This road map looks like a bunch of loose ideas (except for one point).
Re. 1. Adding the ability to change colors and shapes for all widgets is OK. Although it's not a matter of life and death. However, the idea of using CSS for this purpose is pure fantasy and utopia (and additionally: idiocy). CSS is bloated and causes huge memory and CPU overhead. Adapting existing code to support CSS (Gtk, Qt, WinAPI, Cocoa) is a lot of work. Also, CSS was developed for HTML formatting, not for styling controls in operating systems. It is the complete opposite of what developers of windowed applications need. You can try to develop something similar to styles in Delphi. But that will also take a lot of work. There is still the question of how to handle CSS in Lazarus. Because CSS are human-edited text files, they may contain errors. There is known such a commercial (and very expensive) HTML and CSS editor. It's Dreamweaver. Does he always do well with CSS code? The last time I used it was about 12 years ago. And as I remember, there were problems with it (not that I consider this program to be bad, it's just that it's quite a complicated issue). Unless these files will not be modified by a human, but by a graphical editor. But that still requires Lazarus expansion and a lot of work (time, people).
Unless the originator of this roadmap has in mind creating a brand new portable library. This is complete fantasy and delusion. Just look at how much work it took to create such libraries as FireMonkey or JavaFX (I won't mention Qt, because it's years of work for a battalion of programmers). Yes, a cross-platform library for Lazarus and FPC would be useful. But this topic has been discussed many times before. And creating something new that uses HTML, CSS and JavaScript is a dead end. That's why there are still people who choose Lazarus, because it allows you to build windowed applications without trendy shit like Electron and node.js. Trabants in the GDR were also popular, but the quality was worse (but at least they were simple to build unlike WebKit).
Re. 2. As above (in 1).
Re. 3. Web-Assembly is desirable.
Re. 4. Hmm…
Re. 5. Lack of details as to the intentions of the originator of the road map.
Re. 6. That would indeed be useful. But... it will take at least a few years to achieve the capabilities of open source engines (OGRE, Torque3D, Irrlicht), provided that the project is developed regularly by at least 4-5 people. So it's work for years.
Re. 7. This is the most important point. And that should be in the first place.
Recent
