David W. Fenton quoted part of what I wrote, :
One of the questions to be dealt with, too, is "how much unicode support
does there need to be?"
And asserted
This question demonstrates that you understand neither Unicode nor
the fact that Finale's file format is a database.
There is no halfway. Unicode increases the potential data storage for
each character from 1 byte to 2 bytes. This is not something you can
simply do in a few places and not in others. It would require a
rewrite of the entire database engine underlying Finale, along with
the relevant changes to the file format, and then a complete rewrite
of Finale itself.
Now, I admit that I know almost nothing of the internals of Finale and I
don't know much about the file structure of Finale. However,
I deny that I fail to understand that "Finale's file format is a
database". In fact, I know that since a database is "a structured
collection of records or data that is stored in a computer", /every/
data file used by any software package is a database. Further, I deny
the unsubstantiated claim that there "there is no halfway". Indeed, the
very notion that every part of Finale must support unicode if any part
of it does violates a precept which I have understood to be a
fundamental postulate in Software engineering since at least the mid
1970's: that software should be engineered as sub-units which can
substantially operate independent of one another, in order to reduce
maintenance, and to promote re-usability. Under this precept it would
indeed be possible for one part of Finale--for example, the lyric and
text-block subsytem--to support unicode compliance, while other
parts--for example, the part that deals with the producing the images of
standard music notation--are not. Furthermore, even if the data is in a
structured database, I am not persuaded that one part of the structure
cannot be processed according to one criterion, in this case, "unicode
compliant", while other parts of the strucutre are processed according
to a different criterion, e.g., "unicode non-compliant".
I'm not sure I understand the basis for the assertion that with respect
to unicode implementation in Finale "there is no halfway". The notion of
putting significant parts of a major system like Finale into
independent, sub-units, in order to simplify maintenance, and promote
re-usability an important precept in the science of Software Engineering
since at least the mid 1970's, which as far as I can tell, predates the
earliest alpha versions of Finale. On that basis alone, it is not
inconceivable that the lyrics and text subsytem in Finale may well have
been written with sufficient independence of the rest of the program
that unicode could be implemented there without being implemented in the
remainder of the software.
Now, I do concede that it's possible that the assertions that there is
no way to implement unicode support in the lyrics and text block
subsytem of Finale are correct, and that elements that I perceive as
supporting the idea that partial support unicode (in lyrics and text
blocks) are possible are either elaborate illusions constructed to
confuse the folks at S~, or are simply figments of my own imagination,
but in the absence of a bit more concrete proof that this is the case,
I'm not yet persuaded.
ns
_______________________________________________
Finale mailing list
Finale@shsu.edu
http://lists.shsu.edu/mailman/listinfo/finale
