Rationale and Goals

Rationale

XBMC is currently one the foremost media center applications available, offering a wealth of functionality, flexibility and performance.

One of the most compelling aspects of XBMC lies in a flexible user interface that allows for a wide range of 'Skins' (user contributed front end) that allow XBMC to appear markedly different visually, but share the exact same 'engine' across multiple platforms and devices.

As part of this powerful capability, XBMC relies on supporting 'Art' (images) that are related to specific media.
This 'Art' can be in the form of photos, promotional artwork, CD covers, DVD covers, logos, etc etc that are related to a movie, a tv show, music, or indeed any type of 'media'.

Today (Summer of 2009) XBMC supports two critical types of art…

Thumbs: A 'thumbnail' / iconic image that represents a specific movie, tv show, album, artist etc
Fanart: A more general artwork typically used as a 'wallpaper' or 'backdrop', in support of a movie, tv show etc.

By the Summer of 2009, it became clear that XBMC users and skin designers wanted to push the boundaries of XBMC art a little futher and introduce more types of media related art.
At the same time (as XBMC became more and more popular), an increasing number of end users found it difficult to configure XBMC (or set out their media storage) correctly, resulting in many support requests on how to make artwork appear in the correct manner.

With this in mind, a small number of XBMC community members decided to try to improve this situation by developing a proposal for how this situation could both be corrected, and indeed improved upon.

The result is a file naming schema for media related artwork.

Goals

1) Offering a totally consistent naming convention for art
XBMC currently supports a number of ways to name artwork, some of which work only in a hierarchical structure for media storage, and some which work in both hierarchical and flat file structures.
This proposal aims to reduce the number of possibilities, whilst at the same time providing a solution that works equally well regardless of how the end user structures their media.

2) Being completely unambiguous about what the art is doing
XBMC currently uses a number of filenames for artwork (folder.jpg, movie.jpg for instance), and these are not always clear to end users what they are for, and how they will appear in the UI.
This proposal aims to rectify that situation with clear unambiguous filenaming

3) Is placeable ANYWHERE in the filesystem
XBMC currently requires artwork to be closely coupled (physically) with the media file itself. Whilst this isn't necessarily a bad thing, it can increase the time taken to 'find' artwork, and does not lend itself very well to having artwork placed in optimised locations (cache or fast access locations).
This proposal aims to offer a solution that does not place any restriction on where artwork must be located.

4) Allows a wider range of art than is currently possible
XBMC currently focuses around two types of art… Thumbs and Fanart. However, there are now a plethora of additional art types available for movies, tv shows, music etc.
This proposal aims to support a far wider (and extensible) range of artwork.

5) Supports different orientations of art
Skins often choose to represent media in quite different styles (e.g. wide banner, portrait, landscape etc) and therefore a thumb or fanart has to be specifically formatted / optimised for the skin. Today XBMC can only support artwork in one orientation. Should a user change skin, the user may be required to reconfigure their artwork collection in order to support the skin.
This proposal aims to offer a solution that allows an end user to maintain multiple orientation styles of artwork that allow new skins to be used without a need to reconfigure their artwork collection

6) Is possible to rapidly port the existing scheme to this proposed scheme
It is the aim of this proposal to offer a solution that can take existing artwork setups and rapidly 'port' them to the new proposed scheme without loss of art, or functionality

7) Allows for art at all levels of media
Today's artwork solution within XBMC is limited to specific 'levels'. That is to say, artwork can only relate to specific TV show, or a season, but not to an episode. In the same way, it can relate to a music artist or album, but not to a track.
This proposal aims to support a higher level of artwork 'granularity' that more closely correlates to specific media objects (e.g. down to episode, or track level)

8) Allow multiple instances of artwork
Today, XBMC only supports one instance of art for a given media file (so it's not possible to have multiple covers, or fanarts etc
This proposal aims to support multiple instances of artwork

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License