If ever there were an appropriate time for cliches, this moment in Second Life history is a prime candidate for “you cannot see the forest through the trees.” In the case of the Emerald “scandal,” the trees have been stared at, chopped down, shredded, reconstituted, and still not yet drained of all perceived viable uses, yet even still the community as a whole cannot take their eyes off the trees for one moment long enough to see the big picture. In fact, I’m not sure most people even realize that there is a big picture, and I find this depressing. Oh I see your perplexed faces behind those computer screens as you read this asking yourselves, “what the hell is this woman on about?”
The trees I’m referring to are the various issues that I’ve seen beaten to a bloody pulp in-world and all over the internet during the last few weeks with respect to Emerald--emkdu, DDoS, data-mining, banning, blocking, user safety, the Third-Party Viewer Policy (TPVP) and its Directory (TPVD), the viability of Emerald, and so on and so forth. I’m in no way delegitimizing any of these discussions any more than one could suggest a forest doesn’t need trees. However, with what appears to be the heart-breaking demise of a viewer enjoyed by roughly 125,000 unique users each day, and the information that has come to light during that process, it’s high time people shift their focus and start looking at the bulldozers surrounding the forest.
Now, before I delve into the big picture, I think it wise to give you some facts about me and my perspective through all of this, since, quite frankly, many of you don’t have the foggiest idea who I am and have no reason to listen to my ramblings or give them any credence whatsoever. I’ve been deeply involved in Second Life for a year now, which for me, is long enough to know a thing or two, but not so long as to be completely jaded or have developed an “I know it all” attitude. If memory serves, I’ve been an Emerald user (erm, that makes me sound like some crack addict *laughs*) for about 6 months. Despite loving the viewer, I still attempted to use Viewer 2.0 when it came out and nearly gouged my own eyes out after 15 minutes. I joined the Emerald Support Team *drum roll please* the week before the DDoS news broke. Prior to getting involved in the Emerald Viewer group, I had no history with any of the Emerald Development Team or the Support Team for that matter. I had no personal vested interest in the Emerald Viewer except for one thing, and one thing only--its community of users, for whom their preferred viewer is what helps make their Second Life experience that much more enjoyable.
My stated vested interest has been the only thing that’s driven me through this time, and for the last couple of weeks, I, along with the entire Support Team, have worked tirelessly to continue to provide the best support we could to our users. Throughout this entire affair I believe I have remained rather even-handed in the positions I’ve taken. I have been both supportive and critical of several members of the Emerald Development Team on both sides of the internal conflict that lead to the end of the project. I never assume anything as fact or truth without having the information necessary with which to base my own opinion or come to my own reasoned judgments.
So after having said all of that, here is my reasoned judgment of the “trees” that make up the recent Emerald events. Although it may have been the actions of one or a few who truly caused the downfall, the Emerald Development Team failed as a whole in allowing a structure to persist that created the opportunity for what has occurred. I’m not passing judgment on anyone personally, and actually have a large degree of respect for most of the team, some of whom I am now proud to call friends. Rather, I have come to the conclusion that those who acted to rectify the wrongs did so too late and after too much sordid history to have any reasonable chance of saving it. This is a testament to what happens when a group of people manage to successfully create something bigger than themselves without ensuring a system of transparency and accountability necessary to serve as a check and balance on individual behavior. For that, and for the breaches of trust that resulted therefrom, the project earned much of what it got in terms of the strict requirements to continue functioning in Second Life. Notably, most of the team rather gracefully accepted that fact, sucked it up, and put nose to the grindstone to try to accomplish the impossible. However, there can come a time when the punishment is no longer reasonably related to the “crime.” I have witnessed Linden Lab change the rules along the way and go too far in their demands, and it became patently clear that this was not simply about user security or trust.
Finally, I come full circle back to the forest. While the vast majority of you are still bickering and debating the trees, Linden Lab has parked bulldozers outside the forest. Let’s examine this a bit closer, shall we? The Lab's original reasoning for requiring Emerald to remove emkdu from its viewer is that it is not GPL compliant:
The Emerald viewer’s closed source emkdu library is not in compliance with the GPL. Bring all current and future versions of the Emerald viewer into compliance with the GPL by omitting emkdu. Use OpenJPEG or other GPL-compatible code.Despite this issue never being raised before or during Emerald’s listing on the TPVD and the reasoned disagreement between those intimately familiar with the GPL as to the accuracy of that assessment, I can still swallow this as a “legal” excuse for preventing what might otherwise pose a security risk to users. Keeping in mind that the emkdu file was used to gather private information from the user (full path names), it is fair enough and understood, although I’m unclear as to why the Lab couldn’t simply use that fact alone as the reason for their requirement. That is, of course, unless there are other motives behind the stated justification that go beyond just Emerald.
After issuing their written list of requirements to Emerald and just days before the deadline for compliance, Linden Lab changed the rules. The Lab informed the Emerald Development Team in a meeting that in addition to removing emkdu from the installer package, the viewer code itself must be re-written to remove the calls necessary for the viewer to use any such file, including llkdu. In other words, for those of you who have enjoyed the ability of copying and/or downloading llkdu and dropping it into your chosen viewer installation, the llkdu file would simply no longer be used by the viewer because the code wouldn’t “see” it. Certainly, Linden Lab isn’t concerned about viewer security in utilizing their own Kakadu library, so what is the justification for this new rule? The Lab claims that upon further review, they seem to have “misinterpreted” their Kakadu license, and that the license does not permit distribution of llkdu. Accordingly, Linden Lab informed the Emerald Development Team that this new coding requirement is going to be added to their TPVP. In other words, every TPV will be prohibited from containing code that calls up llkdu for use and will only be considered compliant with the TPVP if it’s using openjpeg or “other GPL-compatible code.” Phrased another way, TPVs containing code that calls up llkdu for use could be considered illegal and malicious viewers, subject to banning. For those of you who question the accuracy of this information, I can only tell you that the members of the Emerald Development Team who were present at the meeting with Linden Lab separately confirmed this information. Notably, this includes members who were on very opposite sides of the conflict that ultimately resulted in the end of the project.
From what I understand, many TPVs have relied upon Linden Lab's distribution of llkdu to improve their rez times and texture handling. There is no question that for some, being forced to use openjpeg in their viewer will not make a huge difference in their SL experience. However, there are many users, myself included, who find it excruciatingly difficult to function in the SL environment using openjpeg, which is not prepared to efficiently handle texture rendering in many ways. Countless users find themselves living in a mostly grey and clouded world, while others suffer with completely corrupted textures. Just ask the rest of the Support Team about the barrage of problems we encountered from users who were running openjpeg after the release of Emerald Viewer Beta 2587 and how easily most of those problems were remedied with use of emkdu or llkdu. From my experience on the Support Team alone, I could write you a list longer than this blog post.
There was a time when the discussions were properly focused on Emerald, what transpired, and what would be done to remedy or deal with the breaches of trust. That time has now passed. With the much anticipated blocking of Emerald Viewers, its users are clamoring to find an alternative viewer they can live with. The vast majority of Emerald’s users came to discover it because of dissatisfaction with the Linden Lab viewer, a movement that increased to a large degree after the release of Viewer 2.0. Therefore, the thought of going back to those viewers has caused sheer panic and chaos. Many are attempting to quell the panic with the comfort that there are several other TPV options available to them, even if not entirely comparable to Emerald. Are you seeing the forest yet? While features are certainly deeply important to users, it’s almost undeniable that the availability of the lldku file to offer comparable or even improved rez times than that of the Linden Lab viewer has become somewhat of a foundational level of functionality for TPVs to build around....but the bulldozers are parked outside.
There is no question that Emerald opened the door to its demise, but if you think for a moment that Linden Lab isn’t milking this opportunity for all that its worth to rid the grid of its competition, I’m afraid you may be sadly mistaken. If you do not start asking the right questions now, the bulldozers will have flattened your chosen viewer before you ever saw it coming. Even if you think I’ve completely bought the farm on all of this, before you pass final judgment, I implore you all to please take a moment and ask yourselves and others the following questions:
- What legitimate user security purpose is served by requiring Emerald to remove a call to llkdu, when that is the file users are to trust as it belongs to Linden Lab?
- Given that Linden Lab has permitted the use of llkdu by other TPVs for quite some time, why are they suddenly interpreting their Kakadu license to prohibit it? I, for one, would love the opportunity to read the relevant portions of that licensing agreement.
- In light of the potential change in the TPVP regarding llkdu, will other TPVs be permitted to license their own Kakadu libraries as Emerald did with emkdu? (Remember that Linden Lab now claims the use of such closed-source libraries would violate the GPL).
- If, in fact, the GPL prohibits use of such closed-source libraries (I’d start questioning this position as well by the way), does this mean Linden Lab will now also prohibit TPVs from using fmod and vivox for media and voice? If not, what makes the Kakadu library different from other permitted closed-source files?
- If you cannot get reasonable and satisfactory answers to the above questions, then you really must ask yourself, what exactly is Linden Lab's motivation at this point with respect to TPVs and do their actions coincide with their public statements in support of them?
If we decide to stop distributing llkdu and/or make changes to our Third Party Viewer policy, we will make an announcement via the official SL blog.
Second recent development--the just released Phoenix Viewer does not contain code that calls up the llkdu file, because they were advised of the intended change to the TPVP as I indicated above.
If at this stage, you do in fact see the bulldozers outside the forest, what are you going to do about it? Don’t ever lose sight of the fact that although we would have no Second Life without Linden Lab, they too would have no Second Life without the users. However, Linden Lab can never be effectively called to task unless and until the users are willing to remind them of what a vacated Second Life would be like. It is, after all, a business and bottom line matters.