I think this may make my points more clear. Hope it helps:

Perhaps the term ‘vector’ art is misleading, and a more explicit title could be ‘processed-effects’ or ‘variable rasterisation’ or something. All operations in a vector package are rasterised at display time, whereas a bitmap is just a pre-rendered, fixed-resolution image. Bitmap editing programs perform operations on these fixed rasterised images only, whereas the vector programs use algorithms to both generate the (bitmap) image and apply effects. Vector processes generate effective bitmaps that are reprocessed for every change in view; take feathering for example: the feathering algorithm is applied on an object for a given view, and the output is essentially a screen resolution rasterised image (bitmap) – but one that is regenerated with each new view. The same applies to drawing Bezier curves; an algorithm is used to determine the output for the view, and once it is determined, the equivalent to a screen resolution bitmap (rasterised image) is generated.

I think I need to make a better distinction between what I mean with:-

1. dealing with bitmaps;
2. dealing with ‘vectors’; and
3. applying effects to both vectors and bitmaps

1. As has been rightly mentioned in responses above, you could include bitmap editing tools in vector packages that work on bitmap objects (objects with a fixed, predetermined rasterisation). These include all the tools to select pixels of similar colour, warp, smudge, and alter hues etc. that are typically found in bitmap editors.

2. This is in addition to all the tools to process ‘vector’ effects on vector (variable resolution) objects that are found in vector packages.

3. But the idea that I poorly conveyed in my first post is that you could conceivably draw/paint etc. objects on top of other objects (vector or bitmap) which applied further ‘vector’ processing (i.e. generate bitmapped effects for the view you are at). An example of this would be to have a collection of bitmap and vector objects on a page, draw over them with, say, the same smudge tool you use on bitmaps, and all objects underneath smudge – including the vector objects. The method I discussed in my first post would allow this (or any equivalent method that is used – say, for generating vector feathers etc).

An example of this is already available: consider the vector feathering of, not a traditional vector object, but of a bitmap. You are applying a vector process to a pre-determined rasterised image. How? By post-processing that bitmap in a way that is independent of its fixed resolution. Hence why you can zoom in to this bitmap, see its pixelation, but not see any pixelation from the feathering processing. In this case, the processing is limited to edges – by definition of feathering, but there’s no reason, in principle, why you couldn’t draw this effect freely with an appropriate tool.

So in general, why couldn’t any effect be simultaneously applied to bitmap and vector (varying, screen resolution bitmap) objects, and all done from a vector package, which, by definition, and unlike bitmap packages, is not limited to working only with fixed resolution objects? This is what I wanted to discuss. I'm sure a certain company (who’s name I said I wouldn’t mention again in this post), is planning on heading more distantly down this road, as they have eluded.