Wagner James Au / New World Notes

4 responses to “Linden Lab Hiring VP of Engineering to Help Define Future of Second Life”

1. 

   Nodoka Hanamura

   March 25, 2021

   On the topic of mobile and streaming, I feel that before they can tackle that, they have to focus on the Elephant in the room, that being performance. I’ve been a passionate advocate for the Lindens to take a step back and rework the viewer to take better advantage of hardware and to fix age old issues such as handling memory and cache.
   And even then, I don’t feel that Second Life can reasonably be made useable for mobile as a full client without dumbing down the platform to the expected standards of mobile, which would only cause SL to suffer as a result. As much as people clamor for a mobile client – in reality, at least for me, mobile games have never been appealing, and on top of that, the lack of screen real estate on most devices doesn’t help SL’s case (or games in general).
   There’s a good reason why gaming on mobile hasn’t evolved much past ports of games from three generations ago and casual games. They’re not only really just timewasters with little long-term continuous play value, but ask me – would you rather play a game on your couch or at your desk in a relaxing chair or on a laptop on your bed, or would you rather be standing, or hunched over staring at a screen as big as your hand? If you chose the latter, then I pity your back (which to be honest is probably far better off than mine). And before you mention handhelds like the 3DS and PS Vita, at least those had larger screens than most mobile devices.
   In reality, while a mobile light client is a good idea, doing what IMVU did by prioritizing a mobile experience is a exercise in wasting resources on something that could be applied to more important issues. On top of that, SL’s viewer code is such a shitshow that Linden Lab is probably having to write a whole new client for SL, that doesn’t use at best, a sizable amount of existing viewer code, and at worst, is written from scratch.
   Linden Lab has essentially made what amounts to 7-layer spaghetti code that has had layers of code plopped on top of or replacing others over the years. We still use a rendering pipeline that doesn’t use multi-core rendering or take more advantage of higher VRAM than 1-2GB outside of certain TPVs who have overridden Linden Lab’s hard limit on that VRAM – Which would be fine if we weren’t living in 2021 and not 2001, and if Mesh wasn’t a factor. The UI is, while I am accustomed to it, is clunky and hard to understand for new users, making onboarding a nightmare. The loading of assets such as textures is a joke, as the viewer fails to prioritize loading assets closest to either the camera or the user’s avatar, and even then, it is horrendously slow to load at times and sometimes takes hard refreshing textures on an object to get it to render properly.
   This isn’t helped by the fact that Linden Lab is quite restrictive on what Officially Approved TPVs can do to mitigate this. While this is reasonable for some circumstances (preventing Hell, some viewers, like Firestorm, are known for having gotten more bloated over the years. Were it not for the fact that it only is available on Windows and it crashes often with WINE, I’d have moved over to Alchemy long ago.
   Look me dead in the eye and tell me that a mobile, full viewer – as foolish as that alone is, will be able to work with SL without them redoing the renderer just for mobile (and hopefully, porting that back to desktop to improve performance if it’s even possible), and on top of that, somehow managing to condense the components of SL into a simple, easy to use interface.
   The best strategy I feel, as much as close friends have questioned it, is to improve performance, try to listen more to the residents’ needs, and focus on Quality of Life rather than cranking out updates that do nothing to mitigate existing problems with the platform, and maybe try to integrate VR into the SL experience within the next 10 years. As VR becomes more mainstream and the cost requirements go down over time, It would be wise to try and compete with next generation platforms (which while I consider them less Virtual Worlds and more 3D Chat Rooms vis-a-vis Worlds.com), Neos and VRChat.
   It says volumes when Second Life runs at poor frame rates in many places, even when it shouldn’t, and yet these other platforms are able to do just fine, in VR no less. This isn’t a matter of “it’s all being streamed to your viewer, so of course it’s going to be a bit slow” – it’s a matter of “Linden Lab has worse procrastination issues than I do, and can’t look themselves in the eye and realize that if they don’t do something to overhaul Second Life not just in their client, but server and service areas as well, Linden Lab is going to be up shit creek with a paddle with holes in it called Tilia.”
   As for streaming, While it failed for Stadia and many others, using SL on streaming platforms would be a good mitigation of the existing performance issues until such a time that they can be remedied properly. For most use cases, it’s handy, but should be a monthly, unlimited solution and not a timed one like Bright Canopy or others.
   But it shouldn’t be used as a crutch. I use a decent gaming PC, with a Ryzen 5 3600, GTX 1060 and 16GB of DDR4 RAM, running SL and Linux both on a NVMe SSD. I should not be having framerate issues in my own house, with draw distance at minimum, at high graphical settings, and I definitely shouldn’t have to dole out cash to properly use a platform that if it were developed responsibly, would have had this problem resolved in some reasonable capacity by now.
   But no, that’s not the case.
   I love SL with every fiber of my being. I’ve made friends, had amazing memories and seen a beautiful, extravagant world that I’ve lived a life in that my real one could never compare to, and I do have Linden Lab to thank for that. But at the same time, I see that my virtual home is being neglected, and while I do like many of the staff from Battery Street, I don’t like how as a collective they’ve failed to fix this mess.
   Something has to be done. Hopefully whoever they hire will be able to put SL back on track.
   Post writing edit: Had a bit of a mental episode (too difficult to explain, essentially shock migraine) just before doing one last check of this. I’m sorry if parts might have grammatical or spelling issues.

   LikeLike

   Reply
2. 

   Mona Eberhardt

   March 26, 2021

   The Lab will need to have a really good look at what Oz Linden had done right during his tenure and hire someone with a strong background in the development of game engines. While SL is by no means a “game” to its users, its engine should be treated as an open-world game engine à la GTA series or something similar: an open world for people to build in and for, for people to take snapshots, for people to create machinima.
   Oz, in collaboration with TPV devs and creative users, led the Lab to advance the engine and the viewer so that certain age-old issues are now either solved or no longer as jarring as before.
   However, there are still features missing from SL, like mirrors. We don’t have volumetric lighting – if you turn on a light within a room and its radius is greater than its size, even if the walls are not transparent, the light will go through them. Linden water goes through solid objects, making it impossible to have underground builds near water level, and don’t get me started on ships, which can’t have decks under the waterline. And so on. We now have camera presets, but the default presets are still the same crap we’ve been saddled with for years – the crap that led to crap avatar design with T-Rex arms, crap builds, crap animations and poses, the crap that ruined SL for architectural visualisation. And the build tools floater really, REALLY needs to be enhanced with capabilities like the ones found in applications like SketchUp.
   SL doesn’t need to be brought “back on track”. It needs to unstick itself from the half-arsed, overhyped thing His Holy Philipness audaciously presented as the best thing since sliced bread, and be brought in line with the long-time expectations of the people who abandoned it because they found its engine and creative tools too primitive and limited.

   LikeLike

   Reply
3. 

   Nodoka Hanamura

   March 26, 2021

   “SL doesn’t need to be brought “back on track”. It needs to unstick itself from the half-arsed, overhyped thing His Holy Philipness audaciously presented as the best thing since sliced bread, and be brought in line with the long-time expectations of the people who abandoned it because they found its engine and creative tools too primitive and limited.”
   You’re absolutely right in this. As much credit as I have to give Philip, he was a bit foolish in how he wanted it to be Web 2.0 and all this jazz. To say Linden Lab needs to quit fucking around and focus on QoL is the biggest understatement on the grid.

   LikeLike

   Reply
4. 

   Clara Seller

   March 27, 2021

   It’s an interesting conversation and a reminder of how lucky SL has been to have so many technically knowledgeable people care about it over the years. Beyond the technical aspect, they really need to hire someone to get the soul of the game back on track.
   Crashing Marketplace over a L$1 Mesh head is cause to weep for all of the Sheeple who have taken over Your World, Your Imagination. Not having that mesh head is the first step in having a head of your own. Stop the madness.

   LikeLike

   Reply