Linden hotline

I actually agree with Jarod here. Not in an "Why the fuck aren't you doing work, LL?" way, but in a "Sheesh, it's been how many months now and they're just now giving us THAT?" sort of view.

For example: llDialog now does 12 options instead of 4. Whoopdie friggen doo. It took them HOW long to add on the option of eight additional buttons? Seven MONTHS? Yeah, that sounds about right. Seven MONTHS to add on eight buttons. Was it really -that- difficult?

How long has XML-RPC been out? It's been broken since the day it was introduced. We were promised that XML-RPC would be fixed... what, a few 1.3.* iterations into the build process, right? Somewhere around 1.3.3 or 1.3.6 I believe it was said. And it's been how long? Ten months now? And we're just now moving into 1.6? And the "fixed by" date kept getting farther and farther away, to the point now that it doesn't even exist?

Gesture "revamps". Bah, I say. Bah. The gesture system was reworked to include the new animation system, and when it was reworked it removed functionality. Unless I've missed the option somewhere, we can no longer set ourselves Away/Busy via a /command. I -know- we used to be able to, because I have my (now empty) /afk command still sitting in my gesture list. Sure, minor useless point, but I certainly didn't ask for a gesture revamp. I could have cared less. Hell, from what little I've used the "new" one, I think I liked the old one.

Animations: Fuck that. Animations are a marketing tool, and nothing more. Sure, I'd like LL to get more money, so it makes sense that animations get implimented. But seriously folks, -animations- are the jewel in your crown right now? Give me a break. Animations are nice for other people, sure, but I could give a rat's ass about them.

llWriteNotecard has been asked for for how long? Since before I came along, definitely. Beta, then? And it's been officially stated that it's "not happening for the forseeable future" or whatever it was they said? Oh. Great. Just kill off one of what could possibly be the greatest tools since the development llListen(). Whatever. I don't care. (see sig)

Ratings: Yeah, 'nuff said.

Frankly, I'm surprised that LL thinks it has a good grasp of the desires of the community. Then again, maybe I'm wrong. Maybe Philip really does get a thousand emails a day asking for "Oo, can you make it so me and my boyfriend can hug each other?" (bad grammar on purpose, of course.) Who knows. *shrug*

To be clear, I agree with Jarod's and Moleculor's positions. I've only been playing 6 months, I'm just starting to try to do serious scripting, and I keep bumping into limitations that cause me to say, "You're kidding, right?" llWriteNotecard is a classic example.

On the other hand, I just found out this week that another game I'm quite fond of has had its sequel delayed for another six months, apparently because the marketing division thinks the first game hasn't sold as many copies as it could, yet. This makes no sense to me whatsoever, because the game series involves specialized hardware only sold with the first game, so a sequel release should only improve sales of the first game, not cannibalize them. This is also at least the second six-month delay in release for this game -- when I bought the first game a year ago, the sequel was supposed to be out a few months later, and I bought it partly based on that premise. That company hasn't bothered to announce the delay in its discussion boards, has a newsletter it hardly ever bothers to send, and generally does a really lousy job communicating with its installed base. There are no Robins and Jeskas and Uncles over there. I once offered to handle community management for them for free, if they'd make the arrangement official so people would know what my status was. They said yes and then never bothered to make the announcement. I don't spend time on their boards anymore.

So while I'd like better response from LL, by comparison to other situations I'm in, LL looks pretty good, even though they still haven't answered my hotline post after two days, hardly ever seem to read or respond to feature suggestions, and when they do respond, they often choose bizarre things to "improve."

I don't play other online games, e.g. MMORPGs. Are their communication patterns better or worse than what we see here?

neko

per bens' answer in this hotline thread, and cory's recent spoon posts at terra nova, i'm of the growing opinion that ll just see all this as from the user perspective. what ben means is that they'll implement outbound xml/rpc when they're sure that you won't be able to use it to damage the community or cripple the economy. those patent answers are substituted regularly when the concept of developing a useful tool flys over the heads of everybody in the office.

for example, outbound xml/rpc can be implemented with a limitation of say 512 bytes. that's plenty of space for a programmer to embed a few initialization codes and search terms to poke at a server. how is that going to damage the community and cripple the economy? it won't. what it will do is take data oversight away from ll. they must be really appalled that these web shopping interfaces are recording transactions that they can't see. so far, they've done everything possible to thwart such efforts.

i'm afraid the answer to your question is "never."

Not in Second Life anyway. XML-RPC allows us to tunnel through Cory and Philip's rat maze, and they can't have that. It lets us bypass a lot of limitations imposed by Lindens. It lets us integrate Second Life into the web. It would let us chat with people who didn't play Second Life. It would completely uncork Cory and Philip's ship-in-a-bottle microcosm, and they can't have that.

Everyone of those features that Philip said was a priority is something that makes Second Life more insular. Want to listen to music, why run iTunes when you can just sit on land with an audio stream? Want to express yourself, use an animation to show it. Nothing they have offered makes Second Life a tool, it makes Second Life a gated community.

So, frankly, I don't expect out-going XML-RPC. That would make Second Life much-too-much like a piece of Internet software, and not like the virtual world Cory and Philip want it to be.

rat mazes aside. ll is not the little innovative startup it was a year ago. the company is too well established for the employees to still be giving us cushy catch phrases in place of valid reasons that useful features aren't being implemented.

if ll were as incompetent at planning and coding as they seem, second life wouldn't be what it is today. so obviously, they're well capable of planning and executing whatever it is they want to do. it's time for them to either come clean with the development community and explain the plan, so we can plan, or bar the door and let us know that we're wasting our time asking.

either solution will work. we of course prefer the former although noone can argue that the latter has worked for most companies over the years. just please stop jerking us around with answers that don't make any sense e.g. we all know you could implement outgoing xml/rpc tomorrow, with very limited capability, and ramp up to a useable point without endangering anything. so tell us seriously why you won't; or tell us seriously that it's none of our business.

Khamon,

If it helps you decide, I successfully negotiated with Cory last October to contract for Linden. However, as the relationship developed, it became clear that the direction he wanted to take SecondLife was pretty far from the direction that I want it to go in, so we moved apart. The idea of being simultaneously a random code monkey and earning a game-dev salary didnt really appeal somehow

Obviously, it would not be ethical for me to state what he wanted me to work on, but clearly if it diverged so far from my own goals that I was willing to terminate the relationship for it, its safe to assume that it had about 0% in common with an API or any of the other things that might interest you or me in having implemented.

Azelda

It is always hard to balance infrastructure features like XML-RPC with new features that make SL look/feel/act cooler. Both benefit developers directly in different ways, right? So we try our best to strike a balance. Since everyone's priority lists are different, and in many cases substantially, we are 100% guaranteed to piss off some people with every release. I guess that's just life in the virtual city.

Regarding outbound XML-RPC - we are going to do it. We need to choose a mechanism that rate limits the traffic. Best thinking is that we should charge L$ for transmissions as a way to do that. Neither is a lot of work - it is on the list.

you can also limit the number of servers an av's scripts can contact to say around one. if we need multiple servers to establish inbound connections, we can contact the one with a code that tells it to tell them to initate contact with inworld scripts.

all we need is a way to get a bit of data out of the system. put the burden on us to manage it all in our own systems from there. the whole point of outbound xml/rpc, and an api, is to force us to develop and support all these cool features for you on our time with our own equipment.

and kudos to you and cory for replying seriously. it's exciting to know that'll y'all prefer this option to clamming up or treating us like children.

No, that's my beef entirely. Streaming audio on the landscape, animations, and junk like that benefit (or appease) the user, not the developer. Things like XML-RPC, an API, and the like benefit the developer because it gives them more options to work with.

This is why I keep harping on an API so much. You could have saved yourselves a lot of effort if we'd had an SL client API months ago. That streaming stuff could have been done in the client: by touching a prim you got a notecard with an address, you could plug that address into the SL client, and it either plays the stream directly or launches something like RealPlayer. With an API, items on your ToDo list, like the new rendering engine, could be done by us developers; they could test out different engines, see which worked best, and then distribute them for you.

Which leads me to...


Absolutely, you can't please everyone. However, from reading what I have from Paul Graham and Joel Spolsky, pleasing the developers should take first priority because -- and here's the clincher -- it allows us developers in Second Life to create better and more stuff for the consumers.

SL is touted as a world where the user is the creator, but a lot of the things SL has created lately seem to be designed in such a way that it feels we're better off sitting back and waiting for Linden Lab to develop the world. Linden Lab needs to please the developers, the developers will please the users. That's how software platforms work, and as much as you're loathe to define Second Life in terms of software, it is a platform.

A darn good one too!

you stop that jarod. i've well earned the title of "most blunt & cynical one" and you're not taking it away from me. seriously lindens, this is as simply put as it can be. content is far far more than prims and textures. we can't build the world if we can't trust lsl to be backward compatible or if we have to wait for ll to implement every little feature.

if you want us to be the creators, you have to empower us to create something we know will work tomorrow; allow us the capability to use our own creation tools and import content; and let us construct our own clients to interpret data streams from your servers. you also have to talk with us about these things, as you have the past week, openly and seriously. otherwise your claim that the residents design and build the world is simply not true. we greatly appreciate your comments & questions and sincerely hope that the dialogue will continue.