Someone killed LSL Warping hack

I have a product base partially on the llSetPrimitiveParams warping hack, that seems no longer to work after this server-side banging 3/14.

Is that one of the many Linden Lab coding glitches or a desired weakening of scripting language?

Most likely they fixed something that was technically broken, but not actually causing any problems (much like when they killed the llSitPos teleporters). When people actively use a bug in code, for non malicious purposes, it ceases to be a bug, and becomes a feature, LindenLabs has yet to understand this fully. My teleporters are broke too. Submit a bug report, and tell all your customers to do the same. Let them know, that 'fixing' llSetPrimitiveParams, was a mistake!

If Linden had actually documented what the functions were supposed to do, they would have a leg to stand on. They didn't, so they're stuck dealing with the resulting mess. The LSL documentation is amateurish and woefully inadequate. As someone who was a principal engineer in a company that sold libraries that had to be documented for customer use, I would have been embarrassed to deliver such flimsy stuff.

LL gets away with it because it's "just a game", and more importantly because we scripters put up with it. (We have little choice.)

Anyway, they made their beds so now they have to lie in them, and honor unintended functionality or else deal with a large community of unhappy customers whose objects no longer work. Note that this means that the more objects that use the unintended side effect, the safer you are. If you exploit a side effect that's not used by others, you're likely to be out of luck when things change.

BTW, I don't think that LL is a bunch of losers. They've created an incredible world, hidden frighful complexity behind a relatively simple interface, and made a lot of very dicey decisions very well. Too bad their good judgement fell apart when it came to documenting LSL. Frankly, I wouldn't even accept that kind of work from co-workers for in-house API documentation. I'd reject the review after the first few pages and say "Come back when you've actually written the documentation."

Agreed. I love SL!!! I am pretty stressed about the stuff that breaks, such as what happened today (and I guess I have been pretty vocal about it), but I still have much respect for everything LL has done that has gone right.

They say they're gonna fix it, but, I got a better idea.

Why not just get rid of the 10 meter limit, when the llSetPos() destination is in <0,0,0> to <256,256,768>?

That'd have the same effect, and cut lag.