|
Yumi Murakami
DoIt!AttachTheEarOfACat!
Join date: 27 Sep 2005
Posts: 6,860
|
04-12-2006 10:45
Dear Linden folks,
The use of multiple script process farms to bypass the inbuilt time delays on LSL functions and the memory limit on scripts, is now well known and documented. A Linden user has acknowledged that there are "ways around" the delays, the general solution is well-known, and it can be easily obtained by someone who asks for help in a number of different community fora.
I thus propose that we've reached the point where the workaround is doing more harm than the possible abuse that the delays were intended to prevent. Anyone wishing to bypass these limits will have relatively little difficulty finding out how to do so, so the limits will no longer be effective at preventing abuse; and the use of a process farm requires objects to contain large numbers of scripts running in parallel and communicating regularly, thus creating lag.
Would it therefore be possible to remove these forced sleeps and add a llMalloc function to allow a script to request extra memory storage?
|
|
Kelly Linden
Linden Developer
Join date: 29 Mar 2004
Posts: 896
|
04-12-2006 15:16
Extra memory space will not happen before mono integration. I do not know when or if it will happen after mono. It is a significant change to how the current script virtual machine runs, which is unfortunate. Rather than spend significant developer time on a system that is slated to be replaced this feature will have to wait for consideration until after, or during, mono integration.
Removing sleeps is a less clear issue. A review of script sleeps may be needed, and would probably not be negated by the move to mono, however the move to mono will require its own review of these delays itself. Even though the work arounds are known, they do limit to some extent, and it is not immediately clear to me that the amount they limit is less than the harm created by the work arounds. For example, the work arounds require that you more or less know what you are doing; it is probably less likely a scripter will accidentally create a processing farm than it is that they would accidentally create a run away while loop. So while the delays in this case wouldn't stop those intent on bypassing them, the cost (in time and effort) may prevent more casual use and make trivial mistakes less harmful.
_____________________
- Kelly Linden
|