There’s really a stigma against failure, especially in America.  It is expected that you keep trying until you succeed, regardless of the consequences of doing so.  While my tale of defeat is nothing of consequence, with little to really be lost from non-success, it kind of makes me sad for people who are not given the opportunity to fail.  And further, to even classify the result as failure when it really should not be.

A week or so ago, I replicated a piece of artwork I have in my house, using my CD collection instead of the cassette tapes that were used in the art.  The picture of the CDs turned out pretty good, I thought, and I was inspired to grow it to a massive scale.  Where my original picture had maybe a couple hundred CDs featured in it, I wanted to scale it up to most of my collection, somewhere on the order of 1500+ CDs.

Over a series of nights, I spent my time placing the CDs in the pattern on the floor.  Keeping the pattern correct and also trying to make sure the CDs varied enough in their grouping was a little arduous.  But, I did persist and came up with a very large, organized mess of CDs.  Then the challenge became how to capture it.

I have a fair collection of photography equipment and so I was able to do some experimentation.  Experimentation was all I could do because I really had no idea how to accomplish the task.  The first attempt was to capture as much as possible in one picture.  I held the camera above the arrangement using a tripod and the self-timed shutter.  This kind of worked except when you would zoom in, you couldn’t read any of the CD spines.  So, in other words, it didn’t work at all.

The real solution would be to take multiple photos and stitch them together.  So, that was my next attempt.  I scanned one row of CDs and took a series of pictures, then took them to the computer to mate them up.  That proved to be very difficult because each picture had to be adjusted to compensate for rotation and zoom and also lighting.  This was proving to be a non-solution as well.  I had a massive number of CDs arranged on my floor and I was running out of ideas to photograph them.

Since the problem with my stitching/panorama concept was consistency, I came up with the idea that I could build a trolly-type of rig to suspend the camera over the arrangement.  This would keep the camera at a constant height and angle where each picture would be the same.  It was a pretty clever idea and made me feel pretty inventive.

So I went to work assembling the rig and shot the first row of CDs.  I took the pictures to the computer and stitching them was actually pretty simple.  This was promising.  I evaluated the size of the arrangement and determined I would have nine rows of photos to stitch up.  This would take days, certainly.  But that’s ok, as long as I made progress.

I shot the second row of photos and brought them in for stitching.  Suddenly, things weren’t lining up anymore.  The first couple photos worked out, then all of a sudden, the scale didn’t fit any more.  Thinking I must’ve shifted the camera somewhere along the way, I re-shot the row of photos.  Again, at the same place, the photos failed to line up.  I wasn’t sure if it was the first row of photos that were somehow misaligned and causing the second row to not match up, or maybe it was just something that was intrinsic to the photos themselves.  I was noticing there was a slight fish-eye effect from the 35mm lens I was using, so the CDs on the periphery of the photos were skewed from the ones in the center.  As I would line up the images on the outer edges, they would be distorted from the ones trying to be matched in the center.

At this point, I had had my CD collection completely dismantled, on the floor, for a week.  This was causing me a little bit of stress.  I was unable to use my listening room for any listening because the floor was consumed with this arrangement.  I was adding new CDs to my collection, but they were in a separate stack, not integrated yet.  My patience was running low, and my prospects of success were low as well.

The next thing to attempt would be to use my 50mm lens on the camera, which wouldn’t fish-eye as much, but that would take much closer images of the CDs, unless I built the rig even higher up, which I wasn’t too keen to do.  So, I accepted failure and began the process of dismantling and reorganization.

And the point here, accepting failure, is the key.  "Failing", or "giving up", is not a bad thing.  There are plenty of other phrases that exist to make yourself feel better about the situation, like "cutting your losses", and something about "reward vs. effort".  those phrases get closer to the reality of the situation.  Right now, this is not something I want to tackle.  It was a good idea, and one I may revisit in the future with an improved vision and more commitment, but I want a return to stability.  There would be no way I could clear my mind enough to consider any means of improvement with everything all out like that.

In the next iteration, if there is one, I would definitely test out some techniques on smaller arrangements, instead of committing fully to a full collection dissection.  That was days of effort to dismantle and it will be days to reassemble, too.  So until next time, fail on.

Well, that was a quick failure.  My previous idea, which was to use older technology to gain access to some data that seemed out of reach, was ill-informed.  Before I even got all the pieces of my $16 project, I figured out how to accomplish what I needed with what I had already.

To start, I was under the impression that my hardware could not read subchannel information from audio CDs.  This is false.  I just need the right software.  And the software was what was causing all of my misconceptions.  I use Exact Audio Copy (EAC) for my CD ripping.  There is an old version, specifically v .95 prebeta 3, that had a feature that would let EAC do direct reading of the track data, which would include the subchannel information.  Because of some legal ramifications of doing so, this feature was removed from prebeta 4 and all future versions of EAC.

I had acquired a copy of prebeta 3 and initially was not able to get it to launch.  Then, I was able to get it to start up by running it in administrator mode.  But then, it wouldn’t recognize when a CD was inserted in my drive.  So, since prebeta 3, code has been improved for things that are essential to its operation on modern operating systems, but code has been taken away for the feature that I needed.  In short, EAC is not going to accomplish what I need under any hardware configuration.

There is another, more modern, ripping tool called CUERipper, which is part of the CUETools suite.  I tried this tool briefly, but did not like the way it handled the ripped files, for one reason or another.  Instead, I kept using EAC.  But now, upon another evaluation, CUERipper will read the actual track data in the same way that EAC refuses to do.  And it does read the subchannel data, which is what I really needed.  That means, CUERipper is the software that I need and all that old hardware is unneeded.  All to waste, I guess.

Still, CUERipper is an inelegant tool and it doesn’t do things the way I want.  However, it is open-source software, and it is actually programmed in .NET.  That means I have the ability to actually change the program to do exactly what I want, how I want it.  And so that is the future plan.  The next few days I won’t be able to do any work on this, but after that… I’ll be able to make my own custom ripping software.  How awesome.

Some small part of me is actually considering re-ripping my entire collection again.  Why?  Well, over time I have replaced CDs here and there but may or may not have ripped the disc that is actually on the shelf.  So I can be sure that my digital collection is out of sync with my physical one.  That effort remains a wait-and-see.