ahead will be available in the ‘next couple of weeks’.
Secondly, we can expect the first installable image of Ubuntu Phone - along with its source code – sometime towards the latter part of February/Early March. This will allow developers to go hands on in assessing the needs and realities of creating a pan-device display server.
It Doesn’t Even Matter
Regardless of what Ubuntu do choose to use – be it their own, Wayland, or Android’s display server – one thing is certain: the graphics performance will be better than what we have now.
Whatever Canonical/Ubuntu end up deciding on we can be assured that they wouldn’t jump ship from what currently works unless a speedier boat was waiting in the water…