naturerecordists
[Top] [All Lists]

Re: Files posted to group web site

Subject: Re: Files posted to group web site
From: "Rich Peet" <>
Date: Mon, 29 Mar 2004 16:51:42 -0000
The system you use works for you and your collection.
We are not whining, we are improving. Metadata is the present and the
future for portability of sound records. It is the reason that BWF is
taking off.

If we are going to go beyond being a series of island collections,
then each sound file most stand as a historical record on its own.
I am building a library. Most visitors will want to read a chapter
from a book or maybe just a page.  It will need to be accessible
across multiple platforms, systems, and locations, not just the
collection but each page.


As I record more I am capturing things I didn't think I would. Such
as the Snowplow recording, or the sound of bubbles within a Tropical
Fish Store, or I90 at rush hour.  I am only one judge of what is
important to record and in 50 years I suspect the values will have
changed in the Naturerecordists favor. I need to treat my recordings
as a small library if any of it will survive beyond me.

A data base is fine. I use one. I just type myself an ongoing
document within a word processor for my database. Search engines are
good enough now that I can rely on them to recover specific
references from a field diary. Flat file data bases will not be the
answer for me.  I have fixed a few corrupted ones with 50,000 plus
records.

Linked is now an old document but worth reading. Developement of
archival systems are moving slow, which they should.
http://www.kfs.oeaw.ac.at/harm/D3_6_3.htm

Rich Peet


--- In  Walter Knapp <>
wrote:
> From: Dan Dugan <>
>
> > I'd prefer discussion here on the list, I think this is of
interest
> > to everyone. I use FWB Catalog ToolKit 5 to catalog files that I
> > archive to CD-ROM. It's fast, but the searching ignores the
volume
> > name, which is a problem because that may contain keywords that
the
> > file names don't, and it doesn't have any means for auditioning
or
> > audio thumbnails. It's advertised for cataloging multi-media
files
> > but it's really designed for images.
> >
> > I hate the idea of committing all the effort to building a
database
> > in some software that may go away. I just checked
http://www.fwb.com,
> > and it's no longer listed as a product, so my misgivings are
> > validated. I would much prefer a standardized metadata system
that
> > could be accessed by different software.
>
> metadata will only be reliable in terms of being supported if it's
used.
> As it stands with audio metadata it's used by a microscopic
proportion
> of users and the proportion is dropping. That's a formula for
having it
> go away. At minimum I'd not have metadata as your only storage of
the
> info. Regular databases are much more long lived, and if replaced
the
> new database software nearly always will read the old formats.
(unless
> microsoft wrote it, then all bets are off) Oddball little slide
show or
> audio player software is usually not readable by anything else
ever.
> Stick with a widely used database format.
>
> I use a separate database keyed to a coded filename. I could
connect the
> two so the file would play, but that assumes I'm going to keep all
that
> bulk organized on a spinning hard disk. So, I play the files
separately
> and find them in a separate database containing all the info. Often
I
> want a sonogram to look at, so play them in my regular sound
editors,
> it's not inconvenient if you work out a naming/organizing system
for the
> files that's logical.
>
> As some may have noticed my filenames start with the year, then the
disk
> number, then the track number. A folder labeled with year and disk
> number contains files from each disk. Those are organized in
sequence in
> a master folder on my hard disk working copy and in my backups.
>
> My field data is initially recorded on field cards, and there is a
> corresponding database in Filemaker format. Filemaker being
something
> designed as a database is very, very good at organizing, sorting or
> finding things in the database. It also can sync the database with
a cut
> down version containing just certain fields that I keep in my Palm
run
> by Filemaker mobile. So I can check the database while out in the
field
> without having to drag a laptop along. I can place a limited number
of
> photos or sounds on the Palm as well. Actually a large number as it
> takes SD memory cards. Filemaker is available across several OS's,
has
> been around for a very long time and is very reliable. It can
handle
> playing soundfiles, though I've not looked into doing it. Might
have to
> use it's scripting ability. Nothing to putting in a site photo if
you
> have them. The same database can even be integrated into webpages
> without going outside filemaker.
>
> I don't understand the whining about having to set up a database,
it's
> very easy in Filemaker, and you can revise it without losing any
data.
> I've added fields, changed display formats and look and all that
with no
> problems. You can, in fact have many different display formats and
> switch between within one database. You will have to set up a
database
> unless you wish to open each file to read it's metadata when
looking for
> something. Let's see, open a thousand sound files to find
something,
> sounds to me like a job for a separate database. Go with a standard
well
> supported database for all that critical data. If you want to do
the
> extra work of sticking the same data in the metadata, that's fine,
but
> it should not be your primary data storage.
>
> As far as I'm concerned it doubles the work to use metadata in a
way
> that will work. And if you don't maintain that separate database
it's a
> huge risk. The separate database is also a lot smaller and more
portable
> than some system containing lots of image and audio files.
>
> Walt
> 



________________________________________________________________________
________________________________________________________________________

<Prev in Thread] Current Thread [Next in Thread>
Admin

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the naturerecordists mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU