THE IMPRUDENCE MANIFESTO THE PROBLEM The Second Life Viewer suffers from a stifling atmosphere of non-change. This atmosphere emanates from Linden Lab, whose attitudes and policies discourage all but the smallest and most superficial improvements. This is the result of the nature of Linden Lab as a corporation; in particular: * A lack of resources to invest in making significant improvements to the Viewer. Linden Lab is stuck treading water; they don't have the manpower to spend on making major improvements to the Viewer. * A burdensome Quality Assurance procedure that punishes any sort of change. The larger the change, the more it's punished. The simple fact that all changes have to pass through such an arduous process discourages anyone from attempting any major endeavor. * A vast throng of paying customers who want the Viewer to remain constant and familiar. If the Viewer changes, they are forced to spend effort relearning it. The most vocal and abusive users even meet change with sneers and insults. Combined, these render Linden Lab ineffectual at dealing with the fundamental usability problems that plague the Viewer: * A cluttered interface that frustrates and confuses both new and long-time users. It is difficult to learn, frustrating to use, and obscures the thing that users actually care about: the world. * Crude tools that force users into awkward and inefficient workflows. The tools bear little relation to the way today's users want to use them. Whether because these tools are outdated relics of the past, or just half-baked non-solutions, very few rise above even the "somewhat usable" mark. * Stability and performance problems that make the Viewer unreliable for any real use. Viewer crashes are an everyday occurance for many users, disrupting their activities, destroying their unsaved work, and afflicting their lives with undue stress and frustration. We don't deny that Linden Lab has made some progress in these areas over time. But such progress is slow and often superficial, because Linden Lab cannot afford to make significant improvement. Encumbered by their own nature, they are forced onto an untenable road of caution, hesitation, and prudence. This is a path that can lead only to stagnation. These fundamental usability problems will not be solved by Linden Lab; their atmosphere precludes it. Even the contribution of source code patches by open sourcers can only address one factor, Linden Lab's lack of resources. Patch contributors are still affected by the burdensome QA process and the community's anti-change attitude just as much as Linden Lab developers are. Any endeavor which relies on Linden Lab to approve and integrate changes will face the same bottlenecks. SOLUTION As I have described, Linden Lab's approach is characterized by three factors that put a damper on significant improvement or innovation: * Cautious, gradual changes over a long period of time. * Dependence on an overloaded central QA department. * Yielding to pressure from large numbers of users to reject change. It stands to reason that a project which removed or reduced these factors would be more free to make radical, fundamental changes to the Viewer. Such a project would exhibit the opposite characteristics: * More significant changes over a shorter period of time. * A scalable, hierarchical QA process to screen and approve changes. * Less need to satisfy users who demand non-change. This is precisely the sort of project we propose to undertake. Because our approach is incautious, and perhaps even reckless, we have dubbed the project Imprudence. Goals The primary goal of Imprudence is simple: to greatly improve the usability of the Viewer. In particular, there are 3 aspects of usability that we intend to address: * Approachability. Improving comfort and ease of use, especially for new or non-technical users. * Efficiency. Improving speed and ease of common tasks and workflows. * Satisfaction. Improving the emotional effect of the software on the user. This is not to minimize other aspects of usability, such as reliability, accessibility, or internationalization/localization. We recognize their importance, but lack the expertise to properly address them. We welcome people with such expertise to join the project and help. METHOD In order to achieve these goals, we propose to combine the open and distributed nature of open source development with the dedication to quality design usually associated with commercial products. The main highlights of our approach are: * Open, public project management. Our plans, goals, and roadmaps are laid out on the table. Our code is published in plain view as it's written. There won't be any sudden bombshells to disrupt your plans. * A pro-change atmosphere. Change is natural and healthy, and it is the only way to make improvement. We encourage experimental change, coupled with evaluation to filter out the negative. Users wanting a static, unchanging viewer should look elsewhere. * Designers, programmers, and users working with each other. Good software requires more than just a team of programmers. It needs designers listening to users, programmers working with designers, and users testing and providing feedback. Every role is necessary and appreciated. * Commmunity involvement. There are many ways to be involved, whatever your interests, skill set, or level of commitment. Contributions are welcomed, not looked upon as burdens. * A modern, distributed development model. The Git version control system makes it easy for programmers to work freely without stepping on each others' toes. Easy and powerful branching and merging tools mean that releases will never be littered with untested, half-baked code. * A scalable, hierarchical QA model. New contributions work their way up a hierarchy of approval, which will start small and grow naturally as load increases. By the time a contribution reaches the top of the hierarchy, it has been tested, polished, and approved by multiple people, and is ready to be integrated. A CALL FOR VOLUNTEERS Imprudence is an open-source, volunteer effort; it depends on people like you getting involved! There are many ways to contribute, and most of them don't require any programming skills or special knowledge -- just some free time and the will to be part of something great. If you'd like to contribute in any way, have a look at CONTRIBUTE.txt. You have nothing to lose, and a better SL experience to gain! SIGNATORIES Jacek Antonelli (August 27, 2008) McCabe Maxsted (August 27, 2008)