[Aiwendil]

I agree in principle with this, but in 2b, do we drop the later ideas and keep the original ones or do we just not mention the original one.
For example the Gil-Galad thing. Would the change be to make him of the House of Finarfin as Tolkien intended or just to leave his lineage uncertain.

I would say that the Gil-Galad change does not fall under 2b. True, it was not actually incorporated into the LQ or GA texts, but it is not a change that requires major revision. The only emendation necessary, as far as I know, is the alteration of a few names at any relevant points. Minor changes like this are not the problem; the problem is with changes "that do not clearly indicate the exact details that must be changed and how they are to be changed." Thus, for example, the round earth cosmology is not to be used, since it is merely a projected change and we would be more or less lost (or, even worse, at liberty) in deciding precisely what textual emendations to make for such a change.

As for the aesthetic issue: I believe that there are two related problems that we are to some extent confounding. There is on the one hand the issue of textual style and readability, and on the other, the matter of such names as Rog. The former are merely features of the text, not features of the story itself. The latter are actual elements of the mythology. There could be a thousand different texts, each written in a slightly different way, that describe the same literal events.

First, the matter of textual style:
Lindil said:

If one only wants to create a canon of 'idea's and events'[like the number of balrog's and wether the dragon's were still mechanical or not] then readability is irrelavant. But if we want a work that can not only stand next to the published Silm but literally replace it in scope quality and accuracy, then we will have to go the extra mile and work whatever literary polish into the archaic language we can.

Yes, it all comes down to this again. I have, from my first participation in the project, favored a 'canon of ideas and events', at least as our first objective. I agree that without some significant stylistic revision, there will be disconcerting changes in style and a certain unevenness about the whole thing. But I prefer to focus first on establishing a canon of ideas and events (that is, establishing the literal 'canon', or truth) and not worrying about the actual style of the documents that present that canon (the literary truth, if you will).

These are all things that CRT had to do to give us a Silm that is as excellent as it is.

Yes, but CRT was shooting not only for a readable Silmarillion, but also for a veritable Quenta Silmarillion. That is, he was trying to create the actual document called "Quenta Silmarillion", a document that exists within the sub-reality of Middle-earth. I have said it before: we cannot create such a document, nor I think can we create a document of much literary worth at all, with this project as it is (and as it has been since the beginning). One of our main objectives is the inclusion of such things as the longer "Fall of Gondolin", or the Narn, things that, by definition, would not be in the Quenta Silmarillion. And above all, there is the problem of proportion. As things are, there are very detailed accounts of Beren and Luthien, the Fifth Battle, Turin, and the Wanderings of Hurin. Then we come to, as CRT says, a cliff. For the rest of the "Ruin of Doriath", there is only a summary account. We then have a very full Tuor, followed by a maddeningly short and compressed Earendil.

It is my opinion that we should retain #7 as it is at present. I think we should deal with the literal canon first, and that stylistic issues should be left out of that as far as possible. It is my hope that this work may be the starting point for a new Quenta Silmarillion, or for a longer, stylistically consistent, one. But at the very least, I think, we can leave stylistic concerns until the end.

Now, in my opinion such things as "Rog" and the mechanical dragons are completely different in fundamental nature. The question of Rog is not a stylistic one. It is a literal question: it is the question of what the character's name is. Nor do I think that principle 7, as it currently stands, necessitates the retention of the name. It states that:

A corollary is that we may not disregard anything written by JRRT unless it is invalidated by one of the above principles; i.e., we must have a REASON for rejecting something.

So, if we have a reason for rejecting "Rog", we can reject it. The principle would forbid us from dropping the name with no reason, or merely because we didn't like it. But if it can be shown that the name would have been changed, then by all means, we should change it.

Lindil said:

Also imo we have already applied this principle of revision by aesthetics or 'feel' by second guessing the mechanical monsters as JRRT never 'specifically' ruled them out. The majority agreed it did not 'feel' right, and that JRRT probably would have abandoned it.

But we did have a justification for excluding them - the fact that, in all subsequent references to the fall of Gondolin, the things are merely called "dragons", not "mechanical dragons". This amounts to a change that can be upheld under principle 2. One could even argue, without reference to the Fall of Gondolin, that the absence of any future references to mechanical dragons is sufficient evidence for their exclusion.

So, in short: I think that the real stylistic issues should wait, and I think that the other issues mentioned are in fact not really stylistic, but canonical.