Linden hotline

Perhaps it might be a good idea, in the future, to have Lindens answer questions, even ones slightly "out of the box," seriously. Just a thought.

Done!

Has anyone else notice that the forum, having been changed to post only, seems to be a great way for people to post stupid ideas that only they want without having them debated in the feature suggestions forum, AND have them replied to by a Linden, while most feature suggestions go unacknowledged.

Is this really what you wanted to achieve, Lindens? Because I have a shitload of suggestions on my personal agenda to make SL what I want to the detriment of those other assho.. residents.

I'd say we need better tools to discuss feature lists.

But then again, I'm saying this for perhaps 3 months or so, asked the Lindens on the Town Hall meetings, and tried to "spread the good work" about all the private initiatives of setting up "feature lists". The net result was, too many ways to post your own ideas, too few ways to make them "official"...

Post Only is better for that forum. To get things debated, we already have FS. H2L is good because it gives Linden a way to respond to single questions without having to read through pages of trolling and bullshit, and it gives us a way to find their responses, also without having to read through pages of trolling and bullshit. If you think something in H2L merits debate, you should post about it here or in General!

Right Huns, but now people are going straight to that forum to get their personal feature wants validated by a Linden without having to have it discussed. And yet most everything in this forum goes largely ignored, or that's certainly the impression. That was my point.

If thats what they want it to be used for, fine. I have 15 months worth of feature suggestions I'm going to post in there, because I want them, and I don't want you lot telling me I can't have them... after all, I have a Hotline to Linden!

Kris, the hotline forum is to ask questions and throw out ideas for a quick response. I've found so many places where people ask questions but they're buried so don't always get answered.

The other forums are still the best place to discuss options and ideas for new features. Plus we're working on a way for residents to nominate and vote for features based on some work Azelda Grace did. That should be available for testing late next month or early April. The goal is to use the tool as a way to help us know what resident priorities are so we can factor that in to our own development planning.

Robin, I hope this tool you're working on is in-game, rather than a forum only tool. While the forums are very active and vocal, and offer some productive discussion at times, they don't appeal to all of the SL population, and you may end up with somewhat of a skewed view by taking them as your only source of information.

If they're in-game, however - wow, what an awesome idea

Yep, that is the idea Travis!

Eventually this MUST take place in-world but the first instance will be web based. No, not in the forum system - completely separate thing.

As soon as we have the tools to deal with that much information in-world in a way that is easy for a human to navigate, make sense of, and interact with then it will move in-world.

Robin, thanks for both the answer here and on the Linden Hotline. I'm really, really glad Azelda Garcia's feature voting site has given you "food for thought" to develop something "official".

As to in-world vs. on a Web site, I definitely haven't any "preference" myself. Ideally, everything should be in-world - but since in-world browsing will be implemented in the yet distant future anyway, I guess the question is moot

The trouble with any voting system where the masses vote is that it becomes a matter of how you ask the question. Suggestion that would make SL a better place with no downsides will get equally ignored and suggestions for more privacy or better security will get noticed.

<rant>
I've made about 20 feature suggestions on this forum and have 300 posts here as well. Of all these threads I've probably seen less then a half dozen come true. I've been here for almost a year.

I have no more expectation that LL will listen any more to any new system then they have the old system. The only times I've seen feature requests come to fruition is when a Dev shows interest. And the Dev's generally don't read the forums; they are busy working on Havok 2 and the new renderer and new permission system and new rating system. Oh wait we haven't seen any of those yet. Then there is XML-RPC still not finished.

And if you were really impressed by Azelda's work I would expect to see some of the suggestions implemented.
</rant>

But maybe I'm wrong to rant, new people have been hired and funding seems secure for the moment. Maybe things are changing for the better. I sure hope so, the track record is abysmal.

Strife, how are you computing that we are going too slowly? We are adding great developers as quickly as we can find them (we've hired 4 in the last couple months alone). The dev team that has produced most of the features over the last year of so has been 6 people. The rate of development in terms of features/dev/time and our quality in terms of getting the order of priority right has been second-to-none, I think. Do you disagree?

1. Because the XML-RPC system has been up for six months, and it's still half broken. You can't implement calls from in-world, which is why most people wanted the thing to begin with: so stores could update websites, so we could send keywords to a web-based search engine for a more robust search, so we could blog from in-world.

2. Because llWrite2Notecard() has't been implemented yet.

3. Because we still don't have an API!

4. Because permissions still suck donkey balls.

They may be churning out new features, but the features are for crap. They're crumbs like animations, the ability to turn off the red-bars, and streaming music that are aesthetically nice, but no where near as juicy as the steaks we're asking about. I appreciate that you have a limited number of coders, but I don't appreciate that they're ignoring 90% of the good stuff people suggest in favor of the piddly things that we could do with out.

I don't know about Strife, but that's how I computer that you're going too slowly.



I disagree. They're second-to-none if you think of Second Life as some kind of psychological rat maze, which I feel you and Cory do. However, for people who want to do Neat Things with SL -- on the order of the kind of stuff you can do with Apache, iTunes, RSS, and such -- the features/dev/time ratios stink.

I dare you to name one thing the devs have given the coders, hackers, and SL app developers in the last four months that they've asked for. Can you name anything?

It just seems slow from our prospective. I've always known the operation was small but wasn't aware that the dev team was in the single digits. It's good to hear that the number has entered the double digits. I've seen some of the scripts written by Lindens before they were Lindens and they are excellent. Maintaining two code branches (Havok 1 & 2) must be mind numbing. Looking at the change log on the Script Wiki it is obvious that allot of good work has been done. I know that the focus for the last few months has been bug fix's and not new features I should have remember this when writing. My rant was skewed (shouldn't rant before sleep). It's easy to criticize others when you don't have all the facts when you really shouldn't. I think the priorities have been in the right order. Allot has been done in the last year:

1. Custom Land Textures
   
2. Animations
   
3. Revamped Gestures
   
4. New Primitive Types
   
5. Extended the Old Primitive Types
   
6. A slew of new LSL functions
   
7. Old bugs fixed (and new bugs introduced and then squashed )
   
8. Streaming Music
   

I was a little harsh in my rant

And when the company president comes and posts on the forums it reminds me the Linden Labs is a different type of software company. When I joined SL the first time I talked with a Dev I was amazed; I was given straight answers. When I was in Uru the Dev's would roll-play their characters (as they were part of the games plot) and would pretend not to have a clue to what you were talking about (leading to creative explanations of bugs; lag becoming highly viscous air). I don't know any other software company so open as Linden Labs; I wish it were like this everywhere.
My apologies for being narrow minded and jumping the gun.

Strife, so what is the alternative? Having a selected group proposing the feature changes "for the greater glory of the SL residents"?

Khamon Fate tried to set up something called the Second Life User Group - an organisation to get together all the residents interested in discussing features and technology related to SL, and acting in a coordinated way to propose common ideas to the Lindens. Actually, almost all types of software platforms have user groups - from Microsoft through Oracle, from Linux to supercomputer users. SL is unique in that we have no real "representative forum" of the community of users. What we have is anarchistic self-representation: I, me and myself, only, alone against all others

I fail to understand how an anarchistic society can organise itself in order to promote a "common front". What will always happen is that small groups who are not anarchistic will be seen and scorned by the rest of the community.

Unless, of course, the small groups grow

We "complained strongly" about this when they disabled it 8 months ago.
llSetPrimitiveParams now correctly sets the color/texture on attachments.

pretty sure we asked for this...
llDialog increased number of options from 4 to 12.

We asked for this.
Editing the position/rotation of an object that is receiving constant updates from the servers is now possible in most cases.

We asked for this.
llRezAtRoot -- Rezzes an object using the last selected root object's location as the rez position.

We asked for this.
Recent Bug Fix: Integer hex roll-over to negative now works. 0xFFFFFFFF==-1

We asked for this.
order of declaration of functions with parameters is no longer an issue. That is, the following compiles:

We asked for about 50% of the changes to LSL at some point or another. Only trouble is most of them are bug fixs that only got delt with after 50+ posts.

When I'm wrong, I'm wrong, and I guess I was wrong.

Jarod: This was what I was getting at the things we *really* wanted haven't come yet.

Gwyneth: We setup a game development group a i think 6 months ago with the goal being to lobby LL for new features that would make game development in SL better. The group contains many of the creators of the games in SL. As a group we collaborated together to come up with a set of recommendations as to what features would be useful and sent it to LL. Many of the features are on Azelda's feature voting page. None of them have been implemented. But thats the past.

It seems to me things are changing for the better.

I'm glad it seems that way to you, and I don't mean to be a downer to anybody, but it irritates me that Philip thinks "getting the order of priority right has been second-to-none." I can understand them being slow, even on a large team it takes time for the wheels to turn. However -- and maybe we're saying the same thing, I often have to say something before I realize I'm parroting -- for him to think that the bits they've given are in "order of priority" just bugs me.

Custom animations, in my opinion, should not have come before XML-RPC was worked out completely. Custom animations are nice for people who want to hold hands, dance, and hump like rabbits; however, they're trivial for Second Life as a development platform.

If Philip said, "We've got a limited number of coders, the greatest demand has been for X, so we've had to do that," I wouldn't mind (as much). The fact that I've never seen him say that (or anything similar), you and other developers have so many requests that are seemingly ignored, and Philip thinking their priorities are in order bothers me.

Perhaps Philip and Robin get five hundred emails a day where people ask for trivial crap like streaming music, and my rants and complaints are based on the false belief that they aren't. Again, I'll cop to if I'm wrong. However, I don't remember ever seeing half as many requests for animations as I have llWrite2Notecard(), and yet Philip says that animations were a priority.

I doesn't jibe for me.

Well yes we do have a few niceties, but the fact remains that SecondLife today is pretty much the same SecondLife when I joined a year and a half ago.

Yes we have custom animations for avatars, but they were never something that I felt was missing.

Yes we have xmlrpc, but wouldnt it be nicer to have a shellserver directly at Linden Lab?

...or to be able to write scripts in a standard language with things like arrays, imports/#includes?

...To be able to lock land down sufficiently to be able to write persistent game worlds relatively painlessly?

Why are there so many basic physics functions missing? Even ODE offers the ability to directly set the linear and angular velocity. Surely Havoc offers this functionality???

Sensors are quite basic.

Etc.

Sure, its good enough to survive, but theres definitely a lot of motivation to create something better.


Philip wrote:

> We are adding great developers as quickly as we can find them

Thats not quite true, because your recruitment pool is essentially limited to Americans.


There is a business opportunity here; and I'm thinking quite seriously about how we can exploit it. Well, you've seen my experimenting already with OSMP. Consider that a prototype


Azelda

I see I wasn't the only person to notice this opportunity.

I heard it was limited to Californians due to some tax laws.

... remember the Wisdom of the Old USENET Posters: when furious, write your inflamed reply, but wait 24 hours before sending it. If it still makes sense the next day, go for it. But you'll find out that 90% of the things you've written were driven by pure passion the next day

Anyway, sticking to the point. Kris sounded pretty angry because now LL has come up with a system that answers age-old questions in minutes, hours or a few years, instead of being left unanswered, simply because the mass of good posts/information is buried deep in the forums and third-party web sites. The (understandable) result is that suddenly all sorts of questions are being asked and answered - and Linden time is to answering them is invested in the age-old questions buried deep in the forums...

How often do we get the following suggestion: "hey, I got a great idea, why can't we subtract prims from other prims? This would be awesome! We'd get many more options! I'm so cool for having thought out this by myself, don't you think?" A quick search on the Feature Suggestions forum would show that this question pops up once per month. Why don't people simply search the forums before posting their "incredibly awesome idea they just had"? The answer: searching the forums and doing research on 60+ third-party sites just to find out if you're submitting an original idea takes lots of time. And quick & dirty forum posting is suddenly not "fun" if you need to do so much homework first. Tell me the truth: how many of you have ever read the posted minutes of a Town Hall meeting with Robin Linden about marketing issues in 2003? And why should you? Because even there several very interesting things were said at that time, that reflected LL's position, and that would reduce forum noise by 10% if people just did their homework and read those things...

Another example: I spent 3 or 4 days with a new resident who was asking me if any of his brilliant ideas for in-world games have been implemented (they all were). After a while I just pointed him to Pirate Cotton's The Game SLave. It's pointless to search for the forums in search of that kind of information, and I can't possibly be aware of everything which is done in-world. So, I'm glad that Pirate has had the amazing idea to try to concentrate all information of that particular kind in a single site! That's the kind of things that I find lacking in all this "feature suggestion" discussions - no information. Just data. But nobody looks at data!

Well, about LL's utter amazing workforce... didn't you all know that there were only 6 devs in SL's team? Weird, I got that information after being 4 weeks in-world (and at that time, it meant counting both Philip and Cory as being part of the dev team). I'm still amazed at the number of established residents who still think that LL is a software development giant the size of Microsoft (or, well, the size of EA anyway...). Again, how did I find out? Search for "Linden" in Find | People, and count the number of developers listed there. They're not afraid to tell the world what they do in SL. You can get the whole hierarchy of Linden Lab from there in an hour or so of homework. It's all public, nothing is hidden, you couldn't ask for more information than that. Really, you can even read Philip's Excel spreadsheets about SL's economy to have an idea if LL is a profitable company or not, since you know how many employees are there, assuming the average wages paid in California and the average costs for infrastructure (co-location). I'd really like to know how many companies in the world are so "free" with their "inside information" as LL. I don't know of any other, but then again, the number of companies doing 3D virtual realities for the Mac (or Linux...) are very few

(then again, LL has 3,000-5,000 volunteers happy to contribute code as soon as they're allowed to, but that's a completely different story, a different thread, and completely off-topic here)

Back again to the point. LL's time is seriously limited. We can help by reducing noise in the features' proposal - ie. make sure we concentrate on a "common set" of desired features, instead of putting up feature suggestions across the World-Wide Web, and "expecting" LL to read all that stuff (and being seriously annoyed that they didn't). the point is, the less LL has to read to get our proposals, and the less WE have to read to make sure we're not simply suggesting anything which has been asked for 12 or 18 months ago (but that we never heard of!). Since we cannot organise ourselves to do so (anarchy and an overall distrust of our fellow residents will prevent that), the only solution is engaging LL's help to get all the features organised. For me, this means an Official Linden Central Feature Suggestion & Voting System - a single place where all the information is stored - feature proposals, discussions, voting, and implementation timeframes. This would probably look like a mix of Azelda Garcia's site and the old USENET proposal system for new newsgroups. Easy to implement, has been around for 20 years or so and thoroughly tested, and there isn't any need of much "technology" - it's just "discussion procedures" with well-established guidelines and timeframes for submitting proposals, discuss them, and vote them (and eventually resubmit them after 6 months if they didn't get passed the first time). Trust me, the system works very well! The only major problem of doing something like that is getting an assessment by the Lindens on how "hard" a feature is to implement, and put a certain "weight" on it. Watching the strategy of actually getting a feature to be implemented will be a very entertaining thing. Imagine the following:

Proposal #1: open source the server code. Timeframe: 1 year. Development team commitment: 100% (full staff just doing that)
Proposal #2: raise the number of avatars in a single sim from 40 to 42. Timeframe: 5 minutes. Dev team commitment: 0.001%

So instead of voting #1, you'd get a feature like #2 "almost instantly", but it's actually pointless. However, if enough people would vote for #1, this would mean that for the whole year we wouldn't get any further features Discussing this and trying to get a commitment from the residents to make them understand why some things are just "impossible" to implement would be an amazing exercise in rhetoric, and I'm all for it

Anyway, the first step was trying to get the "bug queue" organised - and we now can put a checkmark on that And for the record, this is also "old news" - the Lindens are seriously thinking on how to implement "centralised information". Perhaps they'll take months, but it's definitely not forgotten (hint to Philip: outsource that project to the residents You'll get it up and running before the end of the month for as little as a 4,096 sq. m. plot without land tier fees )

- Gwyn, who just happens to know that this is just one of 200,000 other posts

Gwyneth: How does what you are suggesting compair with the feature proposal page on the support wiki.

I don't see a voting system on the feature support wiki. I've added items there, and I've even included URLs pointing to threads here in Feature suggestions where I knew of them, but I don't have a lot of confidence that the support wiki feature page will ever go anywhere. No process is described for prioritizing or responding to the suggestions there.

On the other hand, the Feature Voting Page is a lot closer to what Gwyneth is talking about. I've participated in USENET new newsgroup proposals, both as proposer and voter (I was subscribed to news.announce.newgroups during the rec.arts.music.white-power debacle, for example), and I agree, that's a system that could be referenced for our needs. A difference, however, is that USENET actually was (and still is, I guess) somewhat governed by its citizens, as much as it could be said to be governed at all. SecondLife is goverend by Linden Labs. I think it's great that LL wants our input, and I'm happy to give mine, but at the end of the day, they get to make the call based on what will be best for LL, whether we as users agree or not. We may happen to think that better API and admin tools would really help LL become/continue to be successful, but if LL has done some kind of analysis and it looks to them like custom animations are the ticket to rapidly growing the user base, that's their call.

neko