Intelligent design is NOT an effective corporate strategy

If past experience holds true, this post will not be read. It's going to be fairly long, regardless. If you have ADD or a poor hand-to-post ratio, just read the bold text.


Let me be honest: I'm freaked out. The way Second Life has been going of late, I cannot see it existing five, ten, fifteen years from now. This has very little to do with post-patch trauma, instead extending to the very foundation of the system.




At core, it's a good idea. "Let's empower the users to build their own content so we don't have to. That would facilitate the most compelling worlds ever created."

This is indeed true; the independant developers of Second Life are some of the best in the business.


But, lately, that goal seems to have been lost. Instead, it has been replaced with a more corporate ambition: empowering the users only so much as it cannot hurt others, leaving the intelligent designers (LL staff) to handle the "big" leaps and reap the big rewards. This is a fallacy that will drive Second Life into the ground if it is not addressed.

"LL knows best" is a false cause. The users MUST be able to alter the system to serve their immediate needs. If they cannot, the system falls into disarray.




This relates to several problems, some of them immediate, others that will not be present until further down the road. Here's a brief analysis of some of them.






Problem #1: Competition within your own system.

The GOM-versus-LindeX battle is just the tip of the iceberg. In the pursuit of profits and expediency, you face competition within your own system if users are only given marginal control. This competition is needless, because users that understand the task far better than you and are empowered to do so will ultimately be better at it.

I do believe, having not traded a single Linden, that the GOM was inherently better at managing resident finance than Linden Lab will ultimately be. Simply, they specialized in their field using dedicated staff members that generally knew what they were doing.

The only benefit of a Linden-driven system is one that should have been present for GOM: The ability to trade within the system. The lack of any ability to modify the client or access resident data hampered this process enough to drive the GOM away, a premature but not quite foolish decision on their part.


The inability to modify the software on an opt-in basis is the root cause of this problem. Had the GOM been able to place a hook into the client, their business would continue to flourish.







Problem #2: Inability to please everyone through patches.

Linden Lab employees are human beings. Being human, they have limits, both in their understanding of a particular problem as how one problem relates to another.

As the grid has grown, more residents have extended to features in the system that are broken by completely unrelated changes. "Oops, we patched a damage exploit - there goes your kickass game that used the feature."


The inability to version the system is the root cause of this problem. If residents were able to specify the patch version their content runs under, they would have recourse if something breaks. Developers, whose first priority is stability, need this in order to create content that works.







Problem #3: Load-balancing is impossible when only a few people are doing it.

Second Life is not a distributed system; all of your data is housed in servers that are clustered together and under LL's explicit control. If a server should die or the grid crash, there is nothing we can do.

The inability for residents to maintain their own "Second Life" is the root here. Ultimately, if the grid is to grow, residents will need to be able to maintain the hardware behind it. This is the only reason the Internet works; if it was maintained by three people who balanced the whole thing, it would have died out long ago.







Problem #4: Company Benefit versus System Benefit


Linden Lab is a company. It must pay its employees, its tax collectors, its creditors, and its investors. In doing so, anything that makes this process easier or more lucrative is in the company's best interest.

This places a conflict of interest in the higher levels of corporate culture. It has led to several "contests" and "bids" for real work under the assumption that the residents would be paid peanuts for what it is they do.

Even in the case of the Bedazzle bid, the majority of monies earned were meager in comparison to the actual time investment that went into Wells Fargo. You could make more working an entry-level job than what they were paid.


SLTV is another interesting farce. Bids were taken from residents that offered very real time, under the assumption they would be paid Linden Dollars and be given server rights for their work. They were not paid, and so SLTV was canned.

Do you think these residents are going to be fooled twice? I doubt it.

If you are going to benefit your company, do so through the ultimate realization of the system. Capitalize only on the tenants of it (bandwidth, storage, and niches). Screwing over your investors and main users will only ensure they're not there for you in the future.







Problem #5: Enforcing the Unenforcable

Second Life is a small company with only so much to pay their lawyers. They refuse to go after even the most flagrant offenses to their system, simply because it would be too costly.

And yet, several recent cybercrimes have been committed. A more open system, awash in new capital, would be able to pursue these malcontents and bring them to terms with the real damage they caused.

Justice may be a lost cause, but when someone can crash the entire grid or break into major investments, I know I would want blood.







Problem #6: Your data isn't really yours.

All of your data is owned by Linden Lab under the express terms that they may use it for all PR purposes in an unrestricted sense. Furthermore, they may lose it at will, with no recourse of action on your part. These terms are stated in the Terms of Service that all agree to upon accessing the software.

This is a dangerous fallacy. There must be sufficient recourse to recover ALL of your data at any time, up to and including entire sims. If there is not and your data is lost, you're done.







Problem #7: You aren't allowed to "do anything"

Second Life is a world that says you can do anything. This is not true. While it allows users sufficiently more control than MMOGs, I can think of at least a dozen things that cannot be done without having to majorly alter the system.

And yet, we aren't allowed to do this. Some have, yet they cannot come forward with their inventions in fear LL might delete all of their work.

We should be allowed to do whatever we wish with the client so long as it is in good faith. If we cannot, then "you can do anything" will remain as a false cause.






Conclusion:

As Second Life exists now, it has neither the ability nor the will to become the "metaverse" it first set out to be. In order to make this happen, there must be a paradigm shift for all involved.

That being, we must all be empowered to handle our own data in the ways we see fit. Second Life must allow sufficient access to its inner workings and to competition for the niches forming around it. If they cannot, Linden Lab will spread itself too thin as a company and the entire thing will go to hell in a handbasket.


And, as someone that puts considerable effort into what he does here, I'm worried.


Discuss.

I think you're missing the whole point of HUD.



(bolded in case my post is too long)

But seriously, HUD basically gives you the ability to build your own UI.

It's given us developers a whole universe of options. If you want to redesign SL - go to it.

I'm sure if you start to get very compelling they would be willing to add a preference which says "get rid of LindenLab UI"

In re to the edit:

I'm not trying to redesign LL here. I'm simply pointing out some fundamental problems that will not fix themselves as LL currently plays things. If this means their competition understands them before they do, so much the better overall.

But, as what I consider a time investor to what it is they do, I'd like to see them "get it" before someone else does and sweeps the pot.

I did.

You really only made one valuable complaint as far as I could see that supported your main thesis:



Yes, you have a valid complaint here. I personally think this is more our fault, as content creators, that we're not organising and calling a general strike.

We need backup. How can we continue doing anything for LL without full backup?

But, let's face it, they don't want people export assets and then importing them into competitor B.

We have very different philosophies about most of this stuff, so I'll refrain from comment on most of it. And to be honest, once a post is longer than my hand, I stop reading. If it was posted by someone I respect I start scanning (you got a scan).

However, being that I'm the security field and have seen this whole thing played out in the Internet Security world, I believe LL will eventually prosecute someone. They will wait til the right asshat comes along and can be made a really good example of, and it will make people think twice about bringing down the grid.

The question becomes, though: What level of offense would be considered that bad?

Second Life has seen DoS attacks, grid crashes, script break-ins, and many, many, MANY illegal asset copies in its short life. Aside from someone breaking in and compromising all of our financial data, I can't see a larger offense.


Though, in the long term, you're probably correct.

Beyond just simple backups, we need the ability to ensure those backups are in a format that works if the entire grid were to fail. Otherwise, they may as well be useless when the next patch rolls around.


As for import/export to Competitor B, I completely disagree. Historically, they've been quite friendly to requests for asset imports and exports. This includes my own request for the Blender script I have yet to finish, and the OSMP, which got further than I have yet to.

I can't really disagree with any of the points raised here. However, I am not optimistic about any of them being realised even in the longer term. The current trend of everything being regulated to the point of functional suffocation will probably affect SL in just the same way that it has affected the internet in general, and of course if the longer term plan is to divest SL then a virtual world where "anyone can do anything" (which as you said, isn't true anyway) is going to be a tricky thing to market to a potential buyer, particularly one who wishes to grow the business with minimum maintenance outlay.

I wish it didn't take hours to restart the grid. I also wish we had text-on-prim. Real text-on-prim.

I'm going to agree with most of what Jefferey said.

No, really. He makes extremely valid points, that sadly probably won't be answered in any meaningful way. Oh sure, maybe Philip will roll in and make some vague, nebulous promises about something or somesuch, but the fact remains that there are some serious essential problems with the entire SL system that need to be fixed, or the house of cards will finally topple and die.

Sadly I have no way to fix any of it, I just build buildings.

LF

Round up the developers, form a group, and meet with LL to discuss these issues. Why should the Land Ba... I mean "stakeholders" be the only ones who get some representation.
For extra coolness points call your group Feted Inner Core

Hi Jeffrey--
I enjoyed your post (because I am in general agreement, naturally). With regards to the above quote, are you saying that all SLTV bidders expected payment (but were not paid) or just the winning bidder? I have been having a bit of difficulty finding concrete information pertaining to SLTV's demise, outside of the official LL statement. It seems that LL's version of what ended SLTV doesn't match with your view, so I'm curious to learn more. Feel free to PM me if posting to this thread on the subject is off topic, thanks.

The original poster cites an effective preventative for what he (and I) believe will be the eventual causes of failure of LL/SL.

Unfortunately, none of these concepts are new to the forums or to LL. And as good, right, correct, or well reasoned they may be, LL has shown either disinterest or outright disagreement with these approaches.

To those who have advanced these faults and cures before (most of whom are well experienced in commercial software development) the costs of failing to address them seem blindingly self-evident. Unfortunately, they don't appear so to LL. From this I have concluded that they either don't understand what their customers have been telling them, or the implications of it, or that LL has some agenda with which these ideas are not in accord.

It is very natural to want to help LL succeed for any number of reasons: they invented something cool that merits continuation, the amount of time, effort, and self that players have invested in the game, etc. However, as a colleague of mine is fond of saying: "you can lead a horse to water but you can't make him think". Regardless, I wish you luck in your endeavors.

To scatter the SLTV info further around the forum and to make it even more impossible for people to get the big picture. My group was runner up and we were paid promptly.

That's first hand knowledge. I can't speak for all the second hand, "They never paid the winners" comments out there.

No offense but this just sounds like Jarod Godel's old saw to me. Distribute the servers and break open the client is the majority of what you suggest here.

Distributing the hardware and allowing archiving or even working on content offline is a good idea and eventually must happen.

Releasing an open source client at this particular stage (despite what Jarod says), would destroy the game IMO. And I say that being a supporter of open source software.

Also your vision seems to be of some massive corporatised game where there are a million "flavours" of SL running on a million machines. Of course most people will end up playing the Microsoft version of SL in that scenario.

How exciting, and how close to the original goal of SL would that be?

I agree that the game is not as advertised anymore, but the solution is to get *away* from the monetising of the game, and *away* from the idea that the entire wolrd and every griefer in it needs to play here. If we went back to the smaller game SL used to be, and re-established ourselves as an actual community, instead of just a bunch of people on a grid, then they might actually be able to deliver on those promises again.

I also note that the end point (teleological goal?), of the program you are laying out here seems to be this exact state of the original small community, but replicated on millions of home machines. Perhaps all these things are historically inevitable, but I would prefer for LL to take three steps *backwards* take a deep breath and *think* about the kind of world they want, instead of just chasing that corporate rainbow.

THINGS FALLING THROUGH THE CRACKS..............

One central problem I've seen time and time again is some sort of disconnect that quickly becomes apparent and magnified when you view a world that is quickly inchange and progressing, with the population growing and so many things happening—but like little chicklets crossing the street and getting flattened by a steamroller, a lot the finer strokes of the pen are missed. It's like all capital E-letters showing up as F! Or in cases of effort applied clumsily, transposition: EUCK YOU!

Specifically, by this, I mean a lot of the LITTLE features that make a BIG difference, that are voiced by Resis—here on the SL Forums too!—but not implemented (yet). Practical, pragmatic things that when bundled up, help turn a rusty dagger into Masamune. The little touches, the icing on the cake, the trimming on the tree, WHY IN THE TIMELORD'S NAME IS THE JPG COMPRESSION ON OUTBOUND POSTCARDS SO 1995 (aka BAD)? Who uses this feature? A lot.



So, who can catch them? Who can speak out?

The answers to those speak volumes, because it's about communication.

I'm thrilled gratefully to see what happened with:

1. HUDs
   
2. foot shadows (a nemesis for 2+ years)
   
3. Open for boxes
   
4. Clear Cache button (gawd!)
   
5. etc.
   

and I'd like to see more MoRe MORE of that! And quicker. MORE MORE MORE!









Think of guerillas in the mist.

Some enemy troops are walking down a path, they see a big sign that says "GUERILLAS WILL BE ATTACKING YOU SHORTLY", and they laff it off. Well guess what? A few seconds later, the guerillas attack!

The enemy is surprised and stunned, they struggle to mobilize like drunken apes, they fire into the darkness while several of their forces are quickly gutted and perhaps even bisected, or beheaded. A guerilla quickly confuses them with a stray flare—misdirection! While the enemy scatters to investigate, scattering is not their forte, unlike the guerillas.

A guerilla descends with ninja stars and quickly takes out several of the enemy, and leaps back into the darkness. A bomb explodes elsewhere, eroding the morale of the enemy into subzero levels. By the time this is all over, there's ONE enemy person left in the middle of the savage strike, and the guerillas spare him to weep inconsolably so he can go insane and tell others his tale.



Aaand...

Look at this.



Sumi-e by Carole "CJ" Jordan. Poise, delicacy, elegance, yet an artwork of sturdy, firm bamboo. And there's that big bubble of bubblegum in the background.

What does this all mean?



Think of the game Dance Dance Revolution. To be a master of it, you have to have an incredible overall performance, yet an AAA rating requires you to get each and every little step right too. You have the basket, you have the marbles. There are no holes in the basket, or at least, as few holes as humanly possible.

Yupyup, we have a Feature Voting Tool. And that is all very fine and dandy for BIG things that do take a lot of thinking, and possibly a very formal and extensive process too. That's fine. But again, this is about the LITTLE things that make a BIG difference.

It shouldn't take a Congressional hearing to reword "Release Land" to "LOSE YOUR LAND!" or somesuch thing.

And, as a final kicker, consider the DEBUG menu options. Why is it called "Force Sunset" instead of "Force Daylight"?

Thanks for the first-hand account, Satchmo.

About how long Second Life will be around: There is a 2-D virtural World that just turned 10 years old. I am more than sure when they turn 20 we will be celebrating the 12th annerversity of Second Life.

Anything for you I read Clickable Culture religiously, even if I don't always agree with it.

a lot of interesting posts in here. My comments (not really disagreements) are thus:

1. competition with Linden Lab: I still see GOM as a fairly special case, although it won't be the last example of competition no doubt. Because purchasing L$ is so *fundamental* to the viability of Second Life, LL needed to offer not only a seamless option (which could have been solved by allowing GOM hooks into SL like you said), but also a trusted option. I thought the GOM team ran a nice, trustworthy biz, but I remember being very uncertain about making that leap when I first did it. LL had to get around this trust dilemma.

2. patches: I don't think we want LL, which already has major R&D complications to work through and improve, to have the added burden of version control and support. But then you guys are more technical than me, so that's just a guess.

We have to be willing to accept the fact that our stuff might break in the name of progress. Although I agree with you that too much instability will drive any and all serious developers from SL to the detriment of all.

3. opening up the grid: philip and cory have already stated that they plan on allowing people to host their own sims... what has happened that makes you doubt this now? (that isn't a loaded question, just a question)

4. Company benefit: at face value I am appalled at the price LL pays for resident goods. However, my gut says that the market should continue setting the price for things in SL. People bid based on many reasons, and sometimes bid aggressively because their cost of production is really low (already have something in inventory) or because they view it as a marketing cost rather than a revenue stream. On the other hand, I am appalled that Linden Lab creates new $L to pay for these services rather than buying from Lindex. They are diluting their creators at the same time as rewarding them (minor dilution I know, but the principle is there)

5. data isn't really yours: yeah, this is a problem they really need to fix at some point in time

Hi. My knowledge of the winning bid is essentially second hand, but I believe the person to ask is Fox Diller. He isn't my source, but I've been told he was the leader of that outfit. This pertains specifically to the winning team.

In a nutshell, the winning team had been presented with a contract that had been worded to place all liability on them. Further compounding the problem was the need to be paid in a very large sum of Linden Dollars (L$), which they opted instead to buy through the LindeX than siphon into the economy. And, of course, there was the expectation of multiple hours of "compelling content" a day for what amounted to drug money, and issues regarding the bandwidth the production would ultimately use.

One or all of these seems to have driven the project into the ground. I can confirm this experience by how the Game Dev Contest panned out. While we were ultimately paid in full, the process took well over a month and involved me talking with Reuben over the phone and sending email after email to see their end honored.

It's enough of an issue that I resolved never to participate in the contest again, at any rate.



Simply because I don't see it happening. The push for resident-owned sims, versioning, et al, relies on the need for baseline standards that Second Life itself rests on. I have not seen or heard of any headway in this direction, with the exception of LL's want to store our inventories offline as an encrypted file.

What I'm bothered by are the assertions LL has no concrete gameplan (source: roundtable with Andrew Linden) and some of the changes that focus specifically on limiting resident rights to protect against those few asshats that do crash the grid.

Once you have a standard layer of protocols, versioning the sims would not be entirely too difficult. And, assuming the majority of sims are running on the latest version, global attacks would not be as much as an issue, either.

And yet. I see a tug and pull going on here, between playing Second Life as if it is a game, and playing it as if it is the predecessor of an open standard. This dissonance needs to be resolved if they're to make any headway.



Anyway. I dislike bringing this sort of thing up and presenting LL with a PR nightmare on what I expect is already a shitty week for them. But all the same, I like seeing the information that's there presented fairly to residents. These are all patterns that have been going for some time now; acknowledging the problem is the first step to a solution.

hmmm yes indeed.

the fact that you were so frustrated by the dev contest so as never to do it again ... how much of an understatement is it to say that hearing such a thing is troubling?

In theory, it's an excellent contest. The fact they let us compete for land (which I never would have been able to afford) is enough of a good thing to keep it going in the future.

Further, Reuben and the remaining Lindens were kind enough to set up an entire second sim using monies we would have been paid had we been a full team of four (instead, it was myself and Alan Beckett).



The problem is the process it took to get there. The highlights:

- The contest judging was delayed roughly two weeks due to a scheduling snafu, vis-a-vis a patch. I had to point this out before they realized it.

- One of my major scripts received an "Asset Not Found" toward the end, but I had kept a backup of it.


- The website still does not acknowledge team members. It should read:

-- First Place: Jeffrey Gomez, Alan Becket
-- Second Place: Gary Bukowski, Max Case, et al
-- Third Place: Tiger Crossing, Cuddles Crossing, Arito Cotton, Lucah (forgot the last name)

-- Honorable Mentions:
--- RCR: Satchmo Prototype, DNA Prototype, Digi Vox
--- Podlets: Lumpy Tapioca, Palemoon Twilight, Robbie Dingo, et al
--- Krytterz: Kermitt Quirk, et al


Those are just off the top of my head. Some of them may have changed as the contest went on, BUT the fact the team members are not listed anywhere bothers me extremely. I've pointed this out three to five times now, been reassured it would get there, and it has still not been fixed three months later.


- I had to pry for a month and a half to have the rewards honored, including moves to the final land, monies paid, and account fees defered. I kept all of the emails from this just in case the terms of our year are not made clear, say, eight months from now.

- And finally, the judging methods were not made clear to anyone on just how the final "spread" would be calculated.


Ultimately, it was a lot more hassle than it should have been, and is not something I see improving. Of the devs that competed this year, most have expressed their concern at the process, many to the point of not wishing to try again.

This high attrition rate is a major problem. New blood may very well come to fill the void next year, but meanwhile, your top competitors are gone. Compounding this, how do you ensure a "fair" judging scheme when the Lindens are the ones passing the final judgment? While a touch better than the "Dwell" scheme last year, these are all things that should be addressed in full before even thinking of it next year.


So, it frustrates me that this is all set up as idle PR that, more and more, does not have concrete thought backing it. If this problem is not addressed, I do not have high hopes for Second Life in the future.

Aww that sucks to hear. Primmies was great.

I loved the Game Dev contest and we didn't even win. It was really fun to work hard with a team like that to build a game that hundreds of people enjoyed playing. But then again... we were never in it for the money. I most definately will design a game up for next year.



You make it sound as if it's a game or an open standard. Not all "platforms" are open standards. I can think of the biggest, most ubiquitous platform in the world, and it's very very closed.

I'm all for open source/open standards, and for moving sims to private hosting, but there is a time and place for both and I think LL knows better than the rest of us when that time is. I think one of the worst things for SL right now would be for every tom, dick and harry to be trying to run a SIM on thier crappy DSL connection.

I think there is a really strong arguement for wanting to export content. I have lots of customers who want to prototype stuff in SL and then are disapointed to find out they can't use it in projects outside of SL later down the road. It's a big investment into models that can't be used elsewhere. Another good reason is because I wake up every day in a cold sweat wondering if my teams inventories have been wiped.

But in the over-arching picture, I think most successful companies have become successful because of "Intelligent Design". Every business decision should be carefully weighed and thought through. Even google which started out as a really good alogrithm has manuevered themselves into a blue chip company making decisions about IM, desktop search, news, email, etc...

Yea I can't argue with you there. Every point you made about the contest is valid. I'd throw in the fact, that my team busted our asses to get done on time, while other teams used the Linden SNAFU to build for an extra week or two. That was frustrating.

But it was still an awesome experience, and I'd do it again even knowing it was going to be poorly run. Next time I just won't be stressed and so competitive about the whole thing. I'll enjoy it for what it's worth: working with a team and creating a fun game that people actually play.