Linden Lab founder and chairman, Philip Rosedale, yesterday announced changes to the open-source Second Life viewer program that would enable the fast-tracking of user-contributions to the code-base. The viewer source-code was originally released just a little over two years ago on 8 January 2007, and spurred an immediate surge of development, spurring developments of a variety of software including third-party server opensim.
The procedure for actually submitting patches to the codebase however was clunky, and some contributors abandoned their development efforts after contributed code (some which corrected egregious problems) was left to languish for many months without approval. The new scheme seems set to fast-track user-contributions and eliminate that particular problem.
According to Rosedale, this new project will feature:
- Ability for contributors to directly commit code to a public repository in which the Linden Lab team is also directly working.
- Automated build system which operates immediately on checkins and produces publically-accessible executables.
- Distribution of the new version alongside main SL viewer distributions.
- Automated unit and performance testing with public results.
Patches will still be submitted initially to the public JIRA issue-tracker, and a link sent to the Second Life open source development list. Submissions that raise no objections will then be granted approval by the Lab for direct inclusion in the new repository. Essentially it looks like Linden Lab is creating a fork of the viewer.
We pulled aside Jacek Antonelli, a founder of the Imprudence Project (a popular third-party Second Life viewer), for her thoughts on the announcement.
"I'm not quite sure what to think about it," Antonelli told us, "It's exciting, encouraging, and rather unexpected. I'm a bit skeptical that they'll give it the attention they say they will. I wouldn't be surprised if they just kind of 'forget about it' after a month or two, when their priorities shift."
"Unless they exercise some control over the design, it'll likely be an amalgam of different (and often competing) ideas and concepts. I guess that's not all that different from the regular viewer, actually."
On the whole this would seem to put Linden Lab in a position to cherry-pick the best features and ideas of contributors to include in the official viewer, though with the open approach, it is unlikely that this repository would support any sweeping changes that cross multiple viewer subsystems, like heavily revamped user-interfaces.
Based on that, it doesn't seem like we're likely to see terribly much reduction in development of third-party viewers, and many of those will likely benefit from the availability of tested patches for performance and bug fixes.
What's your take on the new approach? Grail or fail?
|Are you a part of the most widely-known collaborative virtual environment or keeping a close eye on it? Massively's Second Life coverage keeps you in the loop.|