Should we, the residents, motivate the Second Life developers? How?

by Eloise Pasteur Nov 29th 2007 at 9:01AM

Occasionally we get readers write in with story ideas. We are always pleased to accept their ideas and tips of course, but we don't normally operate a "guest writer" spot. However, Bob Wellman wrote in with an interesting opinion piece about Second Life that is printed more or less in full below the fold.

For those of you that don't know, the Tao of Linden says that developers work on projects that attract them and interest them. This is not always the same as the issues that frustrate the residents of Second Life.

What I am about to write is coloured by my RL background as an IT Project Manager. I know just how hard it is to motivate people working on a long term Software project and lets face it they don't come much longer than Second Life.

From what I have read it seems that LL has a nice laid back attitude towards employees. It allows people to pick what they wish to work on from the pile of enhancements and problems identified by Second Life residents and internal staff alike. I am sure it sounds like the sort of place we would all like to work. However, from a motivational point of view, this is a double edged sword:

On the one hand it means that people are working on things they feel good about or interested in and so they are motivated to do a good job off it. Wonderful!
On the other hand there are issues that no one fancies and that no one is inclined to tackle because they are boring or difficult to do. No Kudos in trying and failing. Oh dear!

These second set of issues need addressing too, as far as Second Life residents are concerned, but how are the LL employees to be motivated to address them? Let me, at this point, give an example.

If you look at the problem of alpha sorting textures you will see that it has been open for a long time, and with many votes for it showing how important it is to residents. It is without resolution or even someone taking on the assignment after nearly a year. Why? It is because it is a difficult issue to solve. No one wants to take on a nasty task when one can pick easier tasks to do and you boss doesn't insist you do the hard stuff. Why would they?

[In fairness to the Linden Lab developers this is not a problem unique to Second Life - it is just more noticeable because we have people that don't understand it mixing textures with alpha channels close to each other in Second Life, in games this is addressed by graphic artists and a lot of time and effort. But, it stands as a reasonable example nonetheless. Eloise]

Now, there are many long term issues like this one that get passed over, some even older and more popular, but if Second Life is to be improved, and in the long term survive future competition, they will all need addressing. The question is how do we motivate LL employees to tackle the hard and/or boring issues that are important to Second Life and its residents? It would be nice to think that LL can do this internally, but if not maybe "We the Residents" need to help in the motivation process to get a better world for all of us.

Do we erect statues to hero developers based on the impact of their fixes (the more votes the issue carries the larger the statue)?
Do we have a Residents Improvement fund where we donate money to LL employees as a bounty on certain issues?

I have no idea what the solution is. All I know is the problems needs addressing and with all the talent in Second Life someone reading this article may be able to suggest a scheme to make it happen. So, do you have any ideas that might help? Just Askin'.

Tags: Bob-Wellman, Just-askin', Motivating-Developers, Second-Life, Tao-of-Linden

Reader Comments (Page 1 of 1)

Nacon said on 10:39AM 11-29-2007

I guess you haven't heard of Architecture Working Group, formed by few SL residents and some IBM residents as well.

https://wiki.secondlife.com/wiki/Architecture_Working_Group

Rheta Shan said on 2:03PM 11-29-2007

Should developers be motivated to address issues ranked important by the users ? Most certainly. Can users do this ? I don't see how, as any incentive they can offer is puny in the context of LL being the same developers’ workplace / employer. Customer orientation, I daresay, can only be implemented inside a company

The long and short of it is that to motivate the developers, we have to motivate their managers, and that means motivating the powers that be at LL to listen to us. How we should achieve that is another question entirely, one I fear will not be answered until there is some valid competition for SL in sight...

Rheta Shan
http://rhetasworld.wordpress.com

Bob Wellman said on 2:13PM 11-29-2007

Nacon

Actually I have heard of it but neverthe less I followed the link you knfly supplied to see If I could see anything about how to motivate people to grasp the hard issues fos adevlopmennt but I dodnt see anything there.

What I did see is an example of what all to often happens in software development. Version 1 of a product is developed enthusiastically and the fun/easy stuff is completed leaving only hard to solve problems to fix. No one fancies the hard issues.

Then someone says lest develop a version 2 which will be so good all the hard problems will disappear. Unfortunately when version 2 is finally in some hard old problems have disappeared but other new ones will have taken thier place. They always do.

Unfortunatly no one has addressed the real issue of how do we make any version right? By fixing the hard residue of issues a development leaves us with. At some stage we must face this problem.

I have read the AWG proposals and it looks to be like a great idea. If nothing else it will mean a rewrite of lots of bad code I am sure making the system easier to maintain and enhance. But will it solve every issue.. of course not... only motivation and hard work will do that.

So back to my original question how do we motivate peole to do the hard/improtant stuff rather than the easy/fun stuff they like?

Gigs Taggart said on 2:25PM 11-29-2007

There's a much more direct way you can influence SL Development. I've created a system where money can be donated to open source developers (initially, me), to fix bugs that interest you.

http://www.theseventhsun.com/1107_gigsCorner.htm

If 100 people care enough about an issue to donate, say $10 (US), that's plenty for most bugs.

Bob Wellman said on 3:28PM 11-29-2007

Nacon

Actually I had heard of AWG but nevertheless I followed the link you kindly supplied to see if I could see anything I had missed about how to motivate people to grasp the hard issues left by development after the fun bit is over but I do not see anything there on that subject.

What I did see is an example of what all to often happens in software development. Version 1 of a product is developed enthusiastically and the fun/easy stuff is completed leaving only hard to solve problems to fix. No one fancies the hard issues, so they avoid them.

Then someone says lets develop a version 2 which will be so good all the hard problems will disappear. Unfortunately when version 2 is finally in some hard old problems have disappeared but other hard new ones will have taken thier place. They always do.

Unfortunatly no one has addressed the real issue of how do we make any version right? By fixing the hard residue of issues a development leaves us with. At some stage we must face this problem. Management that expects people to do the hard yards to get thier bonus is an obvious way to do this but is not the LL way.

I have read the AWG proposals and it looks to be like a great idea.
If nothing else it will mean a rewrite of lots of bad old code I am sure making the system easier to maintain and enhance. But will it solve every issue.. of course not... only motivation and hard work will do that.

So back to my original question how do we motivate peole to do the hard/important stuff rather than the easy/fun stuff they like?

On a light hearted noote: I once saw on a project managers wall "the floggins will continue until moral improves". Not the solution but ironically funny. I used to do contact Project Management where I found the morale of IT departments were low because users always got on to them about how bad IT were. I used to explain to the IT staff, solve a problem the users deem important and they will stop moaning and start praising you. "Stop the floggings and Morale will improve"... LOL

qarl said on 4:04PM 11-29-2007

FYI regarding alpha sorting -

it is currently as good as is technically possible. there's no way to improve it in a realtime renderer.

if you think you have an technique that will perform better, of course we'll listen. and of course we might have bugs in our implementation which need to be fixed. but most of what you see is a problem is with the underlying algorithm - it just isn't perfect.

professional game designers have to deal with it all the time. so too SL builders.

Coherent said on 5:05PM 11-29-2007

test

Bob Wellman said on 5:01AM 11-30-2007

Qarl... I take your point that currently it hasnt been done elsewhere and that the algorithm that SL has copied is flawed.

So LL need to invent a new algorithm. All the data needed texture data and postion data is there to support the algorithm so it is possible but HARD to invent the alogrithm.

SL prides itself on being inovative. That involves doing what is not done elsewhere. Surely your not telling me that until a competitors programmers solve the hard issue of the algorithm and SL programmers can copy it, that it cant be done. Cant never achieved anything worthwhile. Thank god Tim Berners Lee wasnt of that mind or we wouldnt be blogging this, as its impossible.

This problem aside are you telling me that every long term, much wanted by the users fix/enhancement is impossible. I hoped you would address the "how do we motivate?" question rather than discuss the specifice example. May I have your opinion on that question please.

qarl said on 5:32PM 11-30-2007

Bob - i don't think you understand - it is NOT possible. basically what you're asking us to do is make 1+1=3.

i understand that the layman doesn't see why this is so. and i can kinda understand why you might not trust LL when we say so - but please, feel free to contact any professional graphics engineer.

and about your larger question - YES - much of resident frustration hinges around these sorts of misunderstandings.

LL is chock-full of smart, motivated people. the reason SL sucks (to use the popular word) is that these issues are genuinely hard.

if you like, let's take this discussion to email. it might be good to discuss more specific examples.

K.

Xugu Madison said on 6:58AM 12-03-2007

It _is_ possible to fix. The algorithms to do it would make your head hurt, and require detecting that the prims are both infront and behind each other, and then splitting each prim into smaller renderable parts which are either only infront, or only behind, but it can be done.