Qie Niangao

Like the poor, the griefers we'll always have with us. If we push them out of the Linden sandboxes, they'll just hang out somewhere even more annoying. Meanwhile, the sandboxes make a handy place to use ARs to thin the herd. In fact, it may be the griefers that keep the Linden sandboxes open at all. I mean, it's not as if there's big money in it for LL to give people an easy way to avoid having to own or rent land. If, on the other hand, they closed the sandboxes and unleashed all those resident griefers on the rest of the grid, land ownership would be a lot less appealing. Maybe, then, we should thank sandbox griefers for their role in maintaining the natural order of the universe.

Yes, we absolutely need a 3.x, but it's a huge mistake to start from this assumption: ... most people love 1.x.. Most people have never seen 1.x. Nor 2.x. Nor even so much as a Second Life-related press release. The next viewer needs to start completely from scratch, except for the server messaging. There's simply nothing worth saving in either of the preceding viewer generations, and huge built-in handicaps. It's simply embarrassing that viewer developers have to push back on true customization, viewer-side scripting, exchangable viewer-side assets such as plugins... all because neither pre-existing architecture can support even the tiniest bit of expandability. Unfortunately, starting on a 3.x architecture is premature unless LL is in a position to state how they're planning to get to mobile devices. It's too soon to be doing any serious re-architecting if they're still looking at Server-Side Rendering vs HTML5 (really a business decision about how much compute power their particular target market will be carrying around, once the separate GPU is a thing of the past).

As I said, I'm no fan of V2 and the TPVs based on it. I think, however, that their suckiness is no more a contributing factor to new user attrition than was the suckiness of V1 and the TPVs based on it. Indeed, I am pretty sure that, if it were possible to measure such things objectively and comprehensively, the two viewer version suck as near to equally as makes no difference. It's just that many of us "cut our teeth" on V1 and simply accepted its misfeatures because we didn't know any better, whereas similarly egregious errors in V2 are obvious to us. I think the exact opposite would happen for folks who started with V2. (I'll refrain from enumerating all the ways I think both viewers suck, uniquely or -- even worse and more pervasively -- in common. Suffice it to say that V2 consumed far too much of LL's working capital and calendar, all for insignificant changes. I'm sure the 80/20 imbeciles would say that the problem was the users not accepting the more drastic changes, whereas the real problem was that they didn't understand enough about the application to change anything correctly, and only through great effort have Linden, Snowstorm, and TPV developers salvaged a usable viewer in which some of their trivial tweaks remain.)

A currency backed by the full faith and credit of not getting hacked. Good luck with that.

There's the viewer URI secondlife:///app/teleport/Region/X/Y/Z although that requires a mouseclick to activate and only works on modern viewers.

I'm no particular fan of Viewer 2, but no viewers are "there yet." Phoenix's crash rate is only slightly lower than V2's. It's really only an illusion of familiarity than any of the available piles of crap are fit for purpose. So it seems to me that if she's okay with V2, I'd let her stay on V2. When LL finally gets sick of supporting parallel interfaces for every central service and pulls the plug on these older viewers (Henri's herculean if somewhat obsessive efforts to backport notwithstanding), at least their long-time users will have gotten good use out of them, but for a new user... there'd need to be a really compelling feature they absolutely needed for some specific reason before I'd suggest spending the time to learn the V1 UI at this late date.

I'd also like to know what this is about. It's not a new phenomenon -- although I haven't seen quite this many in a clump before -- but it's always been mysterious to me. I mean, it's clearly not to give a traffic boost to a Linden water sim, and I've watched the same avatars parked for days at the bottom of some sims, so if they're traffic bots they'd have to be awfully primitive ones. Or if traffic-bots-for-hire, awfully unsuccessful ones. My hunch would be that they're failed alts of somebody who's experimenting with some kind of functional bot software, in which the bots are constantly plowing through data only exposed to logged-in viewers. Across the street and on one of the blogs there's discussion of an attempt to scour membership of Groups, with the cover story of building a social graph of group interconnectedness. On closer inspection, that particular one seems to be a marketing scam for a few old fading clubs, but if one were to do it for real, it might take a few bots to keep it reasonably current.

It's a tricky balancing act between playing to their strengths on the one hand and maintaining minimal tablestakes where they can't hope to compete favorably. An example of the latter category is lag / frame-rate. No matter what the developers do, they can never approximate the frame rates of industry standard-issue games with the insurmountable advantage of developer-optimized content. No way, not even close, ever. Their only hope is to keep the user-developed content streaming at a tolerable rate -- where "tolerable" is a moving target, rising with expectations borne of other games. That forces some finer-grained trade-offs, too. Mesh can be only so good, constrained not only by development calendar but by how badly Mesh creators should be able to shoot themselves in the foot -- along with every observer within draw distance of their creations. Anyway, I agree that their strengths, really, are land and lots of it. I also agree with Alazarin that about the best move they could make in terms of drawing people into increasing landownership would be doubling the premium bonus tier to 1024, to give a plausible "graduation" path from those Linden Homes (which must obviously remain 512 and unimproved or this will never work). Far too many new landowners are getting "stuck" at the Linden Home stage, never making it out to the Mainland nor Estates. Also, before any new user buys any land, they have to stay in-world for more than an hour. That means they have to find stuff that interests them (the Destination Guide was a good move in that direction, but I'm very sure that newbies still don't have access to the kind of one-on-one hand-holding that's needed to get them over the initial hurdles and into any Destination Guide destination. Until last week, I didn't really appreciate how damaging to retention the maturity rating stuff is, because I'd always focused on the hurdle of getting to Adult. What's worse is the hurdle of getting from General to Moderate, if the newbie hasn't been clued-in to that. I'm not sure why this newbie was stuck with General-only settings (is that the default? I certainly hope not!), but hers were set General only, and she was stuck, unable to navigate to any of the green dot-populated places she saw on the Map. She couldn't even walk there without bumping into a mysterious maturity fence, and although the "fence" explained itself, there was no way she could be prepared to understand that explanation, nor what to do about it. I finally managed to explain all that to her, just before she had to log for RL. As helpful as I tried to be and as grateful as she was, I very much doubt she'll ever be back.

... Put those into a prim and deed the prim to your group then people can change channels. but only deed it to group if the land is deeded to group! The prim should be owned by the landowner, whether that's a group or an individual, in order to work reliably. Long term, there are major advantages to having land deeded to group, and there are also advantages to using a scripted prim to control parcel audio and media streams (notably, that makes it easy to control who gets to change the stream without having to mix that up with who's in the group in what role with what permissions). (One other warning about deeding stuff to group: always make sure that the object and everything it contains is set to full perm for the next owner; not doing so is an easy way to brick your own creations.)

In my case, it's purely subjective and wholly unquantified, but lately I'm increasingly feeling that these forums are worth posting to again, and whether there really are more posts or not, there certainly seem to be more posts fit to read, anyway. I admit it: For me, one reason I'm encouraged about the future here is the conspicuous absence of certain posters. They made it feel childish and slightly disgusting to read -- let alone post to -- the official forums, for a reign of terror during which they fed each other and filled the forum with a recursively self-referential swirl of indulgent crap. There was no ignoring them until they went away -- their posts had become the forums. Whatever moderation it took to sideline them, I'm glad they're gone. There is, however, a whiff of that repulsive indulgence in this thread. Not that it needs to be expunged or anything, but one hopes it remains as quaintly irrelevant and out-of-place as it is now.

That "surrounding ocean view" is, however, only a view. Although I'm a Mainland creature myself, I've built my share of private islands, and there's not a heck of a lot one can do about being surrounded by the impenetrable void. That's the huge advantage that USS got, hooking into Blake Sea, although they had a lot of OpenSpace sims among them even before that move. Best of both worlds, though: full Estate control of their sims, and the entire water surround of two continents to sail in. Otherwise, by far the most common (and easiest) thing to do with a private island is to make it the standard-issue tropical beach, leaving just enough shallow margin to paddle a canoe between the shore and some hokey off-sim outcroppings.

But as I understand it, a script reset isn't really what you want to do anyway. Rather, I think you want to reprompt with the Main dialog whenever "Back" is selected from either submenu. So, I'd add the following to the listen() handler, before the test for menu==1 : else if ("Back" == message) { menu = 0; //show main menu, called Main llDialog(id, "Click on button to change number or distance", Main, 10); } You might also want to make this bit into a function, because it also appears almost identically in the touch_start() handler.

The red line marks the current time (whether there's a meeting going on or not). The Community Tools group meeting was cancelled today... you can see it on the linked user group wiki page. That cancellation may have something to do with Amanda's departure from the Lab.

I don't use voice myself, and have SL sound turned off completely about 90% of the time. But that's not why I'm responding. What interests me in this thread is a couple of posters who have circles of friends who co-exist as a mixture of voice and non-voice participants in conversations. I have literally never seen this work effectively. It definitely rules out participation by folks who, like me, usually can't have sound turned on. But aside from that, it just seems to happen that the voice people end up talking to each other and ignoring text, and vice versa. From what I've seen, the conversation simply bifurcates into parallel and non-overlapping voice and text threads within just a few minutes. I'm intrigued that others aren't experiencing this, and wonder what their groups are doing differently to keep the conversation connected across the different media.

Yes, educated landowners are much less likely to use whitelist banlines. Based on my experience talking to neighbors, I estimate that at least 75% of all whitelist banlines are there by mistake or due to a complete misunderstanding of what they do. The fact that the landowner doesn't see their own banlines contributes to those mistakes; they can't see that their banlines only go up 50m Above Ground Level -- which is very often irrelevant to the part of the parcel's space they're trying to keep private. Perhaps another 20% of these things are just the landowner trying to bring grief to a neighbor over some dispute (oblivious to the fact that now everybody else in the sim thinks that landowner is a jerk). That leaves about 1 in 20 banlines doing what they were intended to do. But, lemme tell ya, for those 1 in 20 users, you'll pry their banlines out of their cold dead hands. (Cue Charleton Heston.)

It sure would be nice, however, to know in advance whether a viewer is of V2 generation or not, in order to know whether it's too obsolete to handle the current viewer URI namespace. At the moment, I'm particularly wanting to use secondlife:///app/teleport/ (which is pretty slick, on capable viewers), in lieu of llTeleportAgent().

I quite like the idea of an all-banlines continent. It also seems only fair that landowners should see their own banlines. I avoid the Marketplace as much as possible, but I'd have thought that there'd be HUDs that could at least hint about the sociopathology of parcel settings ahead. I'm envisioning something that would just keep scanning all the 4x4m bits of land in a swath however wide, toward the direction of motion, calling llGetParcelFlags() for PARCEL_FLAG_USE_ACCESS_GROUP, _ACCESS_LIST, _LAND_PASS_LIST, and (not) _ALLOW_SCRIPTS, and whichever side had the nasty-bits set would get a red arrow pointing toward the center of the screen. I haven't really tried this myself. My hunch is that this function would be fast enough and cheap enough to work for all but high-speed fliers. (That's in contrast to llGetParcelPrimCount() and llGetParcelMaxPrims(), which along with sim border detection would be needed to avoid the dreaded SVC-22.)

Luc Starsider wrote: lol. Imagine if you drive through a region which has been split into small, 512m parcels, and every parcel used its own windlight settings. Yeah, but we generally don't drive through parcels, at least not intentionally. It would have a greater impact for folks who fly cross-country, higher than the 50m Above Ground Level whitelist ban ceiling, but I think they're generally trainable to use the (already conveniently provided) viewer-specific settings to ignore whatever's set on the parcels they skim over -- for those few who would actually care. Of course the same viewer-local setting would work for those with very delicate sensibilities, to override the much less frequent transitions that occur when walking from parcel to parcel. I can imagine that the developers would hate it if lots of people used that local override, defeating the whole purpose of the feature. Somehow I don't think that would really happen, though.

Just a couple things to add here. First, Ubuntu came very close to switching from Firefox to Chrome as the default browser in their latest release. That's partially because of perceived future advantages of Chrome's tight coupling to Linux, in Chrome OS, so maybe not much relevant to Windows users. But that's the second point: At the moment, I wouldn't stake anything on Microsoft. They still rule the desktop, but their future in servers is looking very grim. One tiny example of how they've allowed themselves to become vulnerable: Los Angeles is in the process of switching their calendaring system from some Novell-proprietary something-or-other to Google on the cloud. Why, you may ask, is this bad for Microsoft? As enterprises and government migrate to the cloud, there's ever less reason for Exchange. That's less reason for Exchange Server. And less reason for IIS. And before long, that integration among very expensively licensed server components suddenly looks like a house of cards. Microsoft will be lucky if it doesn't pull the desktop down with it. (Google Docs spreadsheets now have pivot tables. I haven't used Excel since.)

Now that I actually look at the code, my weathervane script looks a lot like Kaluura's suggestion, with the relevant line being: llRotLookAt(llEuler2Rot(<0,0,llAtan2(wind.y, wind.x)>), 1.0, 1.0); It's not obvious to me why llRotBetween() is unstable, so Void's explanation sounds as good as any I'd dream up.

Is there some reason we're avoiding llRotLookAt() ?

Yeah, the blog post probably shouldn't have said "landowners" because the vast majority of landowners actually can't use Estate environment settings. And one doesn't need to be a landowner to merely see the effect of the settings on Estate sims. I do not agree that permitting per-parcel environment settings would interfere with immersion on the Mainland. I think I understand the argument, but it's not as if denying this feature to Mainland owners is going to magically make Mainland an immersive experience. What's more, travellers who want consistent environment settings across parcels (and sims) can readily set them local to the viewer (to the Mainland default, if desired), which overrides any land-specific settings later encountered. That said, I wouldn't want to add significant lag to parcel-crossing. It would not be at all important that the viewer immediately reflect a parcel's environment settings, so this might mitigate any lag effects (server- and viewer-side) if environment updates from parcels were "lazy" -- possibly omitted entirely for parcels passed through quickly.

Son of a... I can't find any of them still asking for tips. I haven't seen any of them on the SLRR that don't have a kind of rail car undercarriage (that's very recent). I've now seen both of those again. The second one doesn't really bother me much. It will be better when they conform to their settings, but as I said: we can't really blame somebody for not scrambling to produce a soon-to-be-obsolete prim or sculpty build, with Mesh just around the corner. The first, however, is genuinely troubling.

That is a good point. The flip side is that being allowed to transfer without being allowed to sell introduces a kind of "black market" where payment is made separately from the transfer of the goods. That and the fact that some things simply shouldn't be distributed, and one would want to be able to associate RL identity with the account doing the distributing. Arguably, it may be that uploading is the place to stop content that shouldn't be distributed. The direction seems to be more assets (such as Mesh, textures, etc.) that get explicitly uploaded (in contrast to prims and avatar shapes). I'm not thrilled about restricting all uploads to PIOF-only, but... now I'm not sure what's best, either.

If you own the whole sim, though, you can control sim-side lag on the Mainland, too. The reason it's often so bad on Mainland sims with multiple owners is that Linden Lab is the Estate manager for the Mainland, and Lindens are inordinately enamoured of breedable animals, so they'll rarely do anything about even the laggiest menagerie. Obviously, if one owned the whole sim, one could boot any critter-harboring tenants. Viewer-side lag on Mainland is also sometimes a problem because there are so many different landowners building from different texture inventories. That can be a problem on rental Estates, too, but they're often just the standard-issue tropical islands surrounded by void, so that reduces viewer lag: there are no neighboring sims, so there's nothing to render. (Nothing to see, either, but that may not bother some folks, especially if they're on lower-end machines.)