Peering Inside: Whither avatar puppeteering?
Avatar Puppeteering (also known as Expressive Puppeteering) was shown off at Linden Lab's inaugural Second Life Views session back in 2006. Since then the project has presented the occasional teaser, though the news was handed down in September last year that the puppeteering project was on hold while the focus was on bug-fixing. We asked again in the new year and we were told that it was still 'on hold'.
With recent teasers from Jeffrey Ventrella, the mastermind behind this project (and behind flexi-prims we understand), we did a bit more investigation, and we think that 'on hold' is something of an understatement.
Avatar/Expressive Puppeteering was a system that would allow users to animate their avatars in an ad-hoc fashion, by holding down the Control key on the keyboard, and then simply click and dragging a body-part to a new position with the mouse. Grabbing and moving a hand would allow you to point, wave a hand dismissively, wave hello or goodbye or indicate yourself. Or you could shake your head, cross your legs, all without requiring predefined animations.
The first essential clue we got about the status of the project was Ventrella's new website on the topic. All written in the past tense, and the domain registered to himself, rather than to Linden Lab, it read more like a record of past hopes than current developments.
Indeed, Ventrella left Linden Lab in August last year, we are told. We were unable to press for any additional information given Ventrella's NDA (Non Disclosure Agreement), but it appears that the puppeteering project has been on hold since at least then (and we're guessing quite possibly longer), and we get the increasing impression that unless some minor miracle occurs that it is more or less a dead project for the foreseeable future.
What we can deduce, though, is that Ventrella's focus on new features (after all, that's specifically what he was hired for) was essentially incompatible with Linden Lab's war-footing as it geared up for a single-minded focus on bug-fixes, stability and scalability last year.
We don't know how far along the project actually was, but with all the architectural changes to the viewer, every version seems like it would be making any development work done on the puppeteering project increasingly less likely to be integrated into future versions as the viewer code-base diverges from the code that puppeteering was originally written for.
Our take? The project is effectively dead. To restart it will likely mean throwing away most or all of the work that was originally done and going back to nearly square one.
