I didn’t really know how to phrase this question, sorry. For example, when eyes are added into a character they are usually spheres that are disconnected from the head, but in the same object. Is this bad practice in non-organic models?
If I only have to move a few objects then I just move (append) them separately but if you have a lot of objects and you want to move (append) all of the files to another .blend file then you may want to “join” or “group” the objects to make it easier to move them, otherwise it’s not a problem to keep the objects as separate.
If I build a house for example I leave the walls, windows and doors as separate objects. You can always Group or Join the objects later if need be.
If you explain your situation then it would be easier to answer.
pg
In most cases the eyes are their own object parented to the head.
Again, I’m sorry for my ambiguity. Attachments aren’t working for me at the moment so I’ll try to upload an image later.
As an example, imagine one object containing two distinct pieces which aren’t connected by any edges. The model didn’t start out as two objects, but with two primitives in one object.
That is completely normal. It just comes down to functionality. Just depends on what you what do do with the mesh. There are many things that could determine what you want to do. If you have a modifier on the object it will operate on everything in the object. So you would take things out of the object that you don’t want affected or keep them in and use vertex groups. Even for a robot there is no reason you could not have it as a single mesh with separate parts and then animate them with bones and weights. Each mechanical part would get a 100 percent weight of the bone controlling it. Just throwing things out there. It really only depends on what you want to do. There are advantages to keeping things separate too, as has been pointed out here.
Thanks guys, you’ve answered all I needed to know.