On top of the recent attachment update bug, JIRA (SVC-7283) (as well as the releated SVC-7294) – there is now another scripting breakage that has been deployed to the grid.
PRIM_PHYSICS used in the likes of llSetPrimitiveParams, no longer functions as expected on server release code 11.09.09.240513, leaving scripted vehicles and animals (among other things) not working, malfunctioning and/or running amok.
A JIRA (SVC-7305) has been raised for the issue, which has a widespread impact. And it has to be said that the JIRA is starting to read like a poorly executed farce, with LL responsible for the script.
- It appears one creator of scripted vehicles running on the mainland which are now effectively out-of-control, and her account has apparently been suspended as a result
- Other creators are reported issues with potentially thousands of sold products which, while there is a potential code alternative available (via STATUS_PHYSICS) – this still means considerable re-coding for many creators and a massive product update (and why should they, when the cause isn’t their fault?)
- Responses from Linden Lab are hinting a fix (again) will not be available for RC channel release until next week, with the implication a full roll-out could be two weeks away – and that, despite the clear severity, there is resistance to accelerating the fix within the company!
Two breaks to basic scripting functionality in Second Life coming a week apart isn’t liable to win Linden Lab awards in the popularity stakes. It’s also liable to have the tinfoil hat brigade nodding to themselves and muttering comments about event numbers, coincidences and conspiracies.
As if we don’t already have enough of that doing the rounds.
Inara Pey: Living in a Modemworld 
Inara,
I don’t know what the “Tin-hat” brigade might do, but recent postings by AnnMarie Otoole in JIRA SVC-7305 make it abundantly clear that there is zero internal communication within those offices of Linden Lab concerned with SecondLife.
It is becoming clearer by the day that there is no sense of urgency in folk other that those directly interfaced with the users (Maestro, Oskar and Andrew now), at Linden Lab, and that not only is there no joined up thinking, there is no internal communication there either.
SecondLife really must have a very low priority in general at LL if such disinterest persists through what is possibly the worst crisis in the recent history os SL.
There seems to be no logic, no connection, no interest in going about soving these issues, other than in those individuals that interact with Second Life residents.
I must assume that to most of the staff, the currrent situation is that a software system run by the company has a couple of glitches in it, and fixes will be developed, in accordance with internal procedures, so that “due process” can be shown to have been followed.
It takes absolutely none of the personal irration, inconvenience and damage to business that is collaterally occurring into account whatsoever.
That Andrew Linden should post what he did is an appalling indictment of Linden Lab’s internal processes, and will stand as a marker for the disconnection between SL and its creators.
LikeLike
The tin hat comment wasn’t a knock at those experiencing issues; more at those who string together theories that LL is out to shut down SL via some back-door means.
This is a cock-up of the first degree (even allowing for any code complxity), and needs to be damned as such – and corrected – but as to anything beyond a gross error on LL’s part? Unlikely.
LikeLike
Oh, and incidentally, Inara, SVC-7283 and its links is not merely to do with updates of worn attachments, it has implications that run much more deeply into the permissions system of SecondLife. It was created by a ham-fisted attempt by one on the Lab’s programmers to close off a potential copying route, and was inadequately QA’d.
LikeLike
Yes, aware of that. From what I’ve heard, it was an issue related to getting around the NO TRANS permission setting on objects – speculation I’ve seen reflected in the JIRA comments as well.
The JIRA refers in that case to issues with updating worn attachments, etc., and as such, that’s the easiest way to refer to it. The lack of regression testing was referenced in the original piece on that particular issue :).
As to wider implications – with both of these issues – that could well be the subject of a broader article down the road.
LikeLike
looking at the policy changes regarding premium accounts, I think this is a step towards a two tier platform, premium memebers ostensibly get better sl performance whle everyone else gets left with. deteriorating simulators, already we see that premium members will have “exclusive” sandboxes that are not accessible to other residents( despite tha fact that plenty none premiums have payment info on file),….( anyone care to run a comparative survay of premium and none premium users to determine any correlations??)
LikeLike