August 24, 2005

I put the CD in OCD

Posted by Arcane Gazebo at August 24, 2005 3:46 PM

I remember being shocked back in May when Amanda at Pandagon said, "A mix CD takes like half an hour, tops, which means that you can pretty much arrange it, drop the disc in and by the time you've finished making your sandwich, it's done." My procedure for mix CDs takes about a week and goes like this:

  1. Open iTunes and start a new playlist dedicated to the mix CD. Create a text file with the same title.
  2. Go through my master playlist looking for songs that fit the theme and intended recipient of the CD, which are added to the new playlist.
  3. Switch to the new playlist. Select the leadoff song, and start arranging the subsequent tracks based on what songs will go well together. Sometimes I'll choose the closing song right away too. There are probably more than 80 minutes worth of songs on the playlist, so I leave the extra ones at the bottom (in case I want to switch them in later) and make a note of where the 80-minute cutoff is.
  4. Note the current tracklist in the text file.
  5. Listen to the playlist all the way through to the 80-minute mark. While listening, make notes in the text file on the overall flow of the CD and the feel of each track in context. I usually mark songs with + or - signs based on whether I like their current position, along with more specific thoughts when appropriate.
  6. Revise the playlist based on my notes, both rearranging the order and replacing songs entirely. Usually the original ordering gets heavily revised, as there will be lots of song pairings that I thought would sound good but sound terrible in actuality.
  7. Repeat steps 4-6 until satisfied. This usually means three or four times. I tend to do one iteration per day so I don't get too sick of the songs.
  8. Burn a disc from iTunes and play the whole thing through on good speakers to get a sense of overall flow. Usually I have done the previous steps on headphones, so this allows me to listen to it in another context. I tend to be pretty happy with it at this stage. Almost done...
  9. Wait! While walking to work in the morning, my iPod shuffles up a song that has to be on the mix. I don't know how I overlooked it! Since the CD is already 79 minutes and 53 seconds, I have to pick a song to drop to fit the new one, and this introduces a perturbation into the (unstable!) tracklist equilibrium. Go through another iteration or two of arranging.
  10. With the final tracklist in hand, go to my music shelves and track down the original CD for each song. I buy most of my music on CD rather than through iTunes or otherwise, so I usually have the albums I need.
  11. Rip the tracks off the original albums in an uncompressed format.
  12. Burn a master copy of the mix from the CD-quality files.
  13. Listen to the master once to make sure no glitches were introduced in the rip/burn process.
  14. Make copies directly from the master for each recipient.
  15. Distribute copies appropriately, file master in personal archives.

Amanda's right that the mix CD is too convenient (compared to the mix tape), but in the sense that it's too convenient to tweak the mix endlessly. With a tape it's much more of a pain in the ass to go back and change things once you've done the recording. My advisor has the same complaint about how computers have changed the way scientific papers are written.

In fact, it just occurred to me that parts of this process are strikingly analogous to the process of editing a paper for submission to a peer-reviewed journal. Just substitute figures and equations for songs, and a limit of four pages instead of 80 minutes. Somehow the mix CD version is a lot more fun, though.

Tags: Music

You're not kidding about the OCD....

In terms of papers, I know that feeling well, although most of my stuff doesn't go to journals with strict maxima. (I should say, though, that the number of iterations I need is larger on average for the situations where I am submitting to such a journal.)

My record is in the 30s (not including page proofs), with nontrivial changes occuring between almost every pair of versions.

High single digits or really low teens (with small changes not causing numerical incrementation) are what usually occurs for my papers.

Posted by: Mason | August 24, 2005 5:31 PM

Make me a mix CD. I'm dead serious. I want to spend weeks listening to it on repeat trying to infer the subtle nuances of the relationships of the songs. Not only the songs on the CD, but the songs *not* on the CD.

Posted by: Lemming | August 24, 2005 5:35 PM

Mason: I think our record was in the low 80's, for a paper that had seven authors and was going to a journal with a four-page limit.

Lemming: Sure! You'll have to report on your analysis when you're done, of course. :)

Posted by: Arcane Gazebo | August 24, 2005 8:53 PM

Of course. I'll try to make sure it's as entertaining as my interpretation of Einstein on the Beach.

Posted by: Lemming | August 24, 2005 9:19 PM

May I never reach 80 iterations... the fact that I've never been on a paper with seven authors helps. (My max is 4 thus far, and 2 is my most common number.)

Posted by: Mason | August 24, 2005 11:11 PM

That paper I whined about? Seven authors, two of whom dislike each other intensely, and 40+ revisions (due to numerous computer issues, I don't have an accurate count, but it took 2 years), before submissions with a couple of really nasty rounds after submission. What an introduction to the fine world of publishing in science.

Posted by: Wren | August 25, 2005 8:13 AM

Back in the day, the purpose of publishing was to disseminate information rather than for career, grant, etc. purposes (and that final stamp of approval that helps with that other stuff). At least we have the arXiv to still keep the original purpose in mind. [I know several prominent scientists who would do away with journals entirely. One of my collaborators got out of a PRE editorship that way. They offered this to him, he stated the above opinion while agreeing to serve, and they withdrew the offer (which is what he wanted). He says what he thinks (consequences be damned), which I respect a great deal...]

Posted by: Mason | August 25, 2005 4:48 PM

PRE? Phys Rev E?

My advisor has been talking about using his NAS membership to circumvent the review process more often. Unfortunately, that still doesn't help with uncollaborative collaborators...

Posted by: Wren | August 25, 2005 9:46 PM

Yes, that's the right acronym.

For PNAS, many people seem to look at track I and track II articles _very_ differently. For him it won't matter, but it could matter for his junior coauthors. (Most likely, nobody will check, but the general vibe seems to be that taking advantage of the NAS membership [track 1, if I am remembering which is which correctly] is considered to be pretty wussy.) Also, this is _because_ of its use to circumvent the review process.

Posted by: Mason | August 25, 2005 9:56 PM
Post a comment