Dear Cristi,
Thank you for making the effort to understand my approach to physics, but I conclude from your questions and remarks that I have not been successful at conveying the the essence of the approach. I will try to be more explicit. Three historical examples will help illustrate what I mean.
(1) General relativity. Before 1911 (let's say) probably no physicist ever heard of Riemannian geometry, and even fewer (which means zero) suspected it might be of any use in physics. Mathematicians themselves, Riemann excluded, thought of it as nothing more than interesting for its own sake, something like number theory (at that time, of course). This is why historians of physics cannot even suggest when general relativity would have been discovered had it not been for Einstein. Maybe not to this very day. It is with deep admiration that I look at Eistein's contribution, but I cannot say the same thing for what followed -- especially since I wasted a lot of time in it myself. Once it had been established that space is not the rigid thing Kant believed it to be, ideas for generalization were not hard to come by. Examples: Increase the numbers of dimensions and then fall back to 3 plus 1 by some imaginative procedure; add structure by considering torsion in addition to curvature; remove structure by falling back to conformal geometry. There is enough here to sustain searching for possibilities for almost a century so far, and probably for a long time to come.
(2) Dirac's equation. The unification of Pauli's spin with relativity was a burning question around 1927. Dirac had a brilliant idea that solved the problem, but since the solution rested on a stroke of genius, rather than on a conceptually justified derivation from some deeper principle, it was not intellectually satisfactory. This bothered Pauli (who referred to it as acrobatics), and even more Dirac himself (who was apparently hoping all his life that something more conceptual will be found). Now, once you have Dirac's matrices and you understand their relationship to the Minkowski metric, it is evident that they belong to the branch of mathematics known as Clifford algebras. An then, unavoidably, this algebra becomes the foundation of a new branch of research with aspirations in physics.
(3) Quarks. I'll start again with a rhetorical question: How many physicists were familiar with the root systems of simple Lie algebras before Gell-Mann put them on the map? My guess hovers around zero (on the positive side, though). But this number jumped up following Gell-Mann's impressive discovery. We now have a lively branch of research aimed at discovering one-to-one mappings between the elementary particles of multiplets and the roots of simple Lie algebras.
Let's now look at these three examples from some distance: Einstein, Dirac and Gell-Mann discovered some profound new paradigms, and their results motivated (and still motivate) a small percentage physicists to search for new paradigms in their wake. If I am not mistaken, this approach cannot boast much success so far (the success I mean has nothing to do with publication counts).
You say "It is not yet clear to me why we should use quantions and not complex quaternions or the Pauli algebra, which seems to contain both the Lorentz metric and the electroweak group.". I' give you two answers.
A general answer: Taking the lessons of history seriously, I could not possible search for the paradigm I am interested in (I mean the idea that will eventually lead to a satisactory merging of relativity and quantum mechanics) in the wake of far-reaching discoveries made by others in the past. I don't think the paradigm in question will be found there, but I do think that the field is crowded enough without me.
A specific answer: First, I am not saying or implying that you should use quantions and that you should not use complex quaternions. By all means, use the latter if you think you'll make a breakthough where many others have not (reminder: breakthrough = unification; if you mean anything else, I'll have nothing to say because, like kiltmaking or fly-fishing, it would be outside my sphere of interests). But there is more: Your question is stated upside down. My objective was not to re-derive Dirac's equation or bust -- and then quantions just happened to do the job. The quantions came from elsewhere as a new number system that seems to be just right to support the unification I am after. But then, my friend Nikola Zovko correctly pointed out that I should take a break from further developing the mathematics of quantions and verify as soon as possible that they are not spuriously relativistic and that they are consistent with quantum mechanics. For the latter, he suggested the verification of Schroedinger's equation, as well as the approach I call "Zovko's interpretation". This worked out beautifully. Both Schroedinger's and Dirac's equations, together with potentials, follow as theorems from the quantionic number system augmented only with Zovko's interpretation.
You mention in your post other things I would enjoy discussing, but not util (if you feel like it) you reformulate them in the spirit of my work, which I just described. I need this to be sure we are on the same wavelength.
But if you need additional clarifications, please ask.
Best regards,
Emile.