Project Management and Metadata

I just did a search on my hard drive and found I have nearly 3000 cakewalk projects dating back to pre-Sonar days. 

Some of these are garbage, snippets, riffs, etc, but the fact that I can't tell the good from the bad highlights the fact that Sonar is in desperate need of an overall asset/media management system.  They have had the project "info" for years, but it never developed into anything that could be tagged/searched/cataloged in any meaningful way.

Some random thoughts on some broader concepts that Sonar needs to keep up with the pro market.

1) Unique identifiers that stay with Tracks

I do a lot of production library work for different supervisors The descriptive working title of the scratch track  i.e. "2009.12.05 introspective Am minimal 92bpm" is almost never the final name of the track they receive.  Add in versioning, re-titling by each of the eventual libraries, different edits of different lengths and you may have 6 different "songs" with 6 different titles, all pointing back to this original track name for the project.  Keeping an Excel spreadsheet updated with project names, edits, re-titles, re-mixdowns, etc. is killing me.

Sonar needs a database-type function that allows each project to have a unique identifier that stays with it from start to finish - something that can be referenced to find the project from  a search function, and something that doesn't change when the project name changes.  A music supervisor calls me with a track name they need a timing tweak on - I need this reference number to retrieve the correct version of the track that was used to create that particular mixdown.

which brings me to the next point:

2)  File/Project Management

Project storage /retrieval in Folders based on names is so 1987.  Imagine if your address book was organized by folders like "Friends", "Family", "Business" and under each folder was a folder with different characteristics like "hair color", "birthdate".  If you wanted to find someone, you couldn't search based on one of these characteristics--you'd  have to remember how you categorized them on that day and look in the proper folder and subfolder. The more subfolders and descriptions you had, the more segmented your search would be.  What about a friend you do business with?

Sonar is still operating under the folder metaphor - it lets you use per-project folder management, which means everything stays nicely within the folder, but that project sits among hundreds, thousands, organized by what - BPM?  Key? Mood? Date? Synths Used?  I find it mind boggling that Cakewalk with all of its technical prowess has not come up with a better method.  Look at iTunes.  I honestly don't care how it manages my underlying file structure in my library - it is indexed to the nines, and I can access any of the 50,000 tracks in seconds, based on comments, tags, metadata, artist, BPM. last played, etc. and not just on one parameter like file location by Artist - Album Name. 

3 Metadata

Similarly, I want to be able to find something like  "all aggressive tracks in the 120-135 BPM range created in the last 6 months".  My current workaround of having the "name" date, bpm and key in my project title is incredibly primitive by this standard.  Not only would searching these tags be an incredible timesaver, every Music supervisor and Library I work with needs and they all need metadata included with the submissions.  I should be able to add these tags, etc. as I am creating the track AND quickly export it to spreadsheets, xml, etc.   Some of them are built into the track - the BPM of the song is already known without me having to enter it.  Why can't we search on it?

4) Thumbnails

I know I have harped on this in the past, but finding the project is only part of the battle.  I have a deadline of 2 hours and I need a particular track.  I've used the metadata search function (that will exist one day...or this Loyal Cakewalk user since pre-Pro-Audio will be on another DAW unfortunately), and have it narrowed down to 15 possible tracks.  Unfortunately, I don't know what they sound like, so I have to open each project. 
With large sample libraries and plugins, it can take considerable time just to load a project to be able to play it long enough to audition it.  I've wasted maybe 20 minutes, assuming every project opens without a hitch.

I was so desperate I've had to develop my own kludgy workaround to create preview mixdowns of projects so I can quickly audition them outside the project.  I create an Archived track called "Preview Mixdown" Bounce the project down to this track, and now I have a track with a truncated filename like "2009.12.05 introspective Am min Preview Mixdown (3).wav.  I can search the wavedata folder for anything with the "Preview Mixdown" in the name.  Drawbacks - they don't stay current, they are huge, If I change the name of the project, the preview keeps its old name and is "lost".  When I make changes and re-mix down, the old files are still there (2009.12.05 introspective Am min Preview Mixdown (1).wav. ,  2009.12.05 introspective Am min Preview Mixdown (2).wav.  etc.).

What I would give to have a thumbnail process managed by Sonar - a decent resolution MP3 file that could contain as much or as little of the song needed to identify/audition the track prior to opening it up...one that stayed with it, could be updated with changes with a simple click of a button "Make Preview Thumbnail"  This would stay with the project and the names would stay concurrent.

Video editing, Audio editing, Photo editing, Word Processing software all have thumbnail previews of files that don't require loading the entire 300 MB project. 

The Database Manager

In my dream world, Sonar's overall project management would take this over and the simple "File open" dialog box would be replaced by a screen that lets me boolean search, sort, preview a thumbnail and/or current mixdown my projects.  This would truly be "Managing" projects rather than just stacking them in semi-random piles in folders with the names they were given at the time they were created.  Right off the top of my head I can think of some of the metadata needed, which in today's XML-driven world should have been integrated into Sonar years ago:

ISRC
Duration of Track
Project Completion Status %
Major/Minor
BPM
Alternate Titles
Mood
Genre
Project Comments
Musicians appearing
Writers
Mixed Down/ Mastered (i.e. ready to send)
Some sort of Status Category - (Is this a riff, Drum Pattern, Sketch, Rough Mix, Final Mix, Remix?)
Copyright
Bit Rate
Plugins Used
Samples needing clearance
Encumbrance (exclusively signed, rights reserved, creative commons, etc.)
Location of Mixdown/Master if done outside of Sonar
Track comments

Conclusion  (and thanks for reading this far).

I should have access to ALL of this info before I even open a project, and should be editable outside the project.  All versions of the track should be easily identified from this main screen - prior versions vs. alternate ideas, and "retired" projects able to be marked read-only.  I'd really like to stop using Excel and Windows Find and Explorer to manage my Sonar projects.

I have huge respect for the folks at Cakewalk and their responsiveness to customer's needs.    I totally understand that many of Sonar' s users are hobbyists, but Cakewalk has positioned Sonar with World class features as the killer DAW capable of professional use on par with every other DAW out there and these issues to me are major omissions (and opportunities).  Sonar is extremely capable and it is like my right hand at this point, but I need it to step it up another notch in this regard. 

I push because I love.  :-)

The Production Music Association has been pushing for a standardized format for Metadata, which would be an excellent thing for Sonar to interface with directly.  (I'm not affiliated with this organization)


METADATA STANDARDIZATION FORMAT


The PMA announces the release of a standard metadata format for use by production music libraries worldwide.
Adopting this metadata standard will allow production music libraries to have registration information stored in a flexible format that can be easily integrated with existing systems such as CWR (Common Works Registration), PRO registrations, etc.  The PMA has pioneered this initiative to establish a standard method for storing valuable metadata in a format that is more universally compliant with all systems.







For the form and additional information go to:


[link=http://app.e2ma.net/campaign/27456.758e83ebcfe968cfcfe8bbed85713cb2]http://app.e2ma.net/campa...e968cfcfe8bbed85713cb2[/link]

[link=http://app.e2ma.net/campaign/27456.758e83ebcfe968cfcfe8bbed85713cb2][/link]

I just started my too-often put aside project of cleaning up my projects disk. I've got a little over ten years worth of projects, and in my case it gets a little more interesting since some were done in CWPA or Sonar, and some were done in Finale - and some used both tools.

All of the demo projects are in really good shape except for maybe some of the scratch tracks, which I could have named better.

Some of the theatre projects are in decent shape, which is to say I could reproduce exactly what I did back then, but if I had to make any changes I'd be in a world of pain.

And then there are some where I may have several GB of sound effects or music clips that I never actually used... or at least it appears that way.

My own compositions run the gamut from being well documented to being worthless because I have no idea what I was thinking.

Proper documentation is important, and no software tool will do it for you, BUT, if the tool makes it easy I think one has a better chance of remaining disciplined.

Indeed... this is an excellent suggestion!

Thanks for the support. 

Feel free to expound upon (or diverge from) the ideas with other suggestions - they are wishful thinking at this point, but with some of the many pro's on this forum throwing their weight behind some of the concept, some might could become reality.

Great ideas; It cost us several thousand $ for a librarian to build a custom data base that does much of what you want and maintaining that data base is a non-trivial ongoing expense.  So, I wouldn't get my hopes up.  Sound's like a great opportunity for a programmer to build an app and make a lot of moola!

This kind of approach would help Sonar not just keep up with the Jones's but leapfrog them.  Similar to Sonar being the 1st to adopt 64bit.  Graphics programs have been using Metadata forever - why not DAW's?