Answer to Update Priorities

There has been countless posts of complaints from residents about what is included in todays update. People have been griping about the extra work placed on flexi prim behaviour on region crossing and categorizing 'pretty lighting tricks' as pointless.

The majority of you dont understand how a developement team works. Programmers are broken up into groups that work on a task. There can be a dozen different groups all working on different projects. Some are working on improvements to the features available for Group Management. Some are working on issues releated to improveing network traffic, inter sim communications and asset handling. There are also other groups who are dedicated to going through the various crash logs, interpretting the information in them and figuring out which are driver related. It is bad management to try to put the entire software developement team on the same features. Different programmers have different areas of expertise and not everyone would even be qualified to do the job. People are placed where their specific skills get the most effecient work done.

In todays case a group working on the noted features happened to finish their job to a point where it would benifit the system to impliment those changes on the grid. There are other teams STILL working on the issues everyone feels are important and as THOSE issues are resolved more updates will occur.

The programming team that finished todays update is probably already working on yet other bug fixes and system improvements. They are probably NOT the issues that some of the community prioritises but thats because there are already people working on those aspects. There are also issues that only so many people can work on the same pieces of code before it becomes counter productive. Allowing the teams to have their assigned tasks is what allows new features to be added, while also allowing both minor and major bugs to get fixed at same time.

Please give the Lindens some slack and take into consideration all points of view before blowing up on them. They just got back from an extended weekend and are well rested ready to work. Dont ruin their good moods (reducing their productivity) with pointless rants and uninformed babblings. Thank you.

The majority of us don't understand? LOL.

Some of us out here know exactly how development teams work and appreciate the concept that you have outlined here. So I think the accusations of "pointless rants" and "uninformed babblings" are questionable. There are some SL users who are experienced professionals in the computer field. We are not all ignorant.

I am fully sympathetic of LL being on vacation and coming back all happy and joy-joy. The world still goes on. We were charged for every day of that vacation; there was no reduction in fees. So vacation or not, SL still continued, interface problems still continued. Vacation is over; back to business as usual. Stop brown-nosing.

The bottom line question: why was a development team assigned to such a project in the first place when there were so many other problems with the 1.9 release? People have been asking this basic question for months with no satisfactory answer. So again, I'll state the simple question of prioritization. When the boat has leaks you don't assign people to painting the staterooms. Everyone bails. All we're asking is for LL to patch the major holes before working to make the boat prettier.

I know exactly how development teams work and appreciate the concept that you have outlined here. So I think your accusations of "pointless rants" and "uninformed babblings" are baseless. As a note: there are some people out here who are professionals in the computer field. We are not all ignorant.

I am fully sympathetic of LL being on vacation and coming back all happy and joy-joy. The world still goes on. We were charged for every day of that vacation; there was no reduction in stipend. So vacation or not, system problems continued.

The bottom line question: why was a development team assigned to such a project in the first place when there were so many other problems with the 1.9 release? People have been asking this basic question for months with no satisfactory answer. So again, I'll state the simple question of prioritization. When the boat has leaks you don't assign people to painting the staterooms. Everyone bails. That's all customers are asking-- attention to patching the holes before making the boat prettier.

Because flexi prims h ave been in developement since SL 1.7.x is one reason. As features approach maturity and the main projects are finished the people who worked on those projects are sometimes the ones who do the bug fixing on that system as they are the most knowledgable to maintain the system. Thus as a continuance of the job they've been doing all along they keep working on that feature to improve it. And will keep doing so I hope.


"boats / leaking / blah ..."

Actually no. Some people bail, some people are readying survival items, some are communicating distress, some people are STRICTLY trying to keep a calm demeanor and organize others so that people dont flip out and go emo. In times of problems the BEST approach is for people to do what gets the most OVERALL done. For a boat what if you have 10 people but only 6 pails to bail with? The other 4 people need to be working on temporary repairs and getting help or everyones doomed.

And that REAL life the analogy works here. SOME of the dev team will be working on the bug based issues people are complaining about. But only so many can be editing the exact same piece of code. So to INCREASE productivity the rest of the coders, instead of sitting around looking over their friends shoulders at their screens, are sitting at their OWN desks working on other things at the same time.

Nice reply and i agree with ya Seronis

The point is that in case of a leaky boat, everyone is doing somthing to help correct the problem rather than trying to repaint the hull.

The problem is that every time they come out with a new feature... that just adds more bugs to debug. I haven't seen a new feature yet that didn't come with its own set of added problems. There comes a point that you stop putting devteams on new features and admit, 'You know folks we have a TON of bugs to correct... so that's all we're going to focus on for the next few months."

It's not like SL doesn't have enough bugs to keep their programmers occupied.

As far as the idea behind your point we both agree. Everyone needs to do something to help. But we differ in how we think that should be accomplished.

Everyone has different priorities. For me with a >2yo system and an outdated graphics card ( GeForce FX 5700 LE ) I check for bios updates a few tiems a year. Do the same with mobo drivers, networking drivers, gfx and all the others. I do a COMPLETE reformat and reinstall at least twice a year and up to 4 times just as 'regular maintenance'. Result is that I might have outdated hardware but it doesnt run with as many glitches as people who are not knowledgable enough to fix them theirselves and must rely on client software that detects your problems and tries to downgrade its abilities so you dont get bugs. Since im responsible with how i run my system I dont consider the current bugs nearly as high on the usefulness meter as adding features that people have been requesting for years.

Do you really think LL should drop work on fixing inter sim communication so that objects to object communication works? Do you think LL should stop developement on the improved Group Management tools we're all anxiously awaiting in favor of tracking obscure bugs? For what little work might still be getting applied to a more uniform physics API so that more advanced physics handling could be added do you think it should be put off?

I dont.

Every one of those things (and more) need to be in developement at all times as capability exists. There are people DEDICATED to the bug finding. It may take awhile but fixing the bugs its a LOT less productive than feature implimentation. So adding things that are missing does just as much improvement as fixing things that are glitched.

Had more to say but major migraines and my train of thought just derailed. I still fully support how LL prioritises their developement cycles (if not actually wanting the specific features that get added. We want physics!! hehe)

The problem with this "answer" is that it's a copout. Any development team THAT out of sync with each other needs better organization. It is the program manager's responsibility to act as a liasion between development teams and make sure they ARE in-sync. Then the program manager needs to communicate with the community manager so end-users aren't left in the dark. Yes, some devs are more skilled in some areas but that's no excuse to not keep the end-user community informed as to SL's development cycle.

And who's testing these skewed bug fixes? I sure as hell don't see them coming out in the preview grid first! And I even double sure as hell don't rely on Linden Lab's testing since it seems to miss even the most basic glaring bugs WAY too often! It's the case of the left hand not knowing what the right hand is doing (and is usually the cause of additional bugs).

SL development obviously needs some serious attention and organization because as it is now it's a confusing mess and only causes residents to get frustrated with LL.

Sorry about the migrane. There is just no silver lining to those things.

Regarding being dedicated to bug finding: if that is the case, if debugging is high on the list, then why are we seeing new, unnecessary things being developed while serious bugs still exist that were first reported months or even a year ago?

Yes, unnecessary. I mean, floppy prims and local lighting may have some application and may look kewl, but they are far from being near the top of the list of user-requested features, much less essential bug fixing. And considering that BOTH of these items serve to add lag factors to an already excessively laggy system, I think that just was not a wise decision. Surely there is a lot more than is in greater need of attention than floppy prims and lighting effects.

I wholeheartedly agree that the entire concept of flexi prims is pointless and a waste of time. All i'm saying is that it was a goal of LL to do it and they have spent quite some time in its developement. And its their right to prioritize their work schedule in this manner.

One useful thing to come out of the flexi prims:

Even if the movements are strictly client side rendering (for server resource effeciency) they stilll have to take into account a lot of various dynamics. This is akin to the start of better ragdoll physics. It is the same concepts when you can take an object, set certain bend points and have those points react to forces in believable and reliable ways. So if developing the flexi's got Linden Labs programming staff (at least one group) a bit more epxerience in understanding how to impliment this CATEGORY of data handling then I support it and cant wait to see if the knowledge can be used on a feature more critical and useful. By having the feature client side only and allowing users to turn it off at whim the staff can continue to learn how to optimize their ragdoll handling to make it faster without it being used in a feature where a bad implimentation would gaurentee harm a residents enjoyment.

Thats another aspect of a developement cycle. The flexi prims and the Mozilla integration are allowing the staff to get experience in important skills by applying them to features that are likely to cause the fewest problems if they mess up. Local lighting being upgarded to be hardware instead of software handled is a very MAJOR enhancement to me personally. It allows my GPU instead of my CPU to take more of the burden. All of these so called 'pointless' features have a bright side. It may not be exactly what some of the community wants accomplished but in the end it will make developeing those features easier because a more diverse framework of options will exist.

Also id love to get a more diverse set of opinions on this besides just yours Wayfinder though I do very appreciate you 'debating' your opinions instead of blindly preaching them. Not too many threads can have a non-heated discusion it seems and its refreshing.

That's nice. I reinstall XP at least five times in a year. My system runs beautifully and is also two years old. Computers are my line of work in many ways (web design, graphics altering, programming, repair, building new computers, network security, etc.). Throughout all of this, SL continues to crash. I am above the system minimum requirements and before this recent sparkle-sparkle update, I was fine. Sure, people have wanted this for a long time, I know I have, but not at the expense of hundreds of people not being able to play (this includes people that pay a lot (unspecified as I am too tired right now to do the research) of money. Sorry, but you boasting about how you take care of your system makes no difference in this matter as I do the same, if not better, and am experiencing issues. If I have made errors in this post, please let me know, as it's 1:18 AM, I've been evicted, have had a fork go through my finger, and am extremely tired.

That's all for tonight.

Good point and one I'm glad someone made.

I see Second Life getting more advanced and complex all the time. With each new update, there are more and more people who for some unknown reason cannot use the system anymore. As you, they have nice, powerful, standard computers that handle just about anything on the market-- except the latest release of Second Life. They must absolutely fume to see all their hard work go down the drain because they can no longer log in.

Seronis said:

I have to disagree. It is their duty to prioritize the needs of the customers who are paying to use this system. They have an obligation to their clients to provide the most stable, usable, friendly, consistent platform possible, because that's what people are paying thousands of dollars a year to have. We do not pay to have them develop things that are "pointless and a waste of time".

If the engine was stable, if the foundation was firm, if the major bugs were fixed and lag was totally under control, then I'd say, "Sure... now let's see what this puppy can really do".

But while major bugs remain, while people can no longer log on, while the data foundation quakes and people crash several times a day, while prims disappear before our eyes and the system is still running on Havok 1 for crying out loud... that is NOT the time to experiment with new, useless toys. That's the time to tighten the belt, pull out the wrench and get to work fixing things.

Lag is awful. Instead of developing additional things to cause yet more lag... they should be working to avoid as much lag as possible. Instead of making changes that cause people with normal, everyday computers to no longer be able to log on, they need to simplify and stabilize the system so that 99.999% of the users can log on easily.

LL is playing games on our dime. They seem to be run by programmers out to have fun rather than businessmen looking out for the good of their clients. At least, that's the impression that is given. And I've heard this same comment repeated so many times on these forums.

That's what people are complaining about: focused attention on non-essentials while ancient needs go un-met. Again, there is no shortage of bugs so that they have to occupy their programmers with laggy flexiprims... and there are surely other user wants and needs that should have been implemented first.

I have to say I agree with a little of both sides of the debate. I'm lucky that I can log on and not crash, but I'm equally frustrated by bugs in the building tools and cache that seem to have been with us since I started (pre 1.6).

Although we pay for a service, I think we need to remember though that SL is essentially a beta product, it's at the cutting edge of what's possible and despite growing numbers we are very few at this particular frontier. SL relies on pushing forward, staying ahead of the competition and in fact, the introduction of those shiny new features because they help drive the creation of new content as well as being the stepping stones for greater technical advancements. We all need to accept some of the risk in participating in this journey.

A balance needs to be struck, and it'd be interesting to see statistics - whilst I have the deepest sympathy for those having problems, if 80-90% of the population are doing ok, then it's a statistical risk LL may be prepared to accept to move SL forward.

I appreciate the points you make. But...

If SL is a beta product... then why are they charging fininished-product prices? (Face it, SL is fairly expensive to use on any level above a newb).

If it's a beta product-- then why are they not fixing the things the beta test comes up with as faulty? It's not wise to "add new features" to a beta product. The purpose is to test it for errors and stabilize the foundation. New features come after the product is up and running and released.

By the definition given above, wouldn't Micro$oft Windows be considered a "beta product"? After being on the market for over 5 years, I don't personally feel SL really qualifies to be considered a "beta" system. By this time, it should be an established, solid product with the main flaws fixed.

Statistically, if 80-90% of the people are doing ok... that means 10-20% of the customer base is not doing ok... which from a business standpoint is unacceptable. Add to it-- are those 80-90% just piddling around, chatting and dancing at parties, or are they the developers, the paying members, those working to make SL function and thus using all the features of SL every day?

Obviously, if a person doesn't build or design clothes or own land, they are not going to be as likely to have problems as those who do. So yes, a large majority may not be experiencing severe problems... because they never use many aspects of the system.

These are all things that factor into the matter. Bottom line: SL still suffers from flaws and lacks that have existed ever since I've been part of it (which is over a year and a half now). These things have been known, reported, ignored, reported again, ignored, until people are frustrated. And when they see things like "floppy prims" and "lighting features" being released while old needs and problems are still unresolved... they have a right to be verbal about it. They're not (as the OP tactfully put it) "ranting", nor are they engaged in "uninformed babbling". For the most part, they are very informed and are expressing legitimate concerns.

I hate quoting entire messages when they're right above me, but I feel it's needed in this case. I completely agree with every one of those points. Also, mainly this is feedback. If no one complains then LL will figure it's not that big of a problem, and will just disregard it. When they realize that it's upsetting their community, thus causing some of us to either leave SL or not boast about it as we have been, then they will work harder to fix these issues.

I called phone support the other day and spoke with two people. The first time I called I was not handled properly, I was told my card was not on the supported list and the support person was rather rude. I called back and got Michael, who actually did something. He pointed out to me that my card is on the supported list, as I had been told by at least 10 other Lindens. Also, he helped me find that it could be my power supply, as it's only 200 watts, which isn't all that much. As I have a lot in my computer, two hard drives, two cd drives, my graphics card, a tv tuner, a built-in graphics card, the motherboard, the processor, etc. - the power supply has to power all of that. As Michael pointed out to me, when doing anything even remotely graphically intensive, my power supply needs to kick out more juice to my graphics card, and at that point it may just not be able to supply it enough power and cause SL to crash. While this isn't what the thread is about, exactly, I thought that it is good to point out that at least some people are working hard to help fix it. I know Hermia, Patsy, and Dee all helped me a lot. Dee has been helping me for about a week now, and we check back with each other, and I am happy to report that she is very dedicated to this issue.

Also, folks, while bragging about some of the Lindens, do us all a favor and when you see a Linden not being helpful, tell Char or Mick, or even support -- and let them know your feelings as to the low-level of support this person gives. However, keep in mind that if you do that, you need to let Char, Mick, or support know when you see a Linden doing good. As for the case of Hermia, Patsy, and Dee -- I have sent in good feedback, as opposed to just sending in negative (that's the word I was trying to think of earlier in my post) feedback. This way, we can get some of the more helpful Lindens moved up, and weed out any non-helpful ones. Not to say that because of one incident a Linden should be fired, that's obviously not the case. But we need to let Linden Labs know who is helping and who is not, so that one day the Linden support that I see so many people complaining about will be much more helpful.

If I've stated anything incorrectly, it's rather early here (actually it's 1:08 PM, but I took a sick day and just woke up), so let me know.

I agree that not AS MUCH time (possibly) should be on the TOTALLY gratuitous features. Just a large portion of the population thinks since the latest release didnt fix THEIR issue they are under a FALSE IMPRESSION that Linden Labs was ONLY working on that one thing. The majority of the pointless babblers at least.

You two have valid and fairly professional points you're making. Im just trying to get the psychotic few to understand a bit. I'm also still deadfast in my opinion that only so many people can be set to bugs and smaller features still need in developement (such as group management).

That said and enough sensible debate from all sides readable above I'll concede that I personally wish we did NOT have as much developement in big projects. But some of them have other benifits (as referred above in how flexi prims will allow the coders to get other similar things improved).

Its sad, but a Happy Medium just doesnt exist. I'm very sorry for a lot of peoples technical issues. But thats the nature of computers. Maybe LL needs to do some prioritising changes, but its a WHOLE LOT smaller than the angriest of us are wanting. Anyone who has read this far has either accepted or ignored the developement side of the issue so i'm probably done in this thread.