Had an interesting discussion with Sebastian Thiel over at Trixter about a tendency I often see myself in many of my freelance ventures. The tendency of artists within and during production to develop tools to help manage their own work. Essentially creating their own little ecosystem, or pipeline, within the already (non)existing pipeline.
I’ve got an example from a recent job. A two month gig producing a fully cg sequence with about 7-8 people involved. The studio lacked any tools for locating or making assets, beyond the traditional word-of-mouth conventions of where to put what. An artist is affected by this vacuum and goes on a mission to develop a miniature library ui, with methods for shot building/generation based on available assets on disk.
Since all of the available assets were created by hand, consistency in names and internal hierarchy couldn’t be guaranteed and thus the programs written upon it had to either back-flip to compensate or disregard anything that didn’t conform to the word-of-mouth convention.
Back-flipping means to compromise and compromising means to introduces bugs. After a while (and 50+ assets in need of updating) I thought “how great, this tool would save me lots of time!” and so the farce begins.
My first experience was, as expected, not perfect and some of the things that suited my colleague well was either unnecessary or in the way of my goals in using the tool. Thats fine, its early, I’m just happy I no longer have to manually reference these assets into my Maya scene each time multiple (unknown, uncontrolled) assets have been updated.
My colleague wrote a tool to help himself produce better work faster and now I’m in the mix relying on his functionality. If the tool turned out well, I’d recommend it to my peers in turn and we’d start the see the light of a better workflow.
In production, you don’t document. Not documenting is what would make it difficult for others to build upon what has been previously created. Understandably, this artist wouldn’t have been able to make many guarantees regarding his software. After all, it wasn’t meant to last – it was meant to simplify this one little thing. How can I build upon that?
Finally, at the end of the project, we, all freelancers, parted ways. Leaving the sub-pipeline hidden and undocumented in the midst of this one project.
Ready to go again?