Avie Worn Greet Script

I want to create a scripted object that I'll wear on my avie to say hello when someone comes within a radius of my avie. As a newbie scripter, I can already kinda see from the other examples how this could be done. The problem I'm having is the llSay command always puts out the script generated text to channel 0 in green, so it'll look funny. How do you change that so it will look like normal chat text spoken by the avie? I've seen scripted objects around SL that seem to do this, but I don't know the trick they use because they aren't sharing.

I don't think you can. ANYTHING a script says should show up as being said by an object - hence the green colour.

The fast ones that do pictures and things in chat are actually gestures. I'm pretty sure there's no way to trigger a gesture from a script, although it is late and my brain's messed up so there might be.

It's possible to make the avatar say something by bringing up a llDialog box and having the avatar choose one of the buttons...

...the avatar will then "say" the button-label. It's a dirty work-around and definitely no good for what you're lookin' to do...but I'd say that's the #1 way people are making objects appear as if they're avatars talking.

Just thinking out loud...

Is there a way to use /me chat command?

"Sydnay Alexander says hello to "detected name""

This may have the same issues as the llSay() but may make the text appear differently...

Is there a way to trigger a gesture from an attached script? Maybe flagg the permissions to yes when attached and then a sensor triggers the gesture that could "say" whatever you like?

good luck with this. It seems the format of the ouputted text is the biggest concern...

Sydney

you can use llSay to do an emote or speach but as it is actually generated by the object it is in green. I have not found a way around this beyond the aforementioned use of a llDialog which requires player approval and is limited in character length as scripts may not execute gestures. It may well not be possible as I doubt the Lindens would want objects pretending to be Avs, especially regarding agents other than their owners and there are no related permissions.

Very good point!

The gesture thing...I think...would be possible though...

...it would just be like a dance-machine that people wear...except that it would trigger gestures...

You can trigger animations but not whole gestures, as far as I know.

I can't see any other way to do it than using llDialog. When the object detects somebody, it pops up a little box saying "Avatar XXX in talk range, what do you want to say?" and then you have buttons marked "Hello!", "Piss off!" etc. It could even make a noise and animate you when you were doing that.

But honestly, unless I developed some sort of keyboard phobia, I'd prefer to just type "hello".

I would agree if it is not automated then why bother. I have never tried anything like this so I will trust in what you are saying.

Hmmm...I'm unfamiliar with gestures...assumed they were just glorified animations...but yea...now that I look at 'em, it makes sense that they wouldn't trigger on an animation-trigger.

At 110wpm I've got no fear of typing, but divide that speed over a dozen people IMing the host and flitting down into your dance area simultaneously and it is a bit overwhelming. I just want an automated assist. Its something us IRC refugees are just used to.

Also, the full-override on chat, for something like an in-line spell checker for those simple, repeated, keyboarding errors (e.g. fixing "teh" to "the" would be soooo nice.

Forgot to mention that I posted the llSay() issue as a feature suggestion and someone else suggested allowing a full chat override, like an animation override, which is a much better idea. That could open things up so that you could have scripts for spell checking, develop chat short hand (e.g., you could expand all those annoying chat acryonyms as you like). There would be a lot of possibilities for new utility and fun scripts that way.

Is having the text white mandatory? IMO I'd prefer to have it remain green as that way people are more away it was an object that said it and not me personally.
This avoids almost everyone who turns up from immediately jumping into a conversation with me, which, although not bad, could become difficult if everyone does it

Just make sure to keep a list of people nearby, so you don't keep greeting them over and over, keep it fairly high-ish like 50 names. Although sensors on return the closest 16 results, people might drop in and out of the closest 16 so would keep getting greeted that way too!

You can set the chat color in your client's edit->preferences menu. So I don't think you have any control over what color something shows up as on someone else's screen. The white for avs / green for objects is the default setting.

Absolutely, i'm storing a list of avies and a timestamp on when they were last greeted. They have to go over a threshold get a welcome and then a welcome back, so if they are hovering at the edge of sensor range they don't trigger a regreet, just update when they were last seen.

Since color isn't going to be something you can control, why not target the message more. Instead of using llSay() use llInstantMessage(). You can then make it so that only the new arrival see's the greeting. It'd be a lot less spammy in busy areas.

Don't worry, when objects IM avatars it just appears as green chat text. It doesn't actually open an IM window.

I like this idea. Makes it more personal and as Harris says "Less Spammy."

Well, that's a useful implicit behavior I didn't know about.