Wayfinder’s Recommendations to Linden Lab– A Challenge to Get it Right 

For quite some time (about 3 1/2 years now) I and others have been regularly making recommendations to Linden Lab of ways to improve their system.  Some of those recommendations have been adopted (Estate Tools, Group Tools, the ability to invite people to a group directly from their avatar).  Other good things have been added by LL themselves.  But many very-much-needed things have been ignored, while exhaustive time has been spent on eye-candy, lag-inducing additions such as flexis, sculpties, voice, videos and most of all– Windlight. *

There are far more important things to consider, things that are essential not only to everyday use, but to growth of Second Life itself.  As a user of Second Life who is online every day, who uses the system everyday, and as the founder of one of the largest fantasy groups on the grid (in other words, someone with in-world experience out the ears), I make the challenge to Linden Lab to address and implement these issues with all due haste, for the good of both its customers and the good of the company.*

At the top of that list we find…

STOP THE MEMORY LEAK.  There has been a severe, graphics-related memory leak on SL for more than a year now.  It causes regular crashes, memory overloads and severe lag.  Any company worth its salt would have stopped such a leak at breakneck speed.  That this has continued for over a year, to be frank, is inexcusable.  Call in a third party consulting firm if you have to, but THIS NEEDS FIXED.

200 GROUPS. Current group limit of 25 is totally inadequate. Preferrable: plan for future growth and change the group-membership database from static to dynamic nature, allowing unlimited groups.  In the time I’ve been online SL has gone from less than 400 sims to over 20,000.  25 groups is not nearly enough and before long… perhaps 200 won’t be.

FAST TEXTURES. Goal: no individual texture at any time should take more than one second to fully rez. Region textures should bulk-load on an even faster basis, with a goal of maximum sim load time 1-3 minutes per 1000 textures. A texture requires 100k or less dataspace. It should load across the internet in the blink of an eye. The current rez rate of 30 seconds or more is inacceptable performance.

STOP INVENTORY LOSS.  SL is at deep-core, a database system.  Losing data is the #1 crime of any computer programmer, systems director, or company.  Inventory costs your customers money, building time, effort.   Can you imagine some huge department chain suddenly losing data on Accounts receivable?  If McDonalds lost data on their thousands of franchises, do you think there is a chance someone somewhere would lose his/her IT job?  You bet.  Data loss can be prevented.  People are losing inventory on a daily basis.  We receive “object missing in database” messages on a regular basis.  Needs to stop.

FIX GROUP CHAT. Stop the timeouts and failed discussions. The inability to perform CHAT properly is simply a disgrace to Linden Lab programming ability. As has been stated before, “If a company can’t get simple chat right, what can we really expect?”

FIX GROUP NOTICES. Group notices should be delivered to all members without fail. The “opt out” option on group notices should NOT be the default, nor be able to be trigggered accidentally. Again, the inability to send a simple notice to a simple list of group members is shameful.

MEGAPRIMS. Remove the 10m prim limitation and expand it to AT LEAST 64m (256m would be far better). Megaprims have been tested, they work, they are not detrimental to the system if used properly (and if old-wives-tech-tales are properly ignored), and they ALREADY EXIST in several forms. So why the excessively-cautious attitude? Make them officially available so we builders can get down to some serious building.

REMOVE ARBITRARY LINK RESTRAINTS. There’s no need for such. Link restraints don’t stop griefers;  all all they do is restrain legitimate builders.  We would love to be able to build a 64m castle or an 80m spaceship and have it all linked and flyable– in one piece. If both link restraints and 10m prim limits were removed, we would see some amazing builds on SL.

IMPROVE FLIGHT. The inherent ability to fly higher than 200m. Users are very tired of having to wear “flight feathers” and similar devices just to fly.

SCRIPT TIME TOOL. Allows builders, sim owners or anyone to measure exactly how much script overhead time is required by an object. Takes into consideration the actual average-operation-time of the object, based on all scripts within the object. Purpose: to help reduce region and asset-server lag. Script time should offer two measurements: server time, client time.

SPECIFIC ROLE NOTICES AND IM. Groups have need to contact specific roles within groups without the entire group being notified. SL is currently sadly lacking in communication abilities– abilities that exist in other worlds far less complex than SL.

DETERMINE PRIM TOUCH. The ability to determine on what face and where on that face a prim is touched.  This would come in extremely handy on prim-based menu systems (such as multiple teleport locations) and control panels (such as spaceship panels and HUDs).  Imagine being able to create a multi-function HUD in one prim, or a multi-location teleporter in one prim.  (NOTE:  This is actually being implemented at this time.  Good for Linden Lab.  Waaay past due.)

I will likely be adding to this list as time passes.  Look for a new date at the top of the list.  Again… there is little time to implement such issues.  Competition is not only coming– it’s already here.
 

* Fact: while sculpties and flexis are visually nice, both greatly increased the “lag” time of an already lag-ridden platform. While Windlight and Dazzle are appreciated by some, just as many rue the day they were brought to the system… and thousands left Second Life as a result.

None of the additions I’ve recommended above are “eye candy” features. They have all been needed for a very long time and have taken a back seat to time-and-resource consuming eye-candy features such as Windlight.  Windlight required thousands of man-hours of programming, hours that could have been spent on far more beneficial things.  Adding eye candy on top of an unstable, buggy platform merely adds to the instability and bugs.  All of the above suggested features are very worthwhile, would greatly improve the SL experience, and need to be implemented not a year or two years from now, but NOW.  SL is about to come up against the strongest competitive market anyone could ever imagine. Linden Lab needs to get on the ball NOW if they are to have any hope of retaining a decent market share.

Summary: We have the environment.  We have the foundation (even if somewhat unstable).  We need the tools and features to make SL work like it should be working.   It doesn’t make sense to spend money on toys when the table is bare.  The SL paying customer base is not growing.  In fact, as of this date, SL has lost 7% of its paying customer base over the last 7 months (easy math: that averages 1% a month).  How much more of a loss will it take to convince LL that it needs to pay more attention to customer needs?  We need improved functionality, useful tools, added VALUABLE features and elimination of unnecessary restraints.  Only by catering to customer needs can Linden Lab hope to retain those customers.  Otherwise, the eventual outcome is predictable.

–o–