[Opensim-dev] Scripting and compatibility

[Dahlia Trimble]

My $0.02 on the delays:
Often they are necessary to protect the quality of the experience from the
excessive zeal of aggressive or inexperienced scripters. For example, if the
llEmail() delays were not there, some scripters could run a spam operation
from a region and the region owner may be blamed. Other cases where limits
are useful is in prim creation and motion, where if no throttling is
available a region could easily be brought to it's knees by a script run
amuck. Some delays could actually improve the perceived performance, such as
in the case of prim motion where sufficient time is available to notify all
observing clients that an object has moved before the next movement occurs.

I vaguely remember some discussion about the necessity of the delays and I
thought the consensus was that they would be configurable in some fashion,
but I haven't followed the issue closely to see what was implemented. As a
region owner, I would prefer the delays/limitations were in place for any of
my regions where end user scripts would be running.

On Sun, Sep 7, 2008 at 12:19 AM, Melanie <melanie at t-data.com> wrote:

> The recent commit that removed a lot of ScriptSleep calls that were
> already commented prompted me to look at which ones are even still
> in there.
> I found that in LSL_BuiltInCommands, most of them are commented, and
> many commented ones have been removed.
>
> The ScriptSleep and it's associated factors had been introduced for
> the reason that we would be able to retain LSL (SL) compatibility
> where desired, yet override the delays where that is wanted.
>
> Removal of the sleep calls, or commenting them, is not really in the
> spirit of that consensus that was found.
>
> Why, even though the factors are there to reduce or completely
> eliminate the delays, are the delay calls being removed?
>
> We should strive to insert, not remove, them, to achieve 100%
> compatibility for those who want it.
> Especially the all important 20 second delay on llSendEmail has been
> commented at some point, even though many existing scripts will
> break in horrible ways if that delay isn't there. The
> llInstantMessage delay is of equal importance, since for both these
> functions, it is customary to build dispatchers in LSL, which depend
> on the delay to function correctly.
>
> Have we now officially abandoned the plan of providing at least the
> option of source code copy/paste compatibility?
>
> Melanie
>
> _______________________________________________
> Opensim-dev mailing list
> Opensim-dev at lists.berlios.de
> https://lists.berlios.de/mailman/listinfo/opensim-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://opensimulator.org/pipermail/opensim-dev/attachments/20080907/ece3c2f7/attachment-0001.html>