Second Life Complaints Bring May 3 Meeting
Linden Lab, maker of the virtual world of Second Life, announced on Monday that it would hold an impromptu town hall meeting with chief technology officer Cory Ondrejka to address the concerns of the almost 3,500 SL residents who have thus far signed on to Project Open Letter, a list of five complaints about the Second Life platform that have long gone unanswered. The open letter is an initiative of Second Life snapshot baron Cristiano Midnight, who’s also one of our co-hosts on SecondCast, and has been a member of Second Life for several years now. The project was started “after I read yet another open letter in a third party forum begging Linden Lab to fix myriad problems that have been going on daily for more than a year, in some cases extending years,” Cristiano writes on the project site. He goes on to point out that LL has closed down all the centralized venues of feedback, such as the company-sponsored forums, that had formerly been available to residents. A bug-reporting system is in place, but LL has not been as responsive to that as many residents would like. In the wake of the letter, LL appears to be listening harder, but it may not yet be hard enough. The platform remains vulnerable; read on for more details.
The town hall meeting, to be held this Thursday, 3 May, at noon SL Time (3pm EST), will give LL a chance to trot out the “progress” it has been making in addressing the five issues raised by the letter: the disappearance of inventory items from users’ accounts (which the letter describes, accurately, as “devastating”), the non-functioning Find and Friends utilities, general stability and performance of the platform, problems with the build tools that require many workarounds, and problems with the delivery of inventory and the functioning of in-world ecommerce.
The letter’s signatories are asking that LL fix these problems before devoting any more resources to new features. And they have a point. Many of the most basic functions of a virtual world or any 3D online environment simply don’t work in Second Life at anywhere near the level needed for broad adoption by enterprise or by a large number of users. (Only about a million people use SL in a given month.) The problem should look especially acute given recent announcements by Sun Microsystems and IBM to the effect that each is building its own enterprise-level system to support 3D collaboration, socializing and ecommerce.
Despite those and other systems in development, though, Second Life remains the only game in town, for the moment — which means Linden Lab has the luxury of doing their development in any order they wish. The residents’ concerns are spot on and should be heard, but it’s hard to envision Linden Lab switching strategies at this point. Not that their position is secure: Second Life remains vulnerable to competition from a new platform, especially one that (a) works better than SL does, which wouldn’t be all that difficult to do, and (b) is tied to either enterprise or media in a way that can direct massive numbers of users to adopt it. It’s that last point that’s most dangerous. A technologically viable competitor will come along sooner or later, but the difference will be made in how that platform ropes in users. Platforms like Virtual Laguna Beach rely on heavyweight media properties; IBM will rely on a deep channel of existing clients. Second Life has neither to rely on — which means it needs to make sure its technology is as top-notch as it can be.
Feeds
Well stated.
Great analysis…I am looking forward to jumping in world for the meeting. Assume you will report on this in case we miss the townhall.
UV
If the limousine liberals are serious about their protest, they should boycott this Town Hall and stage a protest somewhere else in Second Life. If they go to the Town Hall, they are legitmiizing a very flawed an unproductive process of having the game-gods come down from Olympus now and then (mostly then) to speak in one-line sound bytes to the mortals. The events are laggy, confusing, and often griefed, either out of frustration, or as part of a long-term conspiracy against SL by some extremists who believe violence should be used on any effort by the society to become more coherent (and the Lindens don’t acknowledge this conspiracy, which they in part fuel with their tacit support of these griefers in open-source groups).
You can’t do much to change a defacto authoritarian power like Linden Lab in this situation, but what you can *always* do is refuse to legitimize them. If the 3,500 people who signed this letter accept as a sop yet another Town Hall, then they’ll be back signing the same letters next week.
No, the tier-paying residents of SL who make up 80 percent of the Lindens’ revenue and who help make it now profitable need to resist, and not attent a Town Hall that will not offer solutions. They need to ask for more serious political power than just sit passive like morons on laggy town hall benches craning to hear what Yoda will say next.
Instead, they need to ask for a collectively-represented seat on Linden Lab’s board, for the restoration of the community forums with even-handed TOS enforcement of them, and for and end to punishment of dissenters. Those three things are the minimum.
It’s hard to understand why, if IBM and Sun are making little avatar-based simulated office environments to work with, and THAT is the competition (they sound about as much fun as a conference call, an annual budget process, and a company picnic all rolled into one!), then they need to do some hard explanation as to why voice, scultped prims, and “hide me online” were so all-fired important. Indeed, if we scrape away the hype about “the competition,” we see it’s just the very narrow and self-referential agenda of an elitist corps of tekkies and their close personal friends on the grid. that’s not justifiable, but the “Open Letter” and its signers are merely telling us that they *used* to find that process of the internal core and its feting working for them, and now they don’t. They’ve been replaced. By other new, close personal friends.
I am interested in attending this meeting as, for a change, I may be in-world at that hour, but I don’t really expect to hear any amazing news from it; the response to the “open letter” will be basically to pat people on the head and say “we’re working on it” (as concerns the general “this isn’t working” points, and that is almost everything, apart from the things about unverified accounts and restrictions, which will just be ignored - the proposals aren’t based on anything real in any case, just assumptions and rumour, and no developer will take action like that based on the vague suggestions of an outside body. Users specify user requirements, not technical requirements.)
Clearly LL know that the grid is malfunctioning. Clearly they are attempting to do things about it. So, we will have a PR response, and developers will continue to do exactly what they were doing before, and when (if) they solve problems it will be put down to the amazing influence of the open letter. Well, I’m sure that will make some people happy, but it won’t be real.
well, Linden Lab should be well aware of their situation and from what I hear at office hour meetings they are also very active in making the grid more stable. Of course it’s not a thing you do overnight as it involves a lot of ideas and testing and you can never be sure how these ideas do perform under heavy load before testing it out.
And I must also agree that a townhall is not a very productive thing of communicating. I wonder though why not more people (esp. those who signed the letter) come to the office hours. You can learn quite a lot there about what’s going on behind the scenes.
Additionally it won’t help anything if every ressource is now put towards fixing bugs as those might know who read the Mythical Man Month. Also Open GL experts who might work on the sculpted prims might not be of good help when it’s about fixing database load issues.
So in conclusion I think that LL is more than active in trying to fix these things and I also think that new features are needed and possible nevertheless.
Mark Wallace said: “Second Life remains vulnerable to competition from a new platform, especially one that (a) works better than SL does, which wouldn’t be all that difficult to do,”
Hahahahaha! Oh, you kidder, you. Or, for the sake of me not considering you as technologically backwards as Prokofy, I hope you were kidding.
Have you tried Alphaworld or There.com lately, to refresh your memory of what other attempts at the Metaverse were like? Did you see how unspeakably bad Sun’s meeting world looks? Have you tried to use Croquet? It’s insanely hard to make something that does even the bare minimum of Onder Skall’s Big Three a tenth as well as SL does. MMORPGs use preloaded content, shards where you can’t even communicate with your friends on other servers, further split the world into “zones” which are essentially MUD rooms, and have little or no persistent content, all of which lets them scale up fairly easily, but makes them uninteresting to Second Lifers. Eve Online is the only big single-universe MMORPG I know of, and it uses extremely heavy hardware and very simple graphics and an isolated-solar-system model. You can’t make SL with Eve’s solution.
There are many bugs and problems in SL, but if you honestly believe that you can go rewrite SL in a weekend, HAVE AT IT, LADDIE. You will not just fail, and fail miserably, but will soon discover that it’s like scooping up a mountain with a teaspoon. The teaspoon will break, and your limbs will shatter, and the mountain will ignore you.
What Sun and IBM are doing is taking the social lessons of SL and making small VR meeting areas, but they’re not meant to have real graphics, physics simulations, floating gothic castles, furries, escorts, casinos, or scale out past a few servers. They just want to hold meetings and project powerpoint slides on some screens. SL’s overkill for them, but what they’re making is totally useless to us. SL’s the equivalent of a 747 compared to their Wright brothers flyers. Now imagine upgrading a Wright brothers flyer into a 747, while it’s in flight. Now also imagine that the passengers are on the wings, jumping up and down, demanding the pilot personally get up from the controls and bring them coffee RIGHT NOW. That’s you petition idiots.
It is conceivable that someone will be able to make another Metaverse eventually; I can think of maybe a half-dozen companies with the massively-scaled server experience, the programming talent, and perhaps the vision to do it. It ain’t gonna happen just because a bunch of whiny idiots wrote an open letter.
Prokofy’s suggestions are just as futile. If you want a forum, start your own, or use sluniverse.com; what, are your fingers painted on? For someone who hates and fears social networking so much, expecting a “community” board member to be anyone but a major developer they’ve had business relations with, is unbelievably naive. You think the Sheep or someone like them, once they have real political power, are going to be generous to you, Prokofy? You’re trying to vote yourself into the gulag.
Kami, I’m not at all suggesting that I could make a virtual world. Nor that it will happen “because a bunch of whiny idiots wrote an open letter.” But I have to believe that there are competitors out there who are just as competent as the LL team was when it set out, and who are already working on alternate solutions. They may or may not be better than SL. But if LL doesn’t keep its product in top shape, it will be surpassed. LL didn’t have “the massively-scaled server experience, the programming talent, and perhaps the vision” when it started out. It had the vision, and a bit of the talent. The rest was learned and hired on the way.
“Now imagine upgrading a Wright brothers flyer into a 747, while it’s in flight. Now also imagine that the passengers are on the wings, jumping up and down, demanding the pilot personally get up from the controls and bring them coffee RIGHT NOW.”
No. Petition is about not trying to make additional pair of wings, new row of seats, to change the color on the sides of plane and to discuss which flavour of tea will be served while we losing altitude!
Point of petition is to fix the system before making it more complicated. Easy as that. And you and me and Lindens know that.
It is about strategy. More and more companies are talking their dreams of 3D internet, virtual worlds and whatnot. So the LL are getting their positions on the future market by getting more and more users (no I won’t use the word “residents” in this context) - we got some four million registered accounts in last six months. Other thing Lindens do is pumping the features list of Second Life.
We should make a petition that LL hire somebody to make them good company development strategy. No point in pumping numbers of users if system will fail under the weight. Yes, it brings money, but question is if big companies interested in marketing potential of the world will be still interested once that things go too far. And features list is just a list. What is the point of features that are buggy and, even worse, which bugs other features and complete system?
It is true that lot of companies are announcing their worlds. But as we learned so far, that talk is motivated just as all this Linden’s fury is: by need to get a place on the market. So even if they don’t have a product, they will talk about it. OK; some are developing very fast (like PS3 Home and probablly IBM and Sun, but all of them lack what made SL great as it is: user generated content and that nice mixture of game, social platform, creativity tool and luck). So, why are Lindens in a hurry when there is no real runners-up? And even more interesting thing: why they are running from fuss they created in the first place?
Perhaps you don’t realize what Philip Rosedale did before Linden Lab: Real Networks. They were the experts at making massively-scaled server farms with constant streams of data, and making a multimedia experience out of it RIGHT NOW, lest the users start bitching that their streams are all jumpy. In other words, he was one of the only qualified people on the planet to make a Metaverse that worked even as well as SL does.
Aside from the mass hysteria among some users, there’s not much wrong with SL, very very few critical bugs, and the bugs that do crop up get fixed faster than I’ve seen on most enterprise systems with a lot more money riding on them. So LL’s current approach is pretty reasonable: fix bugs where possible, add cool new stuff to raise the bar for competitors, and keep the graphics as current as possible (since that’s a constant “maintanance” requirement to stay in the industry).
The only real danger is that now that one whiny petition has made LL waste several days responding to it, more idiots will make more whiny petitions (Dandelion’s reaction is exactly what I’d expect), dragging SL down to a fiery crash. But hey, they got the captain to come out and bring them coffee.
I’m quite familiar with his history. He and most everyone else at the Lab are an extremely talented bunch.
I think the original post by our gracious 3PointD host is very much on the mark. First off, I look upon what Linden Labs has done with Second Life as an amazing feat, given the multitude of technical hurdles, and trying to ‘please’ multiple constituencies. That being said, there comes a time when the ‘wow factor’ wears off, and for most users that have stayed on Second Life for a couple of months or more, I think that is now the case. Linden Labs IS running a business, as they charge a one-time and thereafter monthly fee to those wishing to secure server space and operational capability to supply a digitized capability. As such, there has to be some expectation of customer satisfaction, as well as quality of service (QOS). The reality is, technically, Second Life is not improving, but getting worse.
For example, since yesterday I have been able to log into the Dell Sim, my AV’s last ‘resting place,’ but once there when I attempt to teleport, the following message is received:
“You have been logged out of Second Life:
You have been disconnected from the region you were in.
Click Continue to look at existing IM and chat.
You will not be able to perform any other operations.
Click Quit to exit Second Life immediately.”
This has now carried over into another day. Perhaps it is an issue solely impacting me, but I doubt it.
Increasingly, it is my belief that we will look back on Second Life as an interesting experiment and ongoing beta test bed for a future Metaverse that will not be owned by any one company, government, or authority, as it simply is too gargantuan of an undertaking for such.
The question becomes does the lack of scaling and other hiccups associated with Second Life begin to turn off users, both early adopters, mainstream, and businesses? I would suggest yes that could be the case, and it might also have the effect of setting back Metaverse development for at least a while–an unfortunate outcome to be sure.
As for the LL Town Hall Meetings, and even the SL Blog, I would agree that they are something of a pressure release valve, in part meant to serve as a venting mechanism. Problem is this only works for so long.
“So LL’s current approach is pretty reasonable: fix bugs where possible, add cool new stuff to raise the bar for competitors,”
That is not reasonable at all. Any wise colonel knows that if there is no threat of immidiate attack he should reorganize, feed and rest his troops.
Bar for the competitors is high enough. It is not only features that counts in the race. As many developers and most of gamers know, it is not just cool graphics and features that makes a game or a virtual world “works” for its users. Just remember Tetris. Even if we tomorrow see new virtual world, with great feats and no bugs at all, do you really think that million and a half residents that regulary log-in will go there? No. The amount of content, familiarity and habit will keep us in SL for a good time. And it is even not all about what has been made inworld. There is a good proportion of fun, busuiness, game atmosphere, creativity… that is hard to copy and which is independant on features. It is not heavy to make MySpace but too many similar projects failed.
Point of petition is not to make Philip to bring us coffee but to state that we can live without in-world voice and even sculpted prims but that it is essential to have stable world which will not crash six times a day. If we talk about competition, bigger threat would be the world that is stable than the world with voice. We had fun with text based games, but never with toys that breaks every hour.
The issue I have here is that there is this persistent idea that there is some sort of fixed pool of “developer time”, which could be devoted to, say, sculpted prims, or could be devoted to fixing sim crossings, and would have equal impact either way.
In practice it does not work like that. If you have somebody whose talents are in graphics and modelling and you say to them “right, work on debugging this XML transfer routine, don’t do anything else for six months”, the likely result is (a) that they don’t do terribly well on that since it’s not their area of expertise and (b) that they leave because they find it terribly dull.
As well as that, the issues involved in fixing things which appear basic and simple are not necessarily basic and simple at all, and even if one has the option of spending time doing them, it isn’t necessarily going to produce results that anyone notices.
From all that I have heard and seen, LL developers are perfectly aware that there are problems with the grid and spend a lot of time trying to solve them, but for heaven’s sake, a few projects where one can say “look, here’s something new!” and stimulate the mind aren’t too much to ask. Google seems to have done well giving people 20% “me time” to work on individual projects.
As a programmer myself I do not have a lot of sympathy for the viewpoint that sees my skills and the skills of my peers as being part of some undifferentiated mass to be applied to whatever end is convenient. That is the sort of old-time “human resources” attitude which produces stagnant companies which go nowhere. Hiring people with specific skills based on business priorities is a different matter, but you don’t tell a painter to start sculpting because “it’s all art isn’t it”.
It is not about developer time, not human resources to be thrown on anything that we, residents or lindens, wish to. Though I agree that “something new” is exciting and motivating for the both parties, there is one simple truth: it is hardly possible to fix something while it moves and changes, especially if that is something so huge, organic and complicated as Second Life is. One does not need to be programmer to know that most of the world’s problems come from server overload (yes, we are testing the edge of today’s technical possibilities) and most of new features are additional weight on servers. It just doesn’t make much sense to do that.
I’ve pretty much said my opinion of LL’s QA and testing processes at https://jira.secondlife.com/browse/MISC-37 , but I found the following comments from cory at the recent town hall response very illuminating:
ref: http://ordinalmalaprop.com/documents/townhall-2007-05-03.html
[12:21] Cory Linden shouts: written with unit tests, whcih help. But the best way for you all to help make SL better is to use the beta grid and report repro cases when you find them.
[12:23] Cory Linden shouts: Patches are tested on the beta grid, but often times bugs do not appear until full load testing.
[12:29] Cory Linden shouts: as I mentioned int he blog post, we are attacking these problems in two ways, to first identity weak spots in the current system and to build a better transaction system as well.
[12:31] Cory Linden shouts: We’ve hit this one a few times. It is the majority of our design work, we post all potential code to Beta grid, and we hope that more of you hit the beta grid to test it.
In Essence:
. Load testing/simulation is one of the key issues for stress testing, which is a no brainer for any scaling transactional system. Solution: Invest in automated test systems developemt, because you cannot rely on free user recruitment.
. Only now are they starting to look at automated load testing systems it appears.
. They have also only just started a test driven development methodology to improve the quality of the development process itself..a concept that has been central to the agile development methodologies for years
. They have been “hoping” [sic] enough users will use the beta test grid to be able to stress test and simulate aforesaid “bugs that only appear under load”.[sic]
It’s a common problem that has killed many companies and products. QA/Testing is often seen as the unglamorous end of a development pipeline, while in reality QA/Testing starts at concept phase and needs to be integral to the WHOLE development process for it to be effective.
Call it Test Driven development or whatever, but the concept of QUALITY is king/queen.
The worst thing that ever happened to the IT/Games industry was the concept [no doubt developed by an accountant under the guise of “we need to release faster to compete”] that you could just keep releasing product updates to your paying customers and get them to debug it for you…for free, and have to put up with the pain of waiting for fixes.
For certain classes of products of a certain small scale and low cost that concept can actually work…but when the product moves into a zone of cost/utility beyond that…look out when the competition arrives. Your so called loyal customer base will drop you like a stone for better Quality of Service, if the product offering and cost/benefits are in any way similar.
With all due respect Walker, these two statements do not square:
Second Life remains vulnerable to competition from a new platform, especially one that (a) works better than SL does, which wouldn’t be all that difficult to do
and
Kami, I’m not at all suggesting that I could make a virtual world.
If you can make the first statement about the relative difficulty of producing a better performing virtual world than SL, one assumes you are making this statement from a locus of technical expertise. Expertise which you appear to disavow in the second statement.
And as to the merits of the statement about it being an easy proposition, well, there are none. If it were easy, someone else would have done it by now. That only one company has, and is struggling with the difficulty of doing so belies the point in its entirety.