Graphical design can be fairly complex. Some vector might contain dozens or even hundreds of control points, and a vectorgraph might contain many of layers of v

Graphical design can be fairly complex. Some vector might contain dozens or even hundreds of control points, and a vectorgraph might contain many of layers of vectors/groups. Rendering the graphic itself and the software UI element might be a challenge for browser engines which is not made for this kind of task, especially on slow computers. I worked for an ad publishing firm many years back, it was common for us to receive vector graphic files that are hundreds of megabyte large. We used Adobe Illustrator and CorelDRAW to handle those files on some fairly fine computers, and yet some graph files from our clients still managed to crash the computer due to RAM exhausting etc. I image it makes sense for an industrial capable software to want to run as native as possible. Another thing is color management. Most graphic software supports RGBA, but for publishers, CMYK is also required. I'm not really sure if browser engines supports these color system well enough (not just color conversion between the two, but also color profile for the monitor), because if something is mess up there, it can be expensive to fix (for the ad publisher).