Number of Groups

These quotes from the second meeting

Robin Linden: First, looking at general characteristics (size and number allowed)
...
Eloise Pasteur: You said in phase 1 raise number of groups? Roughly how high are you thinking?
Kamael Xevious: That's what I mean and hallelujah!
Robin Linden: I'm not sure yet Eloise...it depends on what the devs tell me with respect to the database hit

and from the third meeting

Robin Linden: Jana I want to increase the number of groups.
Robin Linden: However it's a hit to the database so the question is how many to allow.

and from the fourth meeting

Robin Linden: Groups are database intensive

all has me thinking that if our inventory were reasonably capped, if squids were allowed only a given number of textures thus limiting us per sm, and a few other reasonable database limitations were imposed, the asset server could afford us a few more groups per av. LL has given up the hippie notion of supplimenting builds and events that, in the end, don't represent self supporting uses of the software (killer apps). Internet bandwidth is not free after all; LL are obviously beginning to understand that the general public are beginning to realize that.

Might LL also believe the population to be grown up and savvy enough to understand that a centralized asset server has it's limitations as well? Does it seem reasonable, at this point in time, to tell the population "you can join more groups if you'll agree to limiting your inventories to 15k items and your texture applications to thirty per 16sm?

I'm making up numbers of course. Working values will require some inhouse analysis. My questions are whether or not such limitations will free enough resource to afford more groups per av, and whether or not LL believes the population can swallow them.

This might not be the most appropriate place to bring up the question. And I am not good at outlining my thoughts so I hope I can write them all out in a cohesive enough manor to be understood.

This post (and a few others in the economy/land area) reminds me of a question I ask myself all the time.

What is the REAL difference between the basic account and the premium? I was a premium member when I first started in SL, then after a time it made no real sense to me to remain one. After about a year of being a basic member, I decided to offer up my “premium member” money to the lindens. I don’t own land; I know I could… but renting works out much better for me. But I felt that I loved SL enough that I really needed to put some money where my heart was. I now am a premium member.

But what did it get me? Nothing really other than the ability to buy land and stipend.

I think I understand your idea Kahmon. Finding a way to limit a members usage of the resources here in SL in a fair way. But based on what?

Does it seem reasonable to limit basic members inventory sizes, number of uploads, groups? We already have a limitation on land owned by basic members, why not more than that?

I don’t have much of a computer mind so I don’t know what kind of coding would be needed to implement something like this, but in all honesty I truly believe that premium members need to be differentiated from the basic members in a real quantifiable way.

Theo you have hit the head on the nail we have 10' of thousands of basic members paying virtually nothing to linden labs to use a service. in the same exact manner as premium members can with the exception of owning land.

with the on slot of rental land and mall rentals there is not any real motivation for basic members to be premium members except for owning land and the sense as you put it

and by the way Theo you laid out your thought wonderfully!!

I might be more interested to find out if, while changing how we use groups, they can find a way to make them more server-friendly and efficient. It may be that the early design for groups just didn't take into account how heavily groups would be used. They DO admit that they thought groups would be only used for "book club" type organizations, and were caught off-guard by the advent of big bussiness in SL.

So here's hoping the "one dev" they have on this can really turn things around.

This type of discussion keeps resurfacing, because asset server storage, retrieval and update constitute a scarce resource critical to customer satisfaction with SL. To have a meaningful discussion we need a good solid approximation of the actual performance stats. Is inventory access loading the system far more than textures needed to render an avatar's surroundings? Or vice versa? How does that compare with storage/retrieval/update of groups lists, calling card lists and friends lists?

If these stats have already been posted somewhere many of us would appreciate the link.