Hierarchy in Linked Sets

A quick search of previous posts tells me this has been brought up before, but not much. So in the spirit of squeaky-wheel wishfullness, I bring it up again...

How unbelievably useful would nested grouping be to builders? Forget new building tools -nice as they would be - let's have the ability to create subgroups within larger complex objects that don't get lost when the top level group is undone. It's impossible to estimate how much time this limitation has cost me.

Anyone know whether this is in the works, being dreamed of, or impossible in SL? Or...anyone have a workaround suggestion?

It's not impossible, and I think LL is aware of the need for it

I haven't heard of any workarounds, but there might be by use of scripts?

There is no current way to have hierarchical linking in SL. But the Lindens have been looking into it. I don't have a link handy, but I think it was Blue Linden who posted something on the considerations necessary for making Hierarchical linking a reality, and the work that they had in progress on this subject.

The great question is, 'when, if ever, will some Linden decide this is a high enough priority and that they want to solve this issue?'. Because the way LL operates, Lindens pick their own projects and set their own goals, and there is virtually NO mandated action from the top which insists an individual *will* be assigned to a specific task. A fine structure for a start-up think-tank group in someone's garage, but no way to run a serious corporation.

I've told this story before, but since we're apparently rebooting the discussion here, I'll tell it again.

At the first SLCC, I asked Cory Ondrejka, who was then the CTO of LL, to explain why we didn't have hierarchy, and what, if any, were LL's plans to implement it. He said, "Our original thinking was 'Nature doesn't have heirarchy, so why should we?' That thinking was wrong, and we WILL change it."

That was four or five years ago, and it hasn't changed. Of course Cory is no longer at LL, but that's no excuse, since he was there for three or four more years after he made that statement.

My understanding is that, in addition what Ceera said about individual Lindens choosing their own projects, there are certain things in SL architecture that are such a gawd-awful mess, code-wise, that it would take a complete rewrite of the system, just to implement a very small change. I've been told in no uncertain terms that this is why we still don't have generic outfit layers for the avatar, for example. I would imagine it's the same with link heirarchy. The current setup is probably too deeply embedded for anyone at LL yet to have mustered up the motivation to tackle it completely. It's likely far more monumental than any of us on the outside would imagine it should be.

In many ways, we're still feeling the sting from the fact that SL was just a hack job in the beginning. It was thrown together by just a small handful of people, with a very different purpose in mind than what it eventually became. It didn't matter back then that it was held together with Scotch tape and bandaids. It just had to be demonstrable. Fast forward to the present, and it is what it is. A lot of the things that seem like obvious, fundamental, no-brainer inclusions for any 3D modeling application just aren't there, because it just wasn't initially designed to be what it now is.

With that in mind, I think it's important to appreciate what an absolutely amazing job of fitting square pegs into round holes LL has been doing for the past several years. But even though I know that, I've still got a frustratingly long wish list, just like everyone else. And hierarchy is definitely near the top of it.

All very good; thanks for the replies. I suspected this was the status, but heck...why not rattle the cage occasionally.

I do agree, by the way: having done a fair amount of 3D modeling in (yet another) life, what LL has been able to put together to make content creation possible for absolute beginners is amazing. A two-tiered system of creation tools - one for novices, one for more ambitious builders - would be swell, but certainly the democratic impulse should rule. If keeping it simple makes SL more fun for more people, then that's how it should be.

But I still want hierarchical grouping. Still want it bad.

MISC-428 - Hierarchical linking of prim/objects

MISC-2237 - [hierarchical linking] Requiered additions and changes to objects and scripting, for improving content-creation capabilities.

http://jira.secondlife.com/browse/MISC-428
http://jira.secondlife.com/browse/MISC-2237

Heh, I knew Shack wouldn't be able to resist posting in the thread.

But yeah, Hierarchy seems to be a big problem to squish into the code-mess and would mean potentially significant tinkering with some of the existing architecture. I seen the client code myself and I can't even begin to fathom how much harder it would be to make hierarchy work in the sim' side.

For those scripters though, I had dabbled at implementing a working skeletal hierarchy script for linksets and opensourced it: https://wiki.secondlife.com/wiki/Hierarchics

Very cool.

I need to look more closely, but this will probably save me a lot of time. One of the tasks near the top of my list for Prim Composer is to implement a Builder's Buddy type thing that works with hierarchically grouped prims.

It looks like you've already done the heavy lifting. Thanks for sharing!