Organizing: back to the digital realms...
Sep. 16th, 2009 04:28 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
pseydtonneI've mentioned before that organizing digital files can be even more of a time drain than organizing paper files. For example: a stack of papers in a box usually requires an hour of time, contains a fixed thickness of material and items each of a fixed thickness. You can't suddenly find ninety feet of bound volumes in one banker's box.
Open the main directory on a partition of six hundred gigabytes and you could find anything. That numerical limit could unveil a few movies, or thousands of music albums, or millions of photographs, or billions of text files in no order, or a fascinating combination of the above. My biggest networked share at home falls under that last category.
When I want to sort digital files, I often have to accept much weirder possibilities than I would have with regular papers. Let me describe how I had to sort just one directory yesterday to give you some perspective.
This directory had once belonged to a previous Linux machine. I just copied all of the user files to its replacement machine, labelled the new directory "xubily" and stopped there. I can only vaguely recall why the directory has that name (it must've been a computer running Xubuntu Linux and I came up with a mnemonic based on the 1980s kids' show "Zoobily Zoo").
Okay, so what did I find? Some preference files, only two of which did I need. I find entire subdirectories with single items, none of which were worth keeping. Each of these moments of looking in a folder, seeing a piece of garbage and tossing it took time. When more and more directories had picture files, I had to shuffle files from the command line and create new directories -- just to delete the directories moments later.
Eventually I found one file of 12 megabyte with a catchy name but only a distantly familiar suffix. It had been encoded in an old Mac format (.hqx), which makes the file even bigger but often uses an unzipping tool to open it. I found that I didn't have such a tool installed, so I had to research the tool's name via the suffix, install the tool on the Linux server, get yelled at a few times for mistyping the name of the command to find its manual page, and then run a simple command.
Suddenly I find out I have an MP3 I'd never heard before. I was able to play the song and I liked it but I only had the song title (which was the file name). I looked up the title online to learn the name of the artist. This meant I also needed to edit the ID3 tag on the file to note the artist, album and year. Then I could delete redundant copies of the file and drop it into a folder for future processing.
While I was at it, I grabbed all the files from that last folder and started labelling things. This folder has a bunch of MP3s that may be redundant, have filenames that mean nothing but full ID3 details inside or some combination of anarchy. With those objects sorted, I felt like I'd made a lot better use of the last couple hours.
Thus my biggest problem with digital sorting is that a single, smallish object may require far more attention than an entire giant directory. My only reward is that I've freed a bunch of space when I'm done, that perhaps I've figured out what to do with future items of the weird one's ilk.
Then I'm stuck realizing that I have a pile of files sorted by time and almost nothing else. If I want to make a real organization of these items, I have to decide the order of significance before I can make any progress. All picture files can go into one folder, but how many subdirectories should that have and how significant much the titles be?
That's usually when I stop myself from overthinking and just type up what I've done for the day. That's what you're reading.
-I feel better now, Ps/d
Open the main directory on a partition of six hundred gigabytes and you could find anything. That numerical limit could unveil a few movies, or thousands of music albums, or millions of photographs, or billions of text files in no order, or a fascinating combination of the above. My biggest networked share at home falls under that last category.
When I want to sort digital files, I often have to accept much weirder possibilities than I would have with regular papers. Let me describe how I had to sort just one directory yesterday to give you some perspective.
This directory had once belonged to a previous Linux machine. I just copied all of the user files to its replacement machine, labelled the new directory "xubily" and stopped there. I can only vaguely recall why the directory has that name (it must've been a computer running Xubuntu Linux and I came up with a mnemonic based on the 1980s kids' show "Zoobily Zoo").
Okay, so what did I find? Some preference files, only two of which did I need. I find entire subdirectories with single items, none of which were worth keeping. Each of these moments of looking in a folder, seeing a piece of garbage and tossing it took time. When more and more directories had picture files, I had to shuffle files from the command line and create new directories -- just to delete the directories moments later.
Eventually I found one file of 12 megabyte with a catchy name but only a distantly familiar suffix. It had been encoded in an old Mac format (.hqx), which makes the file even bigger but often uses an unzipping tool to open it. I found that I didn't have such a tool installed, so I had to research the tool's name via the suffix, install the tool on the Linux server, get yelled at a few times for mistyping the name of the command to find its manual page, and then run a simple command.
Suddenly I find out I have an MP3 I'd never heard before. I was able to play the song and I liked it but I only had the song title (which was the file name). I looked up the title online to learn the name of the artist. This meant I also needed to edit the ID3 tag on the file to note the artist, album and year. Then I could delete redundant copies of the file and drop it into a folder for future processing.
While I was at it, I grabbed all the files from that last folder and started labelling things. This folder has a bunch of MP3s that may be redundant, have filenames that mean nothing but full ID3 details inside or some combination of anarchy. With those objects sorted, I felt like I'd made a lot better use of the last couple hours.
Thus my biggest problem with digital sorting is that a single, smallish object may require far more attention than an entire giant directory. My only reward is that I've freed a bunch of space when I'm done, that perhaps I've figured out what to do with future items of the weird one's ilk.
Then I'm stuck realizing that I have a pile of files sorted by time and almost nothing else. If I want to make a real organization of these items, I have to decide the order of significance before I can make any progress. All picture files can go into one folder, but how many subdirectories should that have and how significant much the titles be?
That's usually when I stop myself from overthinking and just type up what I've done for the day. That's what you're reading.
-I feel better now, Ps/d
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
Organizing: back to the digital realms...
Date: 2009-09-16 12:35 pm (UTC)no subject
Date: 2009-09-16 04:11 pm (UTC)I think the one I wound up using the most was a Java applet, MP3 Tagger (http://ostermiller.org/mp3tagger/). It was good at guessing a name based on subdirectories and letting me change those parameters. It also tied into the Windows Context Menu, so I could right-click on an MP3 inside the file explorer and change attributes.
I know I used another one as well, since part of the trick was having more than one approach. I wish I'd kept better notes about what I did then, but I hope that helps.
no subject
Date: 2009-09-17 03:51 am (UTC)what’s wrong with MP3::Tag or MP3::Info? mix those up with a little File::Find sauce and i think you're set :)
-steve
no subject
Date: 2009-09-17 08:22 am (UTC)