Music is a uniquely evocative thing, and whilst technology has brought a huge improvement in sound quality, radically changed the means of distribution, yet I can't help feeling that it has lost a little bit of the magic. I vividly remember buying my first 45rpm single, just as I remember my first real album, and my first CD -- and most of them mI still have. The joy of removing the disc from the sleeve, and carefully placing it on the player to listen to, perhaps whilst reading the sleeve notes...
Even going out to the pub brought music magic -- punching in the track numbers into the jukebox. Where is all that now? Instead, I stare at a screen, trying to scroll through endless lists of tracks, not really knowing what I want to listen to, and getting frustrated beccause everything is too darned slow. I need a new approach.
the Duke brings back the tangible, tactile, music experience -- and makes playing audio a lot simpler and quicker. Whilst it has been planned for years, and intended to be used by everyone in the family, the credit for finally motivating me to get off my lazy arse has to go to the sternest and most ruthless critics of all -- my two children -- who really couldn't manage using Lyrion as it currently stands. So, I move away from anything which requires a display, to something that is cheap and simple to create and use -- but powerful enough to adapt to every circumstance: the humble barcode.
The concept is simple: Audio is brought back to a physical object, upon which is affixed a barcode. The object, of course, can be anything -- but in our family we print a jpeg of the album cover onto a piece of card, stick a barcode on the reverse, and then lamitate it just for good measure. This provides enough protection for general use.
We have tried to optimise the barcode design to deliver maximum flexibility within the minimum search-string length. The printed barcode size should be relatively compact, and yet should have a high-resistance to false readings (ie: error correction should be as high as is practically possible).
Whilst the end-user is free to choose any barcode symbology they feel works best for their own needs, we have gone for the QR-Code. The contents of the code is as follows:
- The URL. This is OPTIONAL, but including it enables the QR-Code to be read by any modern smartphone. Of course, you will need to ensure that the server is actually accessible over the network -- but please don't open up LMS to the public internet!
- The Query.
To save space, this is an array, consisting of three fields:
- Object Type: A single digit, which determines whether we are wanting to play a single, an album, a playlist, a classical work, an episode, or to issue a command to the controller or to the server.
- Search String: A simple array
- Artist Criteria: A complex array, consisting of either one or two arrays. The first holds the artist name(s), with the second (optional) containing their respective role.
The power of this system lies in the simplicity of the query, and of the tolerance to variations of file-tagging within the library. If you have inconsistent tagging, it won't matter if the work that you are searching for is by "Mozart, WA", "Mozart, Wolfgang", "WA-Mozart", or any other possible permutation thereof. Just searching for "Mozart" is good enough. And, if you have works by Leopold Mozart in your library as well, just ensure that the rest of the search string is sufficient to narrow the results returned...
... unless you actually want ambiguity. Yep, we can manage that too. [1, ["Abba"]] will
give you a track from Abba, chosen entirely at random!
Some examples of fully-functioning labels in our home.
Within the 'Manage Plugins' menu, add the 'Additional Repository': https://raw.githubusercontent.com/whitepj/the_Duke/main/repo.xml or https://github.com/whitepj/the_Duke/raw/main/repo.xml