Indexing (Disney comics Database)

[H.W. Fluks]

KNUT:

> I've been looking at the Disney comics database at the web-server a little
> According to the description there should be a zip-file somewhere

Yes, I put a line in the file "legend" about a subdirectory "..index/zip",
but I didn't tell Per about it, and Per didn't read the "legend" file.
So the zip-files aren't there. Per, would you put them there, or send
Knut a copy of sweden.zip, stories.zip, src.zip and root.zip?

> how would I find out what story code a story has in the magazines
> that where printed before the story codes where in the first panel?

One way that hasn't been mentioned is to look at ALL the stories
available. Part of the Database are files with stories, ordered by
story code. When you know the page count, artist, description and
maybe country of origin, you could do a 'grep' or 'search' on the
"stories" files, and see which stories fit. This is what I have been
doing a lot, trying to find the story codes of the Dutch reprints
from the 50s and 60s.

> Dreams, dreams...

Yes, I have been dreaming things like that, too. Maybe some of them
become a reality sooner than we think...

PER said:
> No, there are no Norwegian references as of now.

In fact, there is *one* Norwegian title in the Database ("No:Maretitt").
I happen to have the Norwegian issue that contains that story. 8-)

> Well, I'm interested in having indexes of *any* Disney comics.
> And I think index coordinator Harry mostly agrees with that.

I am willing to include any Disney comics index in the database.
Personally, I once drew a line at stories I found absolute garbage
(like the Italian stuff we get in pocket books, or the Al Hubbard
Fethry stories). But then, there is always someone else that likes
those stories, and most of the time I can be convinced that some of
the "garbage" is actually quite good.
So as long as it's Disney, and as long as it's comics, it is welcome
in the Database.

JO/RGEN said:
> It shouldn't come as a surprise that I've been working on a Norwegian
> index for a while, but so far only in my own format. The main reason
> for this is that I don't like lines of more than 80 characters.
> It should be quite easy, though, to convert it to Harry's format.

Well, it's not that easy if you want to do it right. The layout (which
field in which column) is the *easiest* part. The difficulty is the
story codes. Each story that has a code is listed in the Database
in the story files. When a story is reprinted, the story code should
be exactly like the one in the stories files, so the database programs
can make the connection. For instance, your example entry:

>  1967-20    b WDC313        4              Mim Superselgersken

This story is known in the database under the code "W WDC 313-?1".
The database programs can add extra interesting information to this
entry, like the artist (TS), original title (in this case not known),
and which characters appear in the story (Goofy, the Big Wolf, Scrooge).

No program can transfer the code WDC313 to W WDC 313-?1, because there
is also a W WDC 313-?2 and so on. This would be a lot of manual work
(of course with a lot of help from the computer).

BTW: if the story code is not "correct", this doesn't mean the entry
can't be listed in the database. It just gets a '@' to indicate that
the story is of unknown origin, and you'll miss all kinds of cross
references and additional information.

> The main problem with this format is that
> there are only room for three creators, and not four as in Harry's.

This is only a minor problem.
We only have to write a little program to convert your format into
mine (and vice versa). You don't like lines longer than 80 characters?
OK, write a program that produces short lines. The biggest problem here
is to define a readable format, that contains all the information we
want. Spreading things over more lines doesn't make it more readable,
at least that's what I think.

> A couple of differences from Harry's format is that I have the letters
> that show the "location" of the story in an own column

This is a good idea; I plan to do this in the Database as well (some day).

KNUT again:
> But I could help you with a conversion method
> from your format to Harry's, if you like (I have studied computing science
> and work as a programmer).

I prefer a conversion to the "internal" database format. I have C routines
to write an internal struct to that format, so only routines have to be
made to read the Bangor format.

JO/RGEN:
> Harry sent us (the indexers) a file which explained [the formats] a while
> ago.

This file is also in the ftp directory. It's called "format.legend".
I'll send you (Knut) a copy in private mail.

Of course, as always: any remarks, critique, improvements, are very welcome!

--Harry.