Well what I'm saying is quite obvious, I'd even say it is a truism, yet hn readers often downvote even the most basic statements (probably by lack of understanding the topic).
But here you ask constructively so I'm gonna answer pedagogically:
Skia and Cairo are two of the most used 2D renderers.
A 2D renderer is an angular piece of software used by every GUI software. It has a key role in 1) rendering correctness (anti aliasing, no bug, no tearing) and 2) rendering performance.
2D renderer performance is key for all graphical workloads, from the scrolling/animation performance of your browser to the smoothness of any non cli software.
As humans we strive for progress, we want tomorrow to be better than now, therefore we want more correctness and more performance by 1) improving 2D renderers where it matters the most and 2) making sure that the right 2D renderer is used by softwares.
For example a big reason of slowness of Firefox a few years ago was because they used Cairo, by switching to Skia they got "free" significant performance gains for the pleasure of all users and even for the pleasure of HN readers currently scrolling this thread.
Skia is the only renderer to have a lot of human resources allocated to it (thanks to Google) and this translate by it being the most performant 2D rendering, sometimes by order of magnitudes.
But skia can become even better by integrating new optimizations, Vulkan, NVpath, AVX, etc
So now you can understand the original statements:
Everybody should switch to Skia
This means that new and existing software would significantly benefit from switching from Cairo to Skia (this argument was sufficient before the news but is even more important now that Cairo is unmaintained)
As an end users you will probably get more FPS, accuracy and lower power consumption (which is ecological btw).
This thread about Cairo being unmaintained could attract potential new contributors.
Instead I'm saying to those new contributors:
New human resources if not allocated on Skia should go on next gen renderers like fastuidraw
Because their work if on Skia would benefit order of magnitude more people.
Else they could work on next gen renderers, having talked to the devs of blend2d, fastuidraw, etc I have big doubts they'll ever one day be able to replace skia (too many missing features e.g full SVG compliance) but their performance paradigm shifts could be backported into skia once proved, so contributors could have a huge impact here too.
Yep, that makes a lot of sense. Skia appears to be BSD licensed which is good. Not having used either, are the capabilities of Skia a superset of Cairo?
