A better choice would be to make improvements to the grossly innefficient programming language that is LSL, to make the engine more efficient, and to provide better education to users so they can design more efficient scripts.

Like, for example, being able to set hovertext, client-side prim/texture rotation, particles, etc from the client-side. Those hovertext/animation/particle scripts take up processing time and memory even when they are doing nothing! As well, implement a way to change those properties on specific linked prims, rather than having to have a linked message-triggered property-change script in each prim.

Got this from the blogs:

This is a process that will take some time. To give you a rough idea of our current timeline, we plan on finalizing what the script limits will be in Q1, then releasing the Viewer tools in Q2 for possible enforcement in Q3. Once we have more definite dates, we will let you know.

So awhile then....

In many ways I'm for this. On the down side - it makes my job harder because I have to go through all of my existing scripts and optimize them - trick is I don't really dare start or do any new projects until I have some means of being able to see my results. If I clean up my code now, I have no idea of what sort of net gain I can make.

Once the transition process is complete, though... this can be very good. Creators will not just randomly slap any old stupid script into their things - they'll have to take the time to optimize it as best they can. (There are a lot of old crappy free scripts out there in wide use that with just a few simple changes would be a lot more efficient). FOr new projects, it will be good too - because you can measure as you go. It'll be little different than prim limits. As it is now - when you shop for things, you look and say, "Do I have enough free prims for this? And if so, is the value it is going to add to my place equal to or better than the number of prims it's going to take up." Purchasing scripted objects will be the same way. And as for attachments - you can "have" all your gadgets and things - there is just a good chance you probably won't be able to use them all at once anymore - but in many cases, we have various HUDs and things on now that we might not use for days - they just sit there for when we do want to use them - and that does add resources.

Things will change a bit, but if everyone is reasonable (both LL and the content creators) it could be good. Now we just need to hope that everyone is reasonable. That's probably where I'm dreaming and not rooted in reality. lol

The way I see it, this will be little different from the Havok 4 turn over. It will highlight those creators who use scripts who are willing to change evolve and even advance, and those who have neither the will or inclination to change. Look at how many SL car makers just up and disappeared, or simply stopped producing new product shortly after Havok 4 was implemented because as soon as those who would evolve, adapt and take advantage of what Havok 4 offered starting to advertise they were. I can say in my opinion the cars that are sold in SL now are in general far superior not just in physics optimization, but in construction as well, with many car makers that are left having also adapted to using Scupties to cut down prims needed for geometric detail and moving away from the pre Havok 4 standard of textured box construction.

So, making the huge assumption that LL will actually do something right, this could be very good for highlighting the best creators, and showing the ones that rest on their morals and refuse to evolve and improve their product. Those who are willing to change how they do things, adapt to what is available to them, will generally be the ones continually trying to do everything they create better.