Some Pointed Questions for Linden Lab

As a merchant on Second Life with over 5 years experience, as a member of several large groups, I have a pointed and direct question to Linden Lab:

Why is it that after more than a year, group chat and group notices are not only malfunctioning-- but have become increasingly and significantly worse, to the point that they are for all intents and purposes non-functional?

These matters have been repeatedly reported on pJIRA. They have been voted on extensively. People have literally picketed Linden Lab office hours over these matters.

I have read claims from Linden Lab employees that such are "complex issues". I wince every time I read such questionable statements.

Chat rooms have been doing chat for ages. Online games regularly perform in-world chat, flawlessly. Companies have been incorporating complex chat for over a decade (some far more complex than anything seen on Second Life). Group chat once worked. So who broke it? And why, after breaking it, was it not immediately fixed?

Group notices are not complex by any stretch of the imagination. Flowchart: Go through the list of group members. If the member is online, deliver the notice. If the member is offline, send an email notifying that member of the notice. This is not a major programming puzzle. To be somewhat blunt, sending out a notice to each person on a list is taught in first year computer programming.

So again I ask: why, after more than a year, have these issues not been corrected? They are making it difficult (if not impossible) to do business. They are making it impossible to run groups in any kind of organized manner. Communications are the heart of any company.

Why has Linden Lab not fixed these stowstopper-level communications problems?

per JIRA issue SVC-1507: Notices failing for larger groups
Type: Bug
Status: Reopened
Priority: Critical
Assignee: Unassigned
Votes: 335
Watchers: 28

Apparently, since this issue is "reopened", not "in progress" and still "unassigned", whoever Linden Lab has monitoring and assigning Lab Devs to JIRA issues doesn't prioritize this issue like you do.

I have seen JIRA issues that get votes get attention. So go vote: https://jira.secondlife.com/browse/SVC-1507

I have seen issues that folks call the concierge and technical support about get a dev bump. So treat it like a personal technical issue and make a call to your account support team when it happens to you.

Doubtful that posting to this forum will bring the Linden attention you wish to the issue. (A Linden post here would utterly astound me). This issue has been here before. The JIRA discussion serves as its own forum thread. But this thread here can't hurt, unless the thing becomes a flame or distracts from more workable communications.

There was a massive blog entry just a couple days ago addressing the whole group chat thing. My interpretation of what they wrote is that they accepted that the current situation was a total calamity and that they were re-engineering the whole group chat thing from the ground up to make it far less retarded.

https://blogs.secondlife.com/community/technology/blog/2009/06/15/improving-the-quality-of-group-chat

I agree that getting it right isn't hard. There's only, like, 60 bajillion other "group chat" applications out there that work nigh on perfectly. But you must concede that if they did it wrong in the first place then ripping out all the infrastructure and replacing it with something sensible has got to be at least a little tricky.

Arth: Well, this issue is calling to task LL's failure to deal with serious show-stopper issues, not whether or not they pay attention to forums.

JIRA is pretty much a heavy-handed, heavily-Linden-edited system that obviously has not been beneficial in these areas no matter how many "votes" they get. This forum is an additional area where users can publicly state their opinions without the heavy-handed employee interference of JIRA, and where we can call to task LL employee claims that are (in come cases) totally bogus.



Elanthius: Note from that LL blog:

"Several months ago, we received reports from residents that Second Life's group chat feature was not performing well. "

Several months ago? That's what I mean by spin. Check the JIRAs LL. It's been significantly more than "several months"... and significantly longer than any such fix should take. Communications are the heart of business. You don't lolly-gag when it comes to fixing communcations.

They further claimed that the "old code wasn't scaling well". OK, I can buy that. So the moment you realize the old code isn't scaling... you FIX OR REPLACE it before it becomes a total disaster area. Waiting until the code has gone down the tubes is not responsible system management.

Yes, tearing down the old system and putting together a totally new one might be a bit tricky. What I'm calling LL to task over is taking more than a year to decide that this needs done. I still question how it is that chat worked fine for years, got broke, and subsequently LL failed to fix it after a year of protests. It's a shame it requires people virtually camping on their front lawn and invading their office hours en-masse to even catch their attention in such matters... and it's a shame that group chat had to go from occasional timeouts to total disaster before Linden Lab decided that yeah, this just might be a problem.

We have seen repeated claims from Linden Lab before about "we're working on it", "it's not easy", "it's complex" and promising fixes. Well, after over a year we're tired of promises and spin stories and claims. These are major issues that are making merchanting and management of groups on SL virtually impossible. Customer patience is not just worn thin-- it's GONE. We are tired of hearing that it's been turned over to "working on it Linden". We don't need results ten excuses from now, next quarter, next year... we nee this fixed NOW. We need them to put their core, key people on this, burning midnight oil if they have to, because this is affecting the very core of our investments.

So I'm saying as businessman to business: Linden Lab, get your act together. Stop procrastinating, stop the frivolous "feature" projects, stop messing with trivial menu changes. Put together a list of "top 20 major issues on Second Life" and put every coder in your employment on bug fixing NOW until these major issue bugs are history... then start working on the hundreds of other reported bugs and fix those... and don't bring out a single new additional feature until your product is acceptably functional.

One other post on that blog:

"Because we have a crowded server release schedule coming up, it will be a few weeks before these changes are live on the grid."

A few weeks. They're going to let chat be totally borked like this for "a few weeks" without putting out a new release with the supposed fixes they've made?

Imo that is irresponsibly sad. As Arth said, it's pretty obvious Linden Lab has not considered this issue a serious priority... which indicates to me just how out-of-touch they are with their customers. For us, this is likely the top priority on the grid right now. It just amazes me that Linden Lab hasn't seen fit to push this to the top of the priority list themselves.

Electron Cleanslate: ""timely"??? This issue was open in the Jira on 27 November 2007. For the benefit of your loyal SL users please explain your definition of "Timely""

Fred Rockstown: "When a bug as disruptive as this one pops up, you should all drop your pet projects and work day and night to get it fixed. Only when it's fixed and QA has approved it for deployment, you can go back to working on your own little projects."

I won't bother quoting the dozens of other highly negative responses to Linden Lab on that blog. Needless to say, customers are ticked, mightily.

I'll tell you exactly why this isn't getting fixed in a reasonable amount of time.... there are a few issues at hand here.

Issue #1: The whole group chat system is flawed. It's not scaleable, it doesn't work well at it's heart, and the only option is to rework the entire group system from scratch.

That in itself isn't really that big of a problem, but it brings us to...

Issue #2: Everything in this silly game is tied into that massively flawed group system that needs to be redone from scratch. Ownership (objects/land), permissions(objects/land), accounting (as it related to group liabilities and income), and so on. Virtually every little thing in there was hooked into the group system in some way.

Thus, it's not a matter of just redoing groups and calling it good. New code means new hooks into everything.

That said, the solution could be rather simple if the old groups system that is the heart of how SL all "sticks together" stays the way it is, but a new system of "social groups" (give it another name without the word group in it, please - and social probably isn't a good word either... but... anyway)... these "Social Groups" would be set up in a way designed specifically for group chat - be it open or for product support, for sending announcements, and so on. Being in these groups is separate from anything else and thus it can be redone.

if I want to buy land and give people permissions on my land or other things along those lines, they need to join by "Old Style Group". If I want my customers to be able to chat amongst each other, receive update notices, get support, and so on, they join my "Social Group".

It's not a perfect solution by any means. But it is viable and could be done. It's not at all surprising that LL hasn't tackled the group issues as they are now - if they do, they have to rework virtually every hook and every aspect in the game - and it breaks a heck of a lot of stuff too.

(TIP: Getting a new groups system in place that is separate from the old groups would also allow both to operate side by side and, over the course of a year or two, you can redo the various hooks of ownership and rights and begin to phase out the old group system over time).

Lemme make this real easy for you.

Linden Lab does not care one bit.

They're bringing in loads of uninformed people with every new feature addition, and they get money from these newbs AND from the people who still pay them while their services fall apart. Why spend money maintaining it if the old users are still paying you? This is why I downgraded to basic today - LL is undeserving of my money.

For about the last month, I get only one line of group chat then it automatically closes itself. How much I've missed, I can't say but, obviously, very frustrating.

I am afraid you're right Fred. LL has consistently made decisions based on imagined profit rather than customer satisfaction. As a result, after six years, their profit line isn't so great. Compared to other online activities, Second Life is small peanuts.

LL has put profit as job-one even when it meant extreme harm to their customers (witness: Open Space sim fiasco). They always take the easy way out: they just recently limited "takeable" items to 1000 prims (as opposed to fixing the core root of the issue). How many problems did that cost customers? (I know it cost me a bundle of headaches.)

Linden Lab seems to be unaware of the fact that the way to significant profit is to create a product that actually works, rather than new flashy toys that just further bog down the system. The way to score more customers, satisfied customers, paying customers-- is to gain a good reputation by having a good, solid platform and a reponsive support team (as opposed to attitude-laded drama-queen employees). The way to majorly increased profit is by having a reputation as a company with a rock-solid product-- not a reputation as a bug-infested virtual brothel.

So I have another question for Linden Lab:

How is it that your priorities are not the same as customer priorities?

Company priorities should be identical to customer priorities. We're paying the bills. When our priorities cease being met-- as seen above, we stop paying the bills.

Like you Fred, I refuse to spend a dime with Linden Lab. I once had a premium account and a full sim. I cancelled that status and went back to basic for one reason: Linden Lab attitude and failure to meet customer needs.

Point on "scaling":

OK Linden Lab claims they have trouble with group chat due to "scaling" issues. Ignoring that their scaling issues should have been addressed four years ago when the company started growing...

How does scaling in any way affect delivery of a simple group notice to a list of people?

That is a singular process, a very basic transfer of data. Read the names from the list, send the notice to each member, if they're offline deliver an email.

In what way does that involve "scaling" by any stretch of the imagination.

What, there weren't 500-member groups 5 years ago? Or perhaps the new multi-core multi-gigabyte RAM computer systems are less powerful than they were 5 years ago? The internet is less efficient and communications lines are slower than they were 5 years ago?

So what is it about sending a simple notice to a list of people that is so amazingly difficult that Linden Lab can't seem to accomplish that Computer-101-level task?

Scaling my tooter. That excuse is as old as Second Life. If a company has "scaling issues"... the wise course is to address and fix those issues before they become major catastrophes. Failure of chat grid wide is what I'd call a major catastrophe. It makes Linden Lab programming staff look like a bunch of junior high schoolers. Correction: these days many Jr. High schoolers do a better job at coding.

So again I ask them the pointed question: What are they doing over there at LL that they can't accomplish such basic tasks as group chat and notices?

Almitra: It's the same story all over. The result: most of the major groups with whom I work have become frustrated and just stopped trying to use group chat. They're not holding events, because they can't announce them.

My merchant friends and associates are reporting all-time record-low sales. May merchants report not even being able to make rent. They can't announce new products, they can't host merchant events, they can't keep in touch with their customers.

Linden Lab is destroying the grid. Their focus on toys and failure to solidify the platform is finally coming back and biting them in the tail. Unfortunately, it's biting their customers as well.

When large groups and merchants can no longer function-- what does Linden Lab think that is going to do to their land sales and company profit line?