Qie Niangao

This would explain that new car in the parking lot.

I'm not understanding quite what you're trying to do here. Getting the names of all the regions in a continent is not that difficult (see the Region_name_from_global_coordinates cap ) as long as you don't have to visit them for some other reason. Because you're also trying to plot paths, however, maybe you also need to visit them. Note that if you're hitting that from within a script, you'll need to be careful about pacing the calls to llHTTPRequest(), which are throttled to about one per second. (I notice on the wordpress site that you're having problems with "http request script does not work while we are flying above 50 meters" which doesn't sound right; instead it may be the effect of that throttle.) Anyway, if it's not clear from the other responses, the relevant land restrictions are set per parcel, not for a whole region (although there are some single-parcel regions, I suppose). The thing is, these settings change all the time, as does the layout of parcels within regions, so mapping these a priori is kind of a lost cause. Instead, the path finding through the maze would need to be done in real time... and, of course, it is very possible to encounter a destination surrounded by restricted land, for which you really "can't get there from here."

That's certainly good news. We still need some advice from LL about how Arbor can best help them to efficiently and effectively maintain sanity (and profitability) on the Mainland. To a layman, it would seem that a list of locations of abuse, submitted by a reliable source, would be easier to work down than a bunch of disjointed individual abuse reports from random submitters. I know that sounds a little Arbor-FICish, but there has to be some way that this group can both help the Lab and advance its mission of improving the Mainland experience for everybody -- very much including the Lab itself. The group still has an important role, even if we have to execute it by "AR parties" again, as inefficient as that always seemed. I sent an alt out on a brief tour of Atoll roadways; although it was supposed to be just for fun, he found plenty of abuse (including a nasty bit of LL trademark infringement bundled with a particularly egregious microparcel scam). I could submit these as individual ARs, but this seems the long way 'round the barn.

I'm not sure where you should post this to get the kind of information you really seek. I suppose scripters are "actual users" of a sort, but we're certainly not representative of the vast majority of script users. The idiosyncracies of SL script execution may also get in the way of generalizing much from what you observe here. An uncharacteristically large part of the effort in developing SL scripts is devoted to squeezing out lag, squeezing into memory, and working around bugs and limitations, with very little room for compromises that would improve modularity, reusability, nor even readability. It's a bit like programming for real time embedded systems -- and that's somewhat odd, given that SL scripts are almost all user-generated content. That said, there are many open source programs that might be useful to your research. Many of them are in the library or archives, however, so it may be redundant to list them. Nonetheless, in case it's of any use, some widely used, re-used, and adapted scripts: OpenCollar; Lockguard MLP; Lex & LearJeff's Sit Target Positioner; nPose Argent Stonecutter's "flight feather" Void Singer's Zen resizer ... and as I'm typing, I realize there are so many that any listing will be absurdly idiosnycratic, so... I think I give up.

This is one problem that should be gradually ameliorated by the new process of bringing abandoned land quickly to market. In the past (and there's still a huge stockpile of these), abandoned parcels were left with auto-return turned off intentionally, and without changes to the build permissions, leaving the land ripe for all kinds of trouble. Another problem caused by this is the exhaustion of all spare prims from Governor Linden's holdings in a sim, when somebody fills up an abandoned parcel with as much crap as they can. That's a big deal on sims with roads, railroads, and waterways. It would help if all the intended public pathways belonged to LDPW instead of the Governor; I don't really understand why they don't.

A Linden definitely needs to delve into the parcel sales history for this sim and figure out what the hell happened here. I don't think it's going to be a simple answer. Looking around the sim there are microparcel clusters with descriptions of being abandoned at different times in 2010. Some of these clusters were set for sale, and some not, but those that are for sale still show descriptions of abandonment prior to the auto-for-sale process. Looking at the details, I'm pretty sure that the people named in those descriptions as having abandoned the land weren't the people who did the cutting. Maybe those descriptions don't get cleared by the process? If not, then perhaps the ones for sale were all in one auctioned parcel, purchased by some drunken dimwit, diced-up, and the pieces abandoned into the new automated process. That's the kind of mess that makes Simeon's jira so important. On the plus side, I see that a microparcel in a sim neighboring mine that had been bought two days ago and got a spinning ad sign and price jacked-up to extortion level, which I AR'd, has had the sign removed and the sales price dropped to only an absurdly optimistic level (which happens to be slightly less than the fool paid for it). It happens that this same would-be adfarmer doof bought a few of the scraps in Skutter; those microparcels are not currently for sale, so I'm thinking some enforcement message is getting through.

Thanks, Raudf. I had no idea it worked like that; I always thought that if you downgraded to Basic, any land was immediately and automatically abandoned, and any group tier contribution withdrawn. I guess I thought wrong. I also mistakenly thought that any L$ purchases were cleared through the charging institution before the L$ balance was credited, but even if that were true, there's always a chance that credits could be backed-out later (fraud, insufficient funds up-chain, or something). Anyway, back to the main point: I'm not trying to promote leniency for leniency's sake -- rather, I'm casting about for whatever policy: least discourages new Premium members, and to a lesser extent: most encourages the once-paying member to eventually return (either as Premium or as an Estate landowner), and to a much lesser extent: discourages dropping Premium status in the first place. To me, the main thing about #1 and #2 is the unboundedness of the liability that a Premium member can incur if payment lapses. In the case of the OP, they're hit with the full year renewal because they were on the annual renewal plan, but I think it's even worse than that: as I understand it, the Premium fee* continues to be charged against the account, so that to restore the account (even if only to downgrade), one has to pay membership for the entire time the account was inaccessible because of non-payment -- meaning that the barrier to returning just keeps getting higher. That's a huge disincentive to ever come back, and the risk of it occurring is a strong reason never to go Premium in the first place. Instead, there should be a cap on continued charges. I agree that 30 days is as good an interval as any, so I think a reasonable approach is to limit charges to one month of Premium fee (about US$10) plus one additional month of tier ("additional" because tier is billed in arrears for the peak usage of the prior 30 days, so by the time there's a billing problem, tier is already a month in arrears). And that should be as far behind as a lapsed Premium can fall. After that, they revert to Basic automatically, their land is reclaimed by LL, and if they want full access to their (now Basic) account again, they have to pay that fixed balance due. This isn't that different from the current approach, really, but it removes the negative incentives by which the current policy discourages both new and returning Premium members. I'm not sure I completely understand your proposal, but I think I like the idea of freezing a lapsed account's ability to transfer funds to another account. __________ *I don't know, but I've always supposed that tier fees also continue to be charged until the land is repossessed. No idea how that works with group tier contributions.

Hey, Dres (or anybody else who knows)... okay, I admit I'm clueless: how does a Basic member run up a negative US$ balance? Is this some wackiness of Marketplace, maybe?

The "stick" may sound useful, but if fear of the stick is driving people to avoid the carrots, there's a problem. First, I must admit that I have absolutely no experience with other MMOs, so it may be common industry practice, but I really don't think it's working very well in SL. I don't know whether that's a problem with the fit of the policy to SL's land economy, a problem with LL enforcement practices and systems, the frequently unreliable charging mechanisms, or something altogether else. My fundamental criterion for a Premium billing policy is that there should be no way for a Premium member to find themselves in a worse situation than a Basic member. As long as there's a risk of that being the case, it's going to be difficult to sell Premium memberships. Sure, they can be required to pay up overdue Premium fees in order to get their land back. That's fair. And yes, there's a chance that they may have gotten unpaid use of that land for the duration of any grace period, but either they pay up the entire balance or they lose all the land. (Incidentally, I now think my "ETA" was a little brain-damaged. One can't use such a scam repeatedly to save any money because one either pays the whole balance due or loses the land.) Actually, I think I said that they should be able to dispose of the land during the grace period. On further reflection, that's probably wrong. Maybe that would be okay during those first seven days for fixing charge info, but after that, the land isn't theirs to sell until they pay the balance due. I can even see putting the account on hold during that interval, so as not to complicate the logic about setting land for sale. Anyway, once a former Premium member fails to pay long enough that LL sells off their land, there's little point in trying to coerce them into paying the overdue balance by holding the account itself hostage. I just can't imagine that has ever once worked, but even if it does on rare occasions, it presents an unreasonable risk to those thinking of becoming Premium. There are too many things that can go wrong in RL. Suppose that two weeks before a renewal data, a Premium member unexpectedly goes away to hospital, or is otherwise physically unable to connect to the Internet for several months. Returning, that member may be disappointed to learn that their land now belongs to somebody else, and in the cosmic scheme of things, having to pay the full balance due for all that unused time is probably less traumatic than whatever caused their extended absence, but I honestly don't see the point of posing that risk to potential Premium members. I absolutely never would have become a Premium member as Qie, had I appreciated this risk in advance. Instead, I'd have set up a landowning alt and made very sure he never had any other responsibilities. It's just too risky. And that's no way to attract more Premium members. (I could tell my billing horror story--probably most everybody has one who has been here as long as I have--but it's long and boring and I've told it before. Suffice it to say, the period between billing cycles is not relevant to a grace period. Such a period commences only when, suddenly and through no fault of LL, the member, nor the financial institution, somehow, a payment mechanism that hasn't changed for five years simply stops working. At that point, there had better be a lot of time free on one's calendar immediately to fix the problem. Anyone to whom this has not yet happened is living a charmed existence. I, too, was blithely oblivious to this risk, for five years.)

AdminMenu = TRUE

Because there's to be a parade, silly! :smileytongue: And a MUSIC EXTRAVAGANZA. Which, now that you mention it, might be worth listing for the benefit of those planning their (next) weekend. Cribbing from a poster: 12:30 - DJ Marx 1:30 - JoeUSA Shelbyville 2:30 - Christov Kohnke 3:30 - Yaro Pinion That's not to mention the thrill of being in the presence of multiple Bay City residents, simultaneously! (Or, almost equivalently: "if you have to ask....")

There's a recent, high-profile report of being confronted by this (hopefully obsolete) practice of wiping Premium accounts: "Linden Lab has informed him that in 30 days his entire account will be wiped... including a significant investment in inventory." This may well be a miscommunication (or two, given that it's hearsay), but the perception still exists that Premium membership is a very dangerous thing. I realize that WEB-2647 has been closed for a long time, but there seems to be confusion about SVC-6251, and I really think the practice itself needs to be changed to be clear and consistent. The current practice is very incompletely described in the cited Delinquency Policy, now that I actually read it. The part of the process described there is fine, but it's not where things go off the rails for Premium members. Assuming that the account termination + inventory wiping is not actually happening any more (which was a disastrously bad practice), there remain a couple of problems, as best I can figure out: After a Premium account goes into arrears, it appears that the entire auto-renewal amount must be paid to bring the account back to paid-up status. That's probably a mistake. If the member was on the annual plan but wants to only renew on a monthly basis to revive the account, that should be possible. At some point, the Lab repossesses any land held by an in-arrears account. That's fine, but at that point, the account should revert to Basic and any amount due should be dropped. Yeah, holding the account hostage is some incentive for paying the balance due, but I'm guessing that actually works rarely enough that it doesn't offset the disincentive it poses to ever becoming Premium in the first place. To be honest, I think the simplest and most coherent process would be for Premium accounts to immediately revert to Basic whenever payment lapses, commencing a 30-day (or whatever) grace period during which time they must either re-establish a Premium account (on any renewal plan) or dispose of any land holdings before LL reclaims it at the end of that grace period.* (Given the vagaries of charging, this probably should be preceded by the current preliminary 7-day interval just to rectify payment info such as expired credit cards.) I also want to add that I've gotten extraordinarily good response from Billing phone support. I wish I'd taken down the representative's name the last time I had to call, because he really deserves commendation for thorough and professional attention to the customer. This is in stark contrast to the screwed-up automated system that had to be worked-around, and a painful Live Chat experience that may result from those reps having very strict limitations on their own processes, which filter down to some quite unreasonable demands on customers. The point is: anybody with a billing problem should contact Billing directly, and never try to go through other support channels. _______ *ETA: This "grace period" would need to be restricted to no more than once in a 12 month interval or something like that. Otherwise a daring landowner could pay tier only ever other month.

Somebody has to say it: There is another possible explanation. If the land is outrageously overpriced, and it's situated such that being access-restricted poses a tremendous inconvenience, it may be the seller's attempt to bully a neighbor into buying. This was briefly a fairly common practice for adfarmers gone bust, trying one last desperate scam to unload their roadside microparcels before the Lab started cracking down on them for it. On the other hand, if it's land somebody might actually want for a price somebody might actually pay, then yeah: the seller just screwed up. Incidentally, if it's possible to TP to the parcel, only later to be ejected from it, then the culprit is a "security" script, not restricted access parcel permissions. That's only of possible interest because this seller seems likely to need detailed hand-holding through the process of minesweeping their parcel.

I haven't looked at the MLP source for a long time, but it's BSD-licensed, so even if there's no such message built-in, it would be possible to craft it. For this particular application, however, it may be preferable to convert to nPose and do away with the poseballs altogether.

Pussycat Catnap wrote in part: Why is there your so-called market? Because people buy things to socialize and play with each other. Yeah, but two things: First, for a very significant part of the population, the subject of socialization and play is the market itself: the ability to buy and sell t-shirts and land. Second, the reason there's any place to socialize and play is that it makes money for Linden Lab. They must monetize the virtual world somehow. Where can they best extract rents from people wanting to socialize and play? I think virtual land is an ingenious way to do that, for all sorts of reasons, not least of which is the fact that many people enjoy the land market as part of how they engage with other Second Life residents (see first point). It's possible that this could all be flipped around. "Land" -- 3D hosting space -- could be doled out on the basis of something other than direct payment, and the Lab could generate revenue some other way. Perhaps user connection fees and/or subscriptions. Or it could go with an "app store" approach and charge 233% commission on the Marketplace. Personally, I think there's very little chance that anything else would be nearly as successful as the current overwhelmingly land-based revenue model. Anyway, the only place where "ideology" necessarily comes into play is this part where Linden Lab exists as a for-profit company. They are free to create any internal economic system, or none at all, as long as they make money for their investors.

Prokofy Neva wrote in part: The Lindens are in business to sell sims and currency. And it's more than fine that some of us, are too. If we weren't, they wouldn't be. It's just that simple. The market sustains those who want to play collective farm -- and that is what is secondary, not the market. The market is primary, and that's how it should be to sustain the world. Indisputably, "Lindens are in business," providing whatever services will generate revenue at a margin over costs. In theory they could do this without any market among residents, nor any fees for virtual land, charging for access to the service only. That's a famiiar enough business model for similar environments. In fact, however, removing the market and market-based land economy would destroy much of the fundamental appeal of Second Life to many of the Lab's customers, current and future. Sure, there would be some customers willing to pay for access to a social virtual world with no market component. If we think for a moment about what sort of service would best succeed with such a model -- what would optimize for that experience -- we'd leave behind many things that seem important features of Second Life. Instead, we'd "invent" Facebook. Or maybe Facebook with a Blender plug-in. Of course, even Facebook has a market component. Certainly Facebook apps / games have internal "markets" sometimes loosely confederated with a shared currency (and sometimes not). The point is that those "markets" are key to the success of even dopey Zynga games. I don't claim to be economic philosopher enough to generalize this to a universal human attribute, that capitalism is wired-in to the species. But just within the little economy of the gaming and virtual worlds biz, success seems pretty linked to the existence of a market -- even when the developers actively discourage the emergence of such markets. Whether or not "land" is allocated by market might be a separable business decision. That is, it could be that people will pay for cauliflowers and not the land on which they grow, which land might be distributed by some other scheme (perhaps a collectivist notion of "greatest need" :smileytongue: ). I think that, too, would erode appeal, not only because people like to participate in a market (see above), so land should be a market too, but because the market is an incentive driving the quality of content, including land, available to be experienced by participants. (This all leaving aside the very practical consideration of whether Linden could change its business model without painting red the balance sheet; I've ranted enough about that elsewhere.)

Phil Deakins wrote: [...] As time goes by, it won't matter, because the population is ever turning over, and before too long, those things will be the norm for the population in general, who won't have known things to be any different. Even now, almost all that ever gets talked about in the Merchants' forum is the marketplace, which, through LL's unfair practises, is rapidly becoming the "normal" way of merchandising for SL - it's sad but true. Yeah, the real oldbies talk about GOM; means nothing to me, though, since I arrived post-Lindex. To be frank, I wouldn't mind LL showing enough interest in Land again to actually try to compete with their residents. What I see happening, rather, is a continued shift to Marketplace, which makes absolutely no sense to me. I can't see how there's enough margin on that miserable little commission to be worth lighting the website, let alone spin off profits to keep the Lab afloat. Meanwhile, it erodes revenue from the Land product, the only cash cow they've got. I really think it's Marketplace that's responsible for the renewed collapse of Mainland prices. I expect Estate sim sales to follow suit again, too. If in-world commerce is dead (abetted by the mess that is Viewer 2 Search), and the only purpose of Land is to hold the stuff people buy on Marketplace, well, that's a market, but it's a heck of a lot smaller market than that to which the Lab has grown accustomed. I don't understand why they'd be so intent on doing away with one revenue stream when they've yet to discover anything else beyond a trickle. One other observation: if the new abandon-to-sale process adds even more downward pressure on Mainland prices, then you're absolutely correct that it would cause more land to be abandoned. It would mean even less incentive to bother feeding it to the bots. Not that there's really any incentive any more anyway. (Nostalgia: I remember signs around Mainland, exhorting folks not to abandon land, posted by land dealers presumably on the premise that any land in circulation was land on which they might make a profit some day. I can't imagine anybody still thinking that they could profit by having more land on the market. And, of course, with the new policy, it doesn't matter: it goes into circulation almost immediately after being abandoned.)

Phil Deakins wrote: Did you ask for your abandoned land back? [...] Nah. I found other things to do with the tier. I was very glad to see the adfarm extortionist gone from the sim, even though I didn't own the immediately adjacent land anymore. But I was pretty mystified that she gave up literally overnight, once I abandoned the surrounding land. I mean, it had stayed as a standoff for well over a year; I'd never contacted her to try to get the land (it only encourages them), and before had only absorbed parcels as the adfarmers had been driven out by LL. So it's not as if she could have had any reasonable expectation that I'd pay her ransom if she held out long enough. I have to suspect that she had reason to think that LL would be less tolerant of her crap when the surrounding land was owned by the Governor than a private resident. Hugsy, that is an interesting case, and yeah: the timing makes it difficult to see as coincidental. About pricing: I really don't think this policy is going to increase the rate of abandonment. Rather the opposite, in fact; I'd certainly never have abandoned any of my land if I knew it would go on the market immediately. Indeed, I kind of rushed to do the abandoning before the policy took hold, knowing it would sit "fallow" for a while that way. As always, it's the bottom of the market that will suffer, if this process succeeds in pushing abandoned parcels to sale faster than the auction system did. Honestly, there is some Mainland that should have no value whatsoever. In fact, even if tier were free, too, some of it is too awful to own. But what I mean is that this really isn't going to have any effect on Nautilus City, Bay City, nor most of Zindra... nor Blake Sea coastline, etc. But yeah, for landlocked, run-of-the-mill Mainland, I can see it raising the "no value whatsoever" threshold.

I kinda doubt that the unusual events around that abandoned 16m² was really a function of this policy. Very rarely, I see some unexplained transfer of abandoned microparcels to somebody not otherwise present in the sim. It's very rare, and I've consciously decided not to probe into it too deeply, partially because of something stupid I did to myself in my "main" region: A while back, I'd gotten very tired of looking at the sole remaining 16 in a burnt-out adfarm that I surrounded and managed to consume, bit by bit, over years. This 16 was set for sale at some absurd thousands-of-L$ price, but belonged to a group that had no tier contributed to it. I submitted a ticket, hoping it would get force-abandoned and I could finish off that adfarm for once and for all. That isn't quite what happened. Harry moved that parcel to a piece of Linden property that he carved out of the SLRR right-of-way, interposed between my property and the SLRR, deeded to one of the owners of the defunct group. I was disappointed (to say the least), but thanked Harry for the assistance. I realized that I'd brought it on myself. Then, that freshly-minted microparcel actually sold, and now has an ad on it. In fact, the ad is technically in violation of the adfarm policy because it includes a rotating texture. I've decided not to press my luck by ARing it. I figure another "improvement" like that and I'll be abandoning the whole d*mned sim. Speaking of abandonment: There was another 16 that I surrounded on 3 sides, again for years, for sale for multiple-thousands-of-L$s (of course) by one of the last adfarmer hold-outs. (Yes, they still exist. You may recognize this one by her faux French name, which I can't say here.) About six months ago, discouraged by the above incident, I finally gave up on that one and abandoned a couple thousand sq.m. around it. Overnight, that 16 had been abandoned, too. Elsewhere on the sim I've still got another 16 surrounded, for sale for L$4000 by yet another sleazeball. If I could be sure it would kill-off that S.O.B., I'd abandon however much land it took.

Route 2 is one of my favorites. See this wiki article for some background.

One thing to watch is that event queues are cleared when changing to a new state. If the script is not using states, then this obviously isn't a concern.

I suspect the confusion here is between Landing Points, which are set on a per parcel basis on Mainland, and Telehub routing, which is used only on private Estate sims. A private estate manager can actually prevent direct teleporting into any individual parcel in the sim, forcing all TPs to a single hub location. That's not the situation in Bay City (nor, as far as I know, anywhere else on the mainland). In contrast, we can TP directly to individual parcels in Bay City today -- our own, other residents', and Governor Linden's. As Marie correctly notes, my concern is with the per-parcel landing points set up for Governor Linden parcels that make up infrastructure such as roads and canals. These wind around and through each Bay City sim; everyone's individual parcel borders them. The problem is that the entire road and canal network and any parks or other infrastructure in the sim share a single landing point. If one wants to TP to the sidewalk in front of a business, for example, it's impossible. One is very likely to instead end up at the opposite corner of the sim and have to navigate through the sim, following the red beacon, to the intended destination. (Try it yourself: open the map, zoom in, and try to TP to the sidewalk next to your parcel. See where you end up.) Ironically, I'm forming the habit of avoiding map teleports to obvious Linden infrastructure, and instead targetting private parcels near my destination, to avoid being trapped by the Governor's wacky landing points. So, if anything, the landing points are causing more such intrusions, at least by me.

Not just to be contrarian: I'd rather move animations into the agent and out of the viewer almost entirely, except for playback. The current round-about mechanism for informing other viewers about an avatar's changed animation makes synchronization a remote ideal, and poses hurdles for extending to a more "physical" avatar. (This is actually orthogonal to where the UI of an AO operates. It's not as if moving the AO to the viewer takes the sim out of the picture; it just pushes the simple AO processing out of the script VM, and saves a network traversal only for the current convoluted data path for animations and animation-states.) I guess I'm suggesting that in addition to making the viewer more extensible, the server-side agent representation could stand some beefing-up, too.

I mentioned it at the Bay City Alliance meeting yesterday, and will send out a notice to that group shortly, so I'll ask here, too: Does anyone object to having those landing points removed from unparcelled Linden land in Bay City, on non-InfoHub sims? I'm increasingly convinced that those landing points seriously interfere with navigating and exploring the affected sims, because they render Landmarks nearly useless and turn teleporting by Map or SLURL into a maze of twisty little passages. One hypothetical objection that was mentioned at the meeting was the possibility that some folks bought land near one of the landing points* in hopes of catching traffic from frustrated travelers. That seems a pretty dicey proposition in the first place (no reason to expect those landing points to remain fixed for long), and I can't imagine it really helps traffic anyway: One very quickly learns never to TP to a Linden-owned destination in Bay City -- the penalty is far too high -- so, if anything, I'd guess that those landing points are places that (experienced) traffic avoids. Note that I don't propose to remove landing points from specific parcels when they've been set up for individual destinations. And I also don't propose to change anything about TP routing at InfoHubs, which (one hopes) is specifically designed for handling the arrivals traffic on those regions. The Landing Points I want to go away are the ones shared by all the road and canal and everything else Linden-owned in the region, especially for regions that are not 100% Linden-owned. _________ *I guess that would be a bit analogous to an old idea of setting up shop near the (128, 128) sim center, to catch those who TP into a sim without specified coordinates, although I'm not sure that ever happens anymore.

Darrius Gothly wrote: [...]Wouldn't it make more sense to move script execution to servers that do only scripts .. and maintain a partition or task for each Avatar? As I understand it right now, when you teleport or move across Sim boundaries, your entire script collection gets copied and run on the new Sim's server. But if script execution was separated onto its own server farm, and you got a partition on the script servers when you logged in, no matter where you roamed in Second Life, your script burder would never need to be moved or copied from server to server.[...] I've seen that proposed before, and it would be interesting to see how it would turn out, but I don't think it's a panacea. It seems it would get some savings at TP time (not having to actually copy scripts and state into the TP'd-to sim) at the expense of a very large amount of communications after TP is complete. A tremendous amount of information flows "through the membrane" between the shared state of the region simulation and the script VM; all that would have to become network messaging, with implications for perceived lag. My hunch is that it would replace one big fat rubber band with a whole bunch of little ones, from which there could be no escape.