Select: 
Category
All
Graphic ambiguousness
Interpretations within context
Differences between sources
Editorial revisions
Corrections & alterations
Source & stylistic information
Notation
All
Pitch
Rhythm
Slurs
Articulation, Accents, Hairpins
Verbal indications
Pedalling
Fingering
Ornaments
Shorthand & other
Importance
All
Important
Main


b. 19-21

composition: Op. 28 No. 17, Prelude in A♭ major

..

In A (→FCGE) the topmost notes of the L.H. part (d1 and higher) are written on the top stave, while in b. 19 the R.H. g notes are placed on the bottom stave:

 (the accidentals and the order of the overlapping quavers in the right and left hands are presented according to A; in FC the  before E was omitted, and GE supple­mented accidentals and introduced some changes in the horizontal arrangement of chords). In FE (→EE) this seemingly illegible notation was simplified by placing the part of each of the hands on a separate stave, which we also adopted to the main text. According to us, the complexity of the original notation resulted in the double Terzverschreibung mistake in b. 21 in GE.

category imprint: Editorial revisions; Source & stylistic information

issues: FE revisions

b. 19-23

composition: Op. 28 No. 17, Prelude in A♭ major

cresc. - - - from bar 20 in FCI

cre - scen - do from bar 19 in A (→FCGE, →FEEE)

..

If FCI accurately represents the notation of [AI], the stimulus to begin the crescendo was initially the five-quaver ascending L.H. sequence in b. 20. According to us, this version may be considered an acceptable variant; on the basis thereof, one can also shape accordingly the course of the crescendo starting – in accordance with the main text – in b. 19.
We believe that the placement of those indications – over the top stave (in A) or under the bottom one (in FCI) – results from lack of space between the staves; in the case of A the reason were also the crossed-out  and  hairpins, which were initially written there (see the next note).  

category imprint: Differences between sources; Corrections & alterations; Source & stylistic information

b. 19-21

composition: Op. 28 No. 17, Prelude in A♭ major

g1 & c2 tied in FCI

g1 & c2 repeated in A (→FE,FCGE)

g1 (quaver) & c2 repeated in EE

..

In the version prepared for print, hence the latest one, Chopin abandoned the ties to g1 in b. 19 and c2 in b. 20-21, since nothing suggests that respective ties could have been overlooked.
At the beginning of b. 19, FE omitted the rest in the lower R.H. voice, and the dot prolonging the crotchet g1. The EE version of the first half of this bar could be an effort to correct this unclear notation. Interestingly, traces of a deleted beam encompassing the first half of b. 19 are also visible in FC.

category imprint: Interpretations within context; Differences between sources; Corrections & alterations

issues: EE revisions , Errors in FE

b. 19-23

composition: Op. 28 No. 17, Prelude in A♭ major

..

In A one can see in these bars crossings-out of dynamic hairpins –  in b. 19, 20, 2nd half of b. 21 and in b. 22 and   in the 1st half of b. 21 and in b. 23. See also b. 24-25.

category imprint: Corrections & alterations; Source & stylistic information

issues: Corrections in A , Deletions in A

b. 19

composition: Op. 28 No. 17, Prelude in A♭ major

f1 in FCI

f1 in A (→FCGE, →FEEE)

..

In the version prepared for print, the top note of the chord on the 5th quaver in the bar, interpreted literally, is an f1, since it is not accompanied by a relevant accidental in this bar. The following arguments prove that it was almost certainly Chopin's mistake:

  •  put before that note in FCI, confirming the authenticity of the version with f1,
  • presence of e1 (played with the L.H.) in that chord,  which, together with f1, would result in an unpleasant cluster – cf. b. 21, in which Chopin avoided a similar cluster by omitting g1.

category imprint: Interpretations within context; Differences between sources; Source & stylistic information

issues: Omission of current key accidentals , Errors of A , Deletions in A