Re: GG Re: Database details -- general

[frank@clark.net]

On Mon, 13 Jan 1997, Kris Shapar wrote:

> What happens when you save the file as a generic file in Word
> Perfect? That should be a straight ASCII file and should not add the
> extra lines you talk about.

I am not expert enough to say for sure. But what I think is going on is
this:

1. The first few characters tells the operating system and other software
programs that there something is dbase format to follow.

2. Then come strings of characters naming the fields and setting their
lengths.

3. Then come the records. They all have to be exactly the same length and
each field within a records has to have exactly the same number of
characters, which will be what you input and then a bunch of spaces to
make up the rest.

I suspect that if you typed up your records in WordPerfect, like this

Field 1      Field 2    Field 3   Field 4    Field 5 Field 6Field 7
Composer     Work       RecDate   Label      Prefix  CatNo  RelDate

Bach         Goldberg   550606    Columbia   ML      5060   560301   

with running out the number of spaces to the next field (the fields may be
of different length, or they may always be dbase's maximum; I don't know),
and then *append* the WordPerfect file, you'll have a dbase data set.

Now the nice thing is that dbase-based programs are all over the place.
You can even get shareware/buyware already structured by record
collectors. I'd do a Yahoo search. In any case, it's easy to add fields
later, as you decide that you want more of them. (I presume the software
will adjust the record lengths.) AND, what is more, you can like data sets
together. So if someone already has an data set of classical music with
many recordings on it, you can just yank up a record from that set, change
it so it displays the data for your own recording, and add it to your data
set. I think you should be able to move fields around in the yanked data
set.

I don't know what is already out there in the way of data sets. I do know
that there is no universal standard on how to identify a piece of music.
The Library of Congress has its own scheme, _Schwann Opus_ another, and so
on. Each composer gets treated differently, as is fitting. Quick: what are
the Cantatas of Beethoven and Brahms? (Answers: Der glorreiche Augenblick,
Op. 136 plus ones about Joe II and Leopold II, WoO 87/88, in the case of
Beethoven and Rinaldo, Op. 50 of Brahms). Quick: Why is the Ascension
Oratorio, "Lobet Gott in seinen Reichen," called Cantata No. 11?

Do you alphabetize by the title in English, the title in the language the
composer *usually* spoke, the title on the first edition (often French for
Beethoven)?

I would start with solo instrumental, chamber music, orchestral, and then
songs and arias by no. of instruments accompanying, then singers with
orchestra, then choruses by no. of instruments, then singer and choruses.
I'd place sacred before secular. This puts opera at the rear end,
*exactly* where it deserves to go!

> BTW, the characters after 120 are also known as the set of extended
> ASCII characters and contain accented characters such as ü (Alt-129).

I just called up www.yahoo.com and scrolled down to computers, had the
whole tree displayed, and looked at the five ASCII sites. I quickly found
a site giving me a whole bunch of *different* ASCII extensions!

Frank