Hans Aberg writes:
 > One then becomes independent of that old PlainTeX once for
 > all, and is free to build up an entirely new consistent logical structure.

You mean that one is free to build up an entirely new _inconsistent_
logical structure . . . !  We all know about creeping featurism.  Even
small subcomponents such as NFSS show how difficult it is to get it
right.  (And the NFSS is not modular - you can't easily add a new
`axis'.  (Note - I said `easily'.))

 > This will perhaps prepare for a new version of TeX, which can take care of
 > the new LaTeX3 structures efficiently.

Let's hope so.  As I said at the beginning, I hope we can progress to
a unified e-TeX/Omega platform.

 >   That is, if it can be made practical... So give it a try, I would say.

Oh, I have an open mind.  I am sticking around for the long haul.  I
just hope we don't lose others along the way.

 > (It is probably more difficult to implement the feature at a later stage
 > than removing it if it does not work.)

Agreed 100%.