Why can't scripting be done via a "gui"?

That said... Intellisense FTW!!! \(^o^)/

OK, so this is actually considerably closer to what I had in mind. Tho, I think it could be extended a bit further. I think a prim, or linkset could be set to respond to some things, like a touch, for example.

3D rad and Game Maker let you set a lot of object properties or parameters with with windows that are not unlike the edit windows we see while editing prims. But they both support traditional scripting.

@Johann. I'm not looking for the same level of complexity. But rather, as Argent seems to have grasped, adding more options to prims that are only available via scripts now. But I'd take it a little further by allowing the prim respond to certain events.

Also, I'm a Dvorak touch typist. To the point that I haven't bothered relabeling my keyboard from standard qwerty in years, because I simply don't look at it.

@ IMMY. OK, so my idea can be taken too far. And I'll admit I've seen the same sorts of problems in Mindstorms.

Umm yeah. That would be nice too.

It already can be. The thing is, once you're doing something active, responding to an event, you really need to be scripting. It's MUCH easier to describe actions as a narrative than hooking together components with wires. People used to have to do that, programming computers with plugboards, and later on programming simple control systems with relays... my wife was doing relay ladder logic well into the '80s. Scripts completely blew away that kind of interface. Or... for laying out circuits, people don't design electronic circuits by taping them out any more, if they can avoid it, they just define the inputs and outputs and get a computer to do the layout.

I've seen what people do using old-fashioned plugboard-style interfaces in Little Big Planet, and it's amazing what they can do... but it's amazing that they're doing it *in spite of* the interface, not because of it.

Yes, Mindstorms is a perfect example. First thing I did was get an NQC environment so I wouldn't have to deal with the pseudo-plugboard interface.

One exception to this is when you're dealing with an inherently dataflow environment, where there's a small number of operations being performed on a stream of objects, without any kind of complex behavior. But even there I've been disappointed in practice. The UNIX command line, for example, would seem to be a perfect fit for a graphical interface but the implementations I've seen have been pretty poor.

One thing that I think WOULD be nice is if they could implement Puppeteer natively on the server. Let me re-arrange the prims of an object, say "save as state 1", and then in script, just say "go to state 1". Yes, I know Puppeteer is a resident project and it doesn't deserve to be GOMmed but it's also really inefficient - it HAS to be, in order to remain as general as it is. Using Puppeteer for a door would be huge overkill, but it doesn't need to be if it can be streamlined into the server.

But, there again, wouldn't it be nice if I could build my own house by clicking to start building, have a front door appear (from a themeset of my choice, including ones made by residents and sold), click on the front door and choose the size and type of room I want behind it, and have it automatically fitted into the primset? Then continue building more rooms by that means?

Honestly, there's so much in SL that could be done so much better than it is; that's why I now just allow myself a little laugh when people say SL has no grinding. And I bet that the real reason why it isn't more convenient is that if it was, people would get too quickly to the state of "having everything they want" - and realize how hollow SL becomes at that point.

The reason you can't build houses in SL using that kind of interface is nobody's written the code to do it, in-game or out-of-game. If there was enough demand for that kind of thing to justify the enormous amount of effort it would require to make it work and work well, someone would do it so LL could GOM it.

And part of that reason is that LL are the only ones who could make it work really well, because it would require storing mass as an abstract and mapping it to prims, which LSL is nowhere near sophisticated enough to do.

I quite fail to understand what the relevance of "storing mass as an abstract and mapping it to prims" is to building a house, not why LSL is unable to do it. The greatest problem with your scheme is not the difficulty of implementing it, but the effort of creating all the possible room templates, of sufficient quality to compete against handcrafted rooms and houses.

To illustrate why this is an issue, consider the Linden clothing model. It is precisely the model you're proposing... one selects an item of clothing, a pattern, and adjusts the size and shape of the clothes with sliders. For some time I made all my clothes this way, and pretended them not so far inferior to those created by hand, using custom clothing textures and prim parts. I was not even fooling myself, and if you stop and think about it I hope you will realize that you really can't believe that any kind of practical "color by numbers" approach to building anything will produce near the variety and quality that one finds in SL.

In order to build the house with the optimal number of prims, you would need to work out what shape the house would ultimately need to be (including decisions made by the user) and fit prims to it. Using prim prefabs for each room wouldn't give the necessary customization and would waste prims. Now, I surely wouldn't like to do that in LSL, but a language with better support for structured data could do much better.



It might not have the same artistic quality, that's true. But you did have _some_ reason why you made your clothes that way, didn't you? And I personally believe there are a fair number of people on SL who would rather have some involvement in the creative process and a rather more generic result (since even a template could be very high quality), than an awesome custom looking house which they had no involvement in creating other than picking it from a catalogue and taking their money out of their purse.

But I know that's not a good attitude on SL. Apparantly, asset quality counts above all, no matter what effect it has on the end-user experience..

A slight side note to the comments in the beginning of the thread about including radar and AO in the viewer: The Emerald viewer already does this.

Also, freebie stuff doesn't necessarily hurt commercial products either here or in the real world. It does force commercial grade developers to make sure they develop superior products because they have to give someone a reason to pay for something they could get free. Most do fine at this.

We now return you to the main topic of the thread .

As I mentioned on another thread, not always. Witness the dominance of MS Office; most users get it for free with their PCs, so its only real competitor is OpenOffice, which has to be also free in order to compete. Technically a competitor could add features that people want, except that.. um.. well, you suggest one for an office suite, that people would actually care about..

1. So what? It's never going to be as good as having actual humans lay out the prims. And it's all going to look the same, anyway, because of the limited palette.

2. Who says the heavy lifting is going to be in LSL? Or even in-world? That's why there's HTTP reading and writing from LSL.

I was cheap.

And the Linden clothing is much easier to design, and much closer to fully custom clothing, than automated house building could possibly be... simply because avatars are more alike than houses.

People have paid plenty of money for games with both of these restrictions. It might be less good-looking, but at least they get to _do_ something. Instead of their entire involvement being financial.



Yes, but I judge LL by the service _they_ provide - and which they charge a great deal for.

Actually, that's not quite correct. Watching that video by no means tells all you can do with Rev. Rev can address objects by ID, name, layer number, object number, and some other ways as well. The language is very rich, much more than that video can demonstrate. There's lots of ways to drop things on other things. And many very difficult concepts are one-liners -- for example, connect via a socket to a server, grab the contents of a file, and display it intact on screen: one line of code. Been making my living with Rev for years and years now, and my Mote Particle Generator was written with it.

Back to my corner now. But yeah, I'd love to see LSL with a Rev interpreter. The power of that language is awesome.

But not for games where you have professional quality builds to compare with. If LL did this, you'd come back and say LL sucks because their home builder doesn't produce homes that look as good as you can build by hand so people who want to "compete" would have to compete with professionals and could never be satisfied.

People CAN do something right now, and some will end up making stuff that's as good as the best. AND it will actually be their stuff. This is just more of your idea that people should be able to be rock stars with simulated audiences, or club owners with people in the clubs never even seeing their club.

IF I'm wrong, and this would be a great product that people would use, why don't you build it and sell it and make yourself a winner?

That's a complete nonsequiter.

I wrote "if there was a demand for this then someone would have done it".

You said "It's too hard to do this in LSL".

I responded "You don't have to do it in LSL, you can do the heavy lifting on your own server".

What the HELL does that have to do with judging LL or not?

But it is still a textual language. You need to actually write code to make the program responsive, right? It's not Amigavision.

Yes, but most will not. Moreover, those people who _do_ end up making stuff as good as the best will very likely have to devote their whole effort to it, which might not be a tradeoff they want to make.



My principle is quite simple. SL has a 70% drop-out rate, and LL don't like that. But if SL endeavours - not necessarily busniesses, but general endeavours, for whatever definition of success - have a 70% or higher failure rate then.. well, what do they expect?

And a 70% failure rate ISN'T HARD. If everyone who owns a club has 3 others dancing in it, that requires at least a 75% non-participation rate for owning clubs. That might not be a failure rate - because some of those non-participants may not have ever _wanted_ to own clubs - but it _could_ be, and you don't know that it isn't!

LL have shown _their_ way of solving the problem. Their way is to bring corporations and other "augmentative" aspects into SL so that people's in-world endeavours don't become the only reason to be in SL. But people here don't like that. And that's fine, to a great extent I agree, but then what do you do about the failure rate? This is why I object so much when people try to convert this into a personal statement that I'm making - that kind of statement can be answered with platitudes in a socially acceptable way (although not an especially valuable or useful way). But if SL is going to be kept alive in the long term without corporations, you have to fight the failure monster.



As soon as we're getting into "what I can do on my own server" we've gone beyond the service LL provide. That's like arguing that (some) shared hosting service isn't really slow, because you can keep your static pages on there, get a dedicated hosting account with someone else and run your scripts on that. Yes, but we're not judging what I can do. We're judging what that hosting company do.

Um, that's the whole point. The service you are describing is one that LL doesn't provide. If it was a desirable service, someone would provide it. If it is a desirable service and nobody provides it yet, then you have identified an opportunity.

I'm saying, it's not a desirable service.

You're saying, you don't want to take advantage of the opportunity, because you don't want to admit it really isn't a desirable service.

C'est la vie. Why should SL be any different?



Retention rates are very low in many virtual worlds / MMOs. Why? Hype. "Your world, your imagination." Hmm.. Well, this looks nothing like my imagination; I'm outta here. People come to SL thinking it is something it isn't, because it is sold as something it isn't for most people. For a while LL was touting its resident numbers (total registrations); now, not so much. Why? Because, for the most part, they are bogus, just as they are in many virtual worlds / MMOs. Many people will come to a new world because they are curious, and then leave when they find out that it isn't what they expected / wanted. To date, there's not a single VW/MMO "for the rest of us" that even remotely has a retention rate of 50% or better (with any significant numbers).



The problem is that LL is pursuing two mutually exclusive goals. Courting corporations is antithetical to courting people for entertainment, which is and has been at the heart of what SL is all about. As a result, their push to corporate-sanitize their entertainment product is meeting with lots of epic fail. There is a LOT less in SL of value to corporations than to consumers looking for entertainment. LL doesn't see it that way, as they have become quite the corporation themselves, forgetting what made SL (and themselves) in the first place.

As long as LL persists in doing everything they can to cut off their own air supply, I expect that "failure monster" to continue growing new heads until it eats them alive.

Yes, you're absolutely right. But I do miss the ease of Rev's syntax when writing LSL and would welcome some kind of interpreter that would allow me to use the language I can write in my sleep. I heard something about LL opening up scripting to other languages. I wonder how that's going and what it would take to get a particular language under consideration.

Why should it be the same? Why is it even valid to compare the two? The only connection between RL and SL is that both have "life" in the name - and SL didn't even have that when it was designed.



So nobody has solved the problem yet. But that doesn't mean it isn't a problem.



Except it isn't. SL has proven that the majority of the entertainment experiences are social, and thus will have majority failure rates because people cannot treat the majority of other people in that way.



SL has changed significantly since then, though.

Because, in both SL and RL, there are people. That is the one variable you can't change, no matter how strange and different you try to make SL from RL.

Just as with the topic of this thread, you make SL too easy, those who are advanced and want to do advanced VR/MMO things in it won't stick around (there are some who, right now as it is today, make this claim).



It is theorized that it isn't a solvable problem. Not only in scientific research sense, but also in a "common" sense: You can't please all of the people all of the time.



I've not seen any proof that social experiences are any more prevalent than non-social ones. Regardless, that's your own well-documented viewpoint, which I don't happen to share (and am not interested in rehashing it here, thanks).



Sure, it has changed, but that fact about it has not.

Do we compare Facebook to RL because there are people on Facebook? Do we compare WoW to RL because there are people on WoW?



And so we need to find a way to enable them to do that while not hamstringing the lower end.



It is perfectly possible for 100% of the people sitting in a cinema to be pleased for the duration of the movie; it is perfecly possible for 100% of the people playing WoW to enjoy it for the duration that they play; why should any other entertainment product be judged lower?



I don't think solitary experiences on SL are that interesting for most users.

For the things that relate to PEOPLE and how PEOPLE behave and respond, YES WE DO! To do otherwise would be, well, silly.



Well, again, when you figure out the magic formula that 100% of everyone, everywhere so far has been unable to come up with, let me know. Until then, I will presume from theory that such is simply unattainable from the nature of complexity theory and human interaction with said complexity.

In short, you cannot make people not be people, by being some imaginary thing you think they "ought to be". That way lies frustration and madness.



Sure, and it is possible for a normal human infant to count every grain of sand in the whole world in a fraction of a second, too. It's only a matter of figuring out the technique, building the tools, and teaching it all to the little primate.

While it may be possible in some fantasy theory you have, it is most certainly not probable in any reasonable sense.

What is "perfectly possible" anyway?



How would you know? Have you done a poll?

People I see playing Zyngo, 7Seas Fishing, or any number of other "solo" games out there seem to be as interested and engaged as their more "social" counterparts out dancing, RPing, or whatever.