In 1977 a re-editing of Ulysses was begun. It appeared to conclude in 1984, when the results were published, together with a further copyright claim, reinforced in 1986 by the appearance of a commerical or trade edition of the work, now sold worldwide. Academically, this achievement can be debated, argued over, criticised. But as a legal claim to a new copyright it can only be challenged when the general James Joyce copyright runs out, next January. [xiv]
It was absurd to imagine that he could deal objectively with the existing third typescript, in terms of making it identical with the earlier two copies, as it is to imagine that Turner could arrive at a varnishing day at the Royal Academy and not bring out his paints and brushes and do more work on his allegedly finished canvasses. 
The fact is that James Joyce was faced with a fair and positive offer from Sylvia Beach in 1921, having seeen that his book would not succeed elsewhere, and he had to address the unusual probllem for a renowned writeras he was then becomingof having a publisher but of not having the book, at least in corrected form and complete. 
Maurice Darantière 
McAlmon muddled some of the text and was surprised when Joyce retained his changes. 
He added no less than 34% of additional material to Ithaca after his announcement about completing the episode and therefore the book.
one typescript page turned into two or three replacement typescript pages which were then inserted in the original sequence. 
.. he expressed extreme irritation over the printers errors and wondered whether these would be perpetuated in future editions. But he never complained about the overall text of the book.
there is no sense in which he viewed what he was currently doing
as other than the finalised version of Ulysses which he wanted to see in print.
[a] view greatly reinforced by the changes which he made on the proofs. [23-4]
In all his correspondence at this time, Joyce referred frequently to the errors in Ulysses and was clearly frustrated by their number. Yet there is never any question of there being deeper textual problems. 
Summary of early publishing history of Ulysses 
John Quinns sale of the Ulysses MS [final typescript], puchased by A.S.W. Rosenbach, a dealer  Sylvia Beachs proofs went Marian Willard [Johnson] to Yale University. 
It is not clear whether Harriet Shaw Weaver made over the literary Executorship to the Society of Authors  Peter du Santoy, principle Trustee .
In deference to Noras wishes, HSW consigned the manuscripts of FW to the BL, and Ireland remains without major Joyce material of any kind. 
The 1934 copyright claim was therefore made for the unpublished balance of Oxen of the Sun, together with the final four episodes.The differences between the parts of Ulysses which were actually copyrighted and teh version of Ulysses as printed in 1934 are more substantial than the differences between the accepted, pre-Gabler version of the book and his revised version. 
The Coming of the Scholars, Chap. 5: Failings of Richard Ellmann, pointed out by Ulick OConnor .
Correction list to Ulysses made by Chester A. Anderson . Scholes corrects Marias waterproof with raincloak, which Joyce had erased in the 1914 edn.
Enter John Kidd.  Rossmann, Ellmann letters, 1988. 
Gabler claims that 1922 comes closest to what Joyce intended, but that it does not offer Ulysses as he wrote it; no analysis of this statement whatsoever. In Afterword, 1986.  continuous manuscript text  By common consent, an editor chooses as the copytext for a critical edition a document text of highest overall authority. This eliminates the first edition of 1922 as copytext for a critical edition of Ulysses. (Gabler) Arnold: How would Gabler deal with Plato? 
Kenner, The Computerised Ulysses, Harpers April 1980. Leopold Bloom Restored, review of 1984 Ulyss, in Times Lit. suppl., (1984).
On the word know to all men  Arnold: It would be a major alteration indeed if the haunting subtlety of Stephen not knowing the word known to all men and searching for it, not as a word but as a reality in his life, were to be known and expressed by him long before he confronts his mother. 
Ellmanns preface appraises Gablers edition as the search for an ideal text, such as Joyce would have constructed in ideal conditions and an attempt to deduce from other versions what the lost documents would have contained.  Paper at 1986 Monaco conference: A Crux in the New Edition, where Ellmann reports on Gablers inclusion of the word known to all men in Scylla: he thought I would be pleased. Ellmann was far from pleased [...] put his doubts to Gabler, but ot no avail. Put it among the prominent variants at the back. .