During my time as a content creator in Blue Mars, one of the most promising features I looked forward to was the upcoming cloud streaming technology through OTOY. Ultimately, OTOY’s inability to deliver and Blue Mars’s poor performance with respect to attracting a larger audience derailed any potential that could have otherwise been realized. With cloud streaming once again on the news front, it appears as though discussions regarding the validity of the business model has overridden the more important narrative of what cloud streaming offers not just for end-users, but most especially content creators. The ability to run Second Life on high graphical settings through almost any device regardless of the specifications is monumental. However, the best case to be made in favor of cloud streaming is its effect on in-world content theft. OnLive encapsulates the virtual world experience into a video stream. Due to the insular nature of such a delivery method, this effectively nullifies the harmful effects of copybot clients as no data, xml or otherwise, is sent to the end user. There is simply nothing to capture other than by way of analogue. For the very first time, texture artists, modelers, animators, musicians and sound designers can now benefit from the same level of server-side protection already enjoyed by programmers. Moreover, Linden Lab is afforded the rare opportunity to hit the proverbial reset button, if you’ll forgive the recent political pun, on their past decision to opensource their client viewer. As everyone is well aware, this resulted in the spawning of many variants that were used for malicious purposes. An official viewer-only policy through the gateway that is OnLive would help limit the capacity for a user with ill-intent to disrupt the experience for others. The caveat? Safeguarding in-world content in this way requires an all or nothing approach - accessing Second Life through the cloud would thus, have to be mandatory for all users. Naturally, the way is barred by a number of issues; cost being chief among these. The current pricing structure will simply not work. - It wouldn’t be out of the question to surmise that processing the vast amount of unoptimized content and compressing it into a video stream requires no trivial amount of resources. One possible measure for minimizing the impact would be to adopt standards which promote responsible content creation. Criteria such as triangle count, textures and memory usage for scripts can factor towards the total “render weight” calculation. When exceeding past a certain threshold, mesh/sculpt geometry and texture resolution would be forcibly reduced or not rendered altogether if no suitable LOD is found. The threshold in question should probably be more generous than current-gen standards for commercial games but still remain within a reasonable limit. This way, processor overhead will be more manageable and thus, lower operating expenses – the primary driver of the current pricing scheme (I think). The result should be appropriate adjustments for a more competitive package, as opposed to the rate of $2.50 per hour. Ideally, it should be free lest the userbase shrink dramatically. - Options should be available to save text logs, screen captures and even brief video recordings to disk. Automatic publishing to popular social networks would be a plus for the more casual userbase. Needless to say, the resolution for the latter two should exceed the quality of the video stream. - Content creators need to retain the ability to backup their own work locally for objects under which they have full permissions. Import and export options should be made available to help facilitate this. - A lightweight text-only client will be necessary for users who want to manage their inventory, estate, groups, contact list, communications and financial transactions with others while keeping bandwidth consumption to a minimum. - The lack of any rigging capabilities for mesh objects has forced content creators to emulate animation by creating multiple instances that act as keyframes from which the visibility of textures are swapped. This is commonly done for vehicle components and limb attachments and the result is a substantial increase in rendering cost. Prior to enforcing rendering cost measures, Linden Lab should consider introducing features like object rigging to provide content creators with the freedom to innovate without having resort to extreme methodologies. Unfortunately, there does exist a set of issues for which I have no workable solutions for: - The double-edged sword that is a single proprietary client that, while providing an added margin of security, is also devoid of the innovation stemming from third party viewer development teams. - Opensim solutions, important for the academic sector, will not benefit from further technological gains made by Linden Lab. - A video stream-only method of delivery is not compatible with external peripherals such as the Oculus Rift. - Commercial viewers such as Lumiya will be adversely affected; depriving the authors of any further income. - A sizable amount of legacy content will be disqualified if rendering cost measures are enforced. - Bandwidth caps will prove problematic for those who do not have access to unlimited packages. Being a resident in North America, I’m familiar with this predicament all too well. Were I to speculate, Second Life will likely never see the day where access via content streaming becomes mandatory. The stiff resistance to such a dramatic change in conjunction with the mass amount of incompatible legacy content may prove to be too insurmountable an obstacle to overcome (to say nothing of paying by the hour). What is truly needed is a new platform, one that is unencumbered by 11 years of political baggage – a clean slate if you will. Something like High Fidelity….