Frank 
> but isn't this ambiguity already present anyway in TeX? and those
> interfaces have to deal with it? E.g., if you write 1+2 without
> explicitly marking it as a math formula you get a text string 1+2 not
> $1+2$
Whether or not text is in math mode is an important authoring issue
that requires explicit author indication.
For example:
\begin{quote}
Everyone knows that the 2 most important numbers are $0$ and $1$\@.
\end{quote}
André Weil in a celebrated book review took the position that one who
wishes to study the history of mathematics needs to know both Greek
and Latin.
It follows that a mathematical historian writing in English needs
to be able to use the Greek alphabet both for quoting brief segments
of Greek text and for creating mathematical symbols.
Ideally the typeset representations should be different depending on
whether or not one is in math mode, just as with 'a', 'b', 'c', ...,
and '1', '2', ... .
But inasmuch as LaTeX does not come with its own fonts, it seems to me
that the pragmatic thing would be to provide dual handling of \alpha
etc. when possible with the fonts at hand and unified handling otherwise.
Presumably somebody in a locale where the Greek characters get regular
text use will have easy access also to the unicode points.
And I think the difference between α (as a representation here
of the unicode point) and \alpha should be erased in normal LaTeX
document classes. (Sophisticated users could still be given a way
to control the distinction.)
I don't know the internals, so I'm guessing in suggesting that to make
these things seamless for non TeXpert authors it may be necessary for
LaTeX to add its own bookkeeping on math mode independent of TeX's.
 Bill
P.S. I assume that this discussion, which began with your questions
about changing over to the T1 fontenc, is about LaTeX3 and not about
the next 2E (unless the next 2E is to be 3).
