Re: Is Sibelius really as bad as this?

[Urs Liska]

Am 29.05.2015 um 10:13 schrieb Richard Shann:

On Thu, 2015-05-28 at 21:16 +0200, Jacques Menu wrote:

Hello Richard,

Find attached what I get after raw import of the XML file into Sibelius 7.1.3 and then export as PDF.

It seems there’s no difference with what you got from IMSLP.

JM

Well, again I couldn't view this in Evince but I could open it with
Iceweasel, and it shows something interesting: in bar 13 the original
has a cautionary accidental in parentheses. 

Denemo's MusicXML import ignores this field (yes! I've submitted a bug
report for this) so I have inserted it manually, getting the attached
typeset LilyPondBar13.png.

The hand-written Sibelius output was particularly bad for this (see
SibeliusHandGenerated.ly), while Sibelius's MusicXML import, like
Denemo's, ignored the cautionary attribute when re-importing its own
MusicXML (see SibeliusImportedFromMusicXML.png attached - this has been
snipped from your file).

Reading this mailing list gave me the impression that Sibelius was a
required format for some publishing houses. 

This is correct. From my own experience and comments by others most (major) publishers require you to submit one of the following:;
- Finale files
- Sibelius files
- SCORE files
- ((((PDF))))

How can this be? 

Good question.
Has to be put also the other way round: How can it be that practically noone accepts LilyPond yet? It can't be the text approach alone, otherwise they wouldn't use SCORE (and sometimes even Amadeus which is very similar to LilyPond in a way).

Do people
go through every dotted rhythm adjusting the positions of the dots by
hand? I just the really terrible ones? Every tie?

I think they do, and they don't bother because it's so easy to do.
However, they don't consider enough (IMO) how unstable these tweaks are with regard to any future changes. And the impact of version control.

When we launched our "Engraving challenges" (https://github.com/engraving-challenges/main) project (which has unfortunately remained unfinished) it became clear that it's a natural reflex for any Finale/Sibelius user to immediately touch their music and "pull it out of the way" to make it usable/readable - the task to *only* enter the plain music seems to be really challenging. And as these tweaks are opaquely stuffed away in the binary file the scores are compromised right from the start (IMO).

Urs

Richard

Le 28 mai 2015 à 20:58, Richard Shann <address@hidden> a écrit :

On Thu, 2015-05-28 at 20:22 +0200, Johan Vromans wrote:

On Thu, 28 May 2015 19:06:08 +0100
Richard Shann <address@hidden> wrote:

Is Sibelius really this bad (ties treated as slurs, 

I frequently rewrite Sibelius scores into LilyPond, and this is one of the
issues that cost a lot of manual preprocessing. Moreover, it makes the
composer 'lazy' -- it doesn't matter whether to create a tie or a slur, and
the command for tie is apparently very easy.

But the composer wasn't 'lazy' in this case - the MusicXML shows that
ties were entered for ties, not slurs, yet they are typeset as if they
were slurs. (That is, Sibelius exported to MusicXML making the
distinction between the slurs and the ties correctly, so they must have
been entered correctly,  but its typesetting is wrong).

(Oh, and the Sibelius-generated PDF does not render with Evince, but
that is another issue I presume).

Both attachments render fine in my Evince.

Ah, that'll be down to my version of Evince being too old then...

Richard



_______________________________________________
lilypond-user mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/lilypond-user

_______________________________________________
lilypond-user mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/lilypond-user