Abstract Factory Blog


31 Jul 2013

Since heading off to StartupWeekend #London on the 12th of July 2013, Pipi has been featured in Campaign Magazine and PRWeek, gotten a website, a Facebook and Twitter account but most importantly –  I now have a team.

We’re a hacker, hustler and designer fulfilling the three D’s of responsibility – Design, Data and Distribution – and we’ve been hard at work trying to figure out how to deliver the first pass of Pipi to you guys.


For the Alpha version of Pipi, I’m envisioning two separate tools for File Handling and Metadata respectively.</p>

Perfect File Handling

As we started to collect feedback from various studios (the deep pockets), technical artists (the integrators) and artists (our end-users), it became clear very early that the first step towards establishing a foundation within a production would be to properly manage their file-system.
Files are super important in any sort of work, especially so in an environment with potentially thousands of files generated, created and modified on a daily basis by the many people throughout a production.
There needs to versioning, categorisation, metadata and above all – consistency. And then there’s the one law that such a tool abides to above all – it must do no harm.
For the alpha, we’ll introduce the Dashboard how it solves these matters.

Meta.. what?

A file cannot tell you why it exists. It cannot tell you about where it came from or what it’s purpose is. For these matters, there is the concept of Metadata, or “data about data”.
One of the first barriers I faced when envisioning Pipi was how to store metadata together with the content I was looking to enrich. As it turns out, this has been done to death in the past and the go-to solution for doing so is to use a database, such as MongoDB.
But to us (you and me), metadata is so much more than just text. Metadata means screenshots per version, reference videos per shot, audio for animators as well as key/value pairs for automation throughout the pipeline. Short text, long text, tables and hierarchies of text, all of that is part of the data about our content.
For the Alpha version of Pipi, I’ll introduce Augment and how it solves this in a way which is both easy to learn (no coding required) and infinitely scalable and adaptable (any OS, any content, any data-format, any size)