The Mesh clothing parametric deformation project has met its funding goal with some forty days to go, now it will be interesting to see what happens here. I blogged about this project earlier this month, basically this is a project designed to enhance the mesh clothing experience but comes at a time when Linden Lab were seen to be stalling on the issue, Jira SH-2734 is where this was discussed.
I’m still unsure about this whole idea of paying someone else to do work when Linden Lab may come up with an alternative solution, but the gauntlet has been well and truly thrown down now. I hoped at the time that some communication would be forthcoming between the team behind this project and Linden Lab themselvesI have no idea if that has happened.
Linden Lab are put in an awkward position by projects like this, they may well have something else in the works to address this issue but have other priorities, supporters of the project however, clearly didn’t feel Linden Lab were showing enough will to make those changes, hence why the project in its fundraising part has been so successful.
I still feel Linden Lab should be opening a dialogue, as I said, they may well have done, I don’t have any insider information, but it will be a great shame if Linden Lab and the project are working independently on solutions here, politics can and should be put aside for the greater good.
Now we will have to wait and see what transpires, hopefully it will be positive for Second Life and Mesh clothing, but I have these nagging fears that this may all end in tears, hopefully it all ends in smiles.
Seems to me like LL would be fully aware of a project of this magnitude. The developer community is a pretty tight knit group. If LL should choose to keep their mouth shut all the meanwhile working on an opposing project, it would certainly not be taken lightly by anyone.
I would imagine what would happen would be more along the lines of Emerald developing jiggly parts and suddenly taking too much of the viewer pie for the lab to feel comfortable with. Next thing you know we have official jiggly parts.
Oh they’re definitely aware of it, links from the Jira and forum were removed, which wasn’t in my view done out of spite but more because they just don’t allow advertising and links to external sites like that and making an exception here could have opened a right can of worms.
The jiggly bits is a decent example, although there was no crowdsourcing for that! LL’s solution probably was superior, but the ball was got rolling by Emerald.
Few points worth making:
* The work has been undertaken in co-operation with LL, rather than in “competition / isolation”. LL have indicated that they are likely unable to pursue a solution (although they’re looking into alternatives), but they’d be willing to consider adopting the developed code if a) it does what it says on the box; b) Karl is willing to sign the Code Contribution Agreement (and he has stated he is perfectly willing to do so). Obviously, this doesn’t guarantee the code will be adopted into the official Viewer – but it’s a positive statement.
* Code contributions to the official Viewer are not a new thing, hence the Code Contribution Agreement. LL have a long history of considering and adapting (even improving) on code from external developers (multi-attach points being a case in point). The only real difference here is a professional code with strong insight into the Viewer and mesh (he developed the initial mesh code for LL) is working on the first step in a solution to a flaw impacting the creative use of mesh.
* This project doesn’t start and finish with SL / LL. OpenSim now has an SL-compatible implementation of mesh rolling out; ergo, this code stands to benefit OpenSim as well as SL, and TPV developers working on either / both platforms.
* The work is really only an initial step; it is possible that further projects and enhancements may well gorw from it; some of these may well be LL-driven, others may equally-well be driven out of the TPV / OpenSim environments.
Obviously there are issues around crowdsourcing, given the way that development work on Viewers, etc., has been undertaken to date – but one might argument that that debate is more suited to what is happening with Kirsten’s Viewer (if at all).
However, where the parametric deformer is concerned, and the groundwork that has been carried out by Maxwell Graf and Karl Stiefvater, this would appear to be a win/win/win situation for mesh, SL and OpenSim.
As long as there is communication rather than competing solutions to the same issue then this has potential to be good, my main concern was regarding there being a lack of communication but you’ve pointed out that this isn’t the case.
The solution proposed by Linden lab was a non-solution involving the use of proprietary Havok. The solution involved with the crowdsourced solution involves something that even OpenSim can comfortably use without hefty licensing.
Thanks, I haven’t seen any proposals for the solution LL may have looked at one day, the additional information Inara and yourself have provided in the comments here make this look more and more the best solution to the issue, which is good because I’d hate to have seen people raise money for this and it end up not being viable, of course we still have to await the finished solution but it looks likely that it is the right way to go.