Homestead Price Increase
You may be aware that Linden Lab are raising the cost for Homestead sims effective 1st July. More details can be found here.
Note that this does not apply to any mainland rentals or our full prim sims Ash, Palm, Maple, Viburnum, Blackwood or Mulberry.
As with the previous increase in January WE WILL NOT BE CHANGING THE RENTAL RATES FOR ANY EXISTING RENTER. The rates for all existing renters will be protected indefinitely.
We will be changing the rates on our unrented parcels. Currently we charge L$2500 for a quarter of a homestead sim. Effective immediately we will be charging L$2750 for a quarter sim or L$5500 for a half sim. Any new renters or existing renters who move to new parcels will be subject to the new rates.
As always if you have any questions contact any of the Arboretum owners - Elanthius Flagstaff, Skye Whitcroft, Liznwiz Wei, Del Peccable
AutoBots
Bots don’t need to be fast they can be smart.
Here’s the thing. As far as I know all landbots currently work on one very basic premise, they search for land and when they see something that is blatantly too cheap they try to buy it as fast as they can. The quickest bot wins. Smartness is in fact not a useful quality because it often slows down decision making.
Human land flippers like to argue that bots will never compete at the same level as humans. I say this is false and I’ve set out to prove it with a new type of landbot I’m calling AutoBot. Instead of being ultra fast and trying to snatch land it actually analyses each parcel makes an intelligent decision about the value before buying.
I’ll be using some simple artificial intelligence techniques i.e. genetic algorithms to train the robot to buy land that is cheaper than it should be and to resell it for the best possible price.
Here’s how it works. I have a set of rules, each rule consists of a number of constraints. For example, here’s a single rule
recdev 0
size 512
multiplier 0.97
slope 1
It means something like this, if you see a 512sqm parcel that is .97 times the base of the market and the difference between the max and min height of the parcel is just 1 meter and it is exactly square then buy it.
Of course you can have an infinite variety of rules like this, I can adjust the size, the multiplier, the slope or remove them as I wish. Each bot will have several, maybe dozens of these rules and it will compare each parcel it finds against them. When it buys a parcel it marks it with the id of the rule it used to decide to buy and then tracks what it paid for it and what it eventually sold it for.
So after buying a few parcels and selling them at the best possible rate the bot ends up with a table like this:
Rule Profit
1 L$10
2 -L$500
3 L$2000
…
We can then use simple roulette wheel selection to pick a pair of “parent” rules. We perform crossover maybe, a little mutation and we end up with a new rule. We can throw that into the mix of rules available to the bot. Maybe eliminate the worst performing rule to keep the set the same size. If you do that often enough and your rule breeding program is well written you should end up with a slowly optimizing set of rules for land buying.
What I have right now is a set of hand crafted rules and a bot that follows them. It’s checking the constraints and buying the parcels, then it’s selling them for the best price it can find using my normal reverse auction technique.
The near future, maybe even this weekend, I will develop this to the next step and have the new rules being automatically created based on the profitability of the existing ones. Once a single bot is semi succesful I can start deploying more. As long as each bot is making a profit it’s safe to create as many of them as I like and have them all buying land and trying to flip it. Competing against each other in a true Darwinian fashion.
The best thing about the basic infrastructure I’ve developed is that it allows me to add other kinds of constraints in addition to the 4 I have so far. Some simple possibilities are:
* Number of other parcels in the sim
* Proximity to Gov Linden Land (i.e. is it protected)
* Percentage of land under water.
* Colour of terrain (this is actually pretty hard to assess)
Gay for Jack!
The long awaited bot blog has arrived and it’s a doozy. I don’t want to gush about Jack Linden too much. I’m not Gay for Jack just yet but this new policy does seem to be very close to satisfying the maximum number of people possible.
Boxes o’ Bots are now banned. Great. No matter what Phil and others say they were a pointless drain on resources and served no purpose other than to violate the search mechanism. To be fair, I’m not against search engine optimisation. Of course I’ve done work on my own parcel to help it appear as high as it does in search. The difference between the other usual methods, paying for picks, keyword stuffing etc etc is that they don’t negatively affect the grid and people trying to visit the sim.
The devil is in the details though. The policy seems wise and fair but the implementation is where it gets a little messy. I’m a big fan of the other new policies Jack has been responsible for. Ad farming and land cutting. With land cutting though implementation has been somewhat wanting. Theres time for Jack to bring it around but we still see land being cut all the time and it’s laughable to suggest that the mainland might be “healing” any time soon. Hopefully LL will do something simple like run through the parcel with the largest traffic and give any bot users the beat down. It seems pretty simple.
There are some grey areas to consider although they shouldn’t concern us too much. Obviously now bots are banned people will go back to camping chairs. But what happens when (not if) the chairs fill up with roaming camping bots? It will likely be impossible to find the owner of the bots and banning them is pretty pointless. More can always be made. If you turn a blind eye then whats to stop land owners putting their own bots on their own chairs? I suppose the only solution is to make the land owner responsible for keeping bots away. He chose to put the chairs there, he chose a model of chair that allows bots, he is the one in violation. This might seem counterintuitive to some and I can see some whining to come but it might be the best way forwards.
But then what if the campers are not bots but real avatars that are afk for hours or days at a time? Is it a violation to go afk for too long? The policy doesn’t say. It seems like this should also be a crime. In my opinion, again, the land owner is the one at fault. Who can blame the hapless noobs for sitting on a chair that gives them free money? But if the land owner hasn’t put in some mechanism to ensure that the campers are actually at their keyboards such as asking questions regularly then the land owner again should be warned, suspended, whatever. Of course I’m not talking about punishing people for going afk. I’m saying if you have a massive camping farm you are skirting the letter of the law and need to go to extra lengths to prove your legitimacy.
These are minor points though. LL will work through the issues and gaming and whatever else the same way they always do (badly, hahaha).
28 March 2009 by ElanthiusThe History of LandBots Part 1
Well, I always figured the day would come. LL are considering taking action to curb land bots. See the latest blog for more details. For the first few years I ran my landbots I always figured LL were going to ban them within the next few weeks and yet it never came. We’ll see where the blog leads us but this seems as good a time as ever to post…
The History of LandBots Part 1
It was September 2006 when I first decided to give land trading a try. I’d been in SL a few months and played around with this and that so I was already very familiar with getting around and land and everything else. I’m generally a pretty cautious guy so I started by flipping 16s. It was amazingly easy to make money. You’d just buy the cheapest 16s you could find. Stick them up for sale for L$4 or L$10 or some nominal amount higher than you paid and people would buy them within a day or two.
Of course the land-landscape was much different back then. The market was totally opaque to non-land traders. All estate and mainland sales were mixed together in the search and there was no way to separate them luckily the search was also one big page. You’d have to hit search and then scroll past tons of sales for L$1 and other bits that looked like they might be estates. You really only had the name and description of the land to go by, and maybe you had memorised the names of some sims. As a result no normal person had any idea what land was worth and land traders would regularly miss certain types of parcels.
Well after a few weeks of hunting through the crappy search I soon got fed up with that and wrote a little bot to search for me, filter out the estate stuff and present the data on an automatically refreshing website. I was apparently the first person who had ever considered automatically filtering land sales. I knew I was the only one doing it because some days at work I’d see an awesome deal and then 6 hours later I’d get home and buy it. That’s not to say there weren’t other land flippers. In fact there were dozens but they just weren’t that smart. They would scramble and fight over every 512 that came up such that a 512 that was 10% cheap would be gone in seconds but massive sims or very tiny parcels would just sit idling forever.
I ain’t gonna lie, we CLEANED UP for the next few months. Flipping massive amounts of land. I distinctly recall very early on buying an entire sim for US$1000 and flipping it later that day for US$2000. That’s probably still the biggest trade I ever did. Now don’t forget although we had automatic searching I was still looking at a website, waiting for something to come up, teleporting to it and buying it manually. Many of the things we bought were invisible to everyone in SL. Other people wouldn’t even see this stuff because it was totally hidden in the crappy search window.
Things stayed like this for a while, until late January in fact. During that time I had improved things a bit, I was receiving IMs about parcels that came up instead of using a website with slurls. I had invited a few of my friends and family to have access to the website so we were all chasing each other and collectively printing kizash. No-one had any idea what we were doing and nobody was even close to coming up with a similar solution.
In January of 2007 LL upgraded the search so you could filter out estate parcels. It was a catastrophe for me of course. For a few weeks I muddled along chasing other land flippers. We’d still win often because my system was more efficient but once or twice we’d see a great deal and miss it. The situation was totally unacceptable. That was when the first landbot was written. It was pretty primitive, just a single bot searching and teleporting around trying to buy land. I always jokingly said that LL had forced my hand. I couldn’t possibly put up with these button mashers beating me to MY parcels.
Unfortunately since the market was a little more transparent and my bots were very slow many land flippers could see what I was doing because something would come up in search, they’d teleport to it and I’d always beat them. Yet somehow I wasn’t even on the parcel. It wasn’t entirely clear at the time whether landbots would be illegal or not. I always represented that I was confident it was legal but you never did know. Really I felt like LL could have decided at any time to outlaw landbots and destroy my business. I felt like I was one or two weeks away from being banned for the whole of 2007.
Everyone knew what I was doing and it wasn’t even that clever. To me it was the obvious move. Why sit around hitting search, search, search, teleport, buy all day long when you can get computers to do that boring crap. Yet no-one thought of doing the same thing. Other land flippers knew how much money was in the land market yet they were all too clueless to create something similar. Of course there was a lot of wailing and gnashing of teeth during this time. Several land flippers were driven out of business by my new system. It was all very sad and everything to hear Sarah Nerd complaining that her livelihood was destroyed or Joy Iddinja complaining that she couldn’t afford to visit her sick mother or whatever. I mean sure, that’s sad, but welcome to capitalism ladies. They all seemed to think that land trading should be like some sort of collective commune where everyone shared the profits equitably and competition was somehow unfair or something. Some of these people were highly regarded by the general population and so they succeeded in drumming up a great deal of hatredduring this year. Bot threads on the forums were a vitriolic bloodbath for me during 2007. I always stood my ground and argued my case, probably to my detriment to be honest but it’s not in my nature to take that kind of shit lying down.
In January of 07 Landbot Hax gained notoriety as some kind of evil demonic landbot from hell. I’m not sure why everyone hated him so much, he had this trick of buying land and instantly setting it back for sale which caused a few people to accidentally pay the higher price. For us it was no big deal though. He beat us occasionally but he mostly focussed on very high priced parcels we didn’t care about anyway.
Somewhere around March of 2007 Zor Zeddmore started his operation which was a touch better than mine. I think it was about April of 2007 when Chikaa Masala, Instant Voom, Bot Hax and maybe a few others started up. This was really the beginning of the Great Bot Wars. I couldn’t believe I’d got away with it for so long so I wasn’t too upset when competitors started challenging me. I had been so heartless towards the other land flippers when I pushed them out of the market that it would have been quite hypocritical to complain about these newcomers. Up until this point I had hardly bothered improving my bots at all there had been no point in it. As I recall I was generally on top during 2007 but never really won 100% of the land I was going for. Besides we all had different parameters. Mine tended to be the most cautious. i.e. I’d only buy 90% under market value. There were other bot runners buying land that was marginally cheap so they’d get plenty of land to keep them busy that I was ignoring. For the next two years various bots came and went some better than others. I kept improving my system, sometimes I was on top, sometimes I cearly wasn’t. Overall we did pretty well keeping up with the competition and the title of “best” went back and forth amongst us all on a fairly regular basis. As far as I know we’re the only name still going from this time.
Lots of people tried stupid stuff to get back at bots during 2007. A popular move was to shoot them or ban them or whatever. All very yawn. There were two attacks that really stuck out in my mind though. Seargent Crossair in May wrote, or paid someone else to write a small bot that would rapidly change land prices between something that was cheap and some ridiculously expensive price. The plan was that bots would see the cheap price, try to buy it and instead pay the expensive price. Unluckily for me he rolled it out while I was asleep and succesfully tricked my bots into paying L$200,000 each for 2 16s. It was a pretty massive loss for me, well it would be for anyone of course. I think I sucked it up pretty well though, although I did write a number of snotty emails to Linden Lab for not giving me my money back. Seargent got a few other bot runners and a couple of real people too with this trick. Eventually he got banned I believe but I’m sure he kept his ill-gotten gains. At least I never saw my money again. LL eventually patched the issue but not before a number of people got ripped off.
The other attack was much more serious and I still suffer from the effects now. Weedy Herbst and perhaps a few other crappy scripters freely distributed “anti landbot” devices. Several land flippers placed them under the ground in tiny invisible prims on every parcel they bought. Many of these things are still around now transferring from land owner to land owner and never getting auto returned. Some are even quite close to my store in Steamboat. The goal of this script was to just teleport home any bots or bot runners whenever they came on the parcel. Unfortunately if a script tries to teleport you home while you are not on the scripts parcel then it throws up a dialog box in the viewer that says something like “Teleport failed”. The pernicious scripts will hammer my viewer with these boxes multiple times a second and the dialog boxes queue up. The result is I either need to shut the viewer down and log in somewhere else or quickly close all the boxes so I can teleport home, then close many more boxes that are still in the queue until I can finally continue about my business.
On the upside for me 2007 was the year when LL adjusted their TOS to make it explicitly clear that modifed clients and therefore bots were absolutely allowed. They practically codified their legitimacy in law. People refused to believe it and put their own little spin on the TOS but I felt like I could relax a little if not completely.
25 March 2009 by ElanthiusTwo new OS sims
We recently brought online two new sims named Fir and Hawthorn. They fill in the two conspicuous holes in our continent. With the addition of these two sims we have connected our two commercial sims to the rest of the continent. These sims are OpenSpaces which means they only allow 750 prims and 10 avatars at a time. We do not plan to rent these out so they will remain available for boating and exploring and will ge left almost entirely as ocean.
LL Reduced To Begging
I haven’t seen the text of the email but apparently Linden Lab have been sending emails to every estate owner who doesn’t own a Homestead. The email apparently reminds them of the wonders of Homestead Sims and kinda begs them to buy one.
Lately Linden Lab seems more and more prone to spamming, I’ve had my own little forays into targetted spamming but the resounding feedback was that it was unwanted to the point of making some people quite angry about it. Fair enough I guess. I learned my lesson I won’t try it again. It looks like Linden Lab aren’t going to be learning this anytime soon though since their spamming increases in frequency.
I suppose the reason for this particular effort is the complete drying up of the Homestead market. There’s no doubt that estate owners are nervous about Homestead sims. Personally I’m quite sure we would have continued expanding into Homesteads sims at a slow but steady pace. Instead for the last few months there has been no consideration whatsoever of buying one. For sure the ones we have are still profitable but who knows what crazy thing is going to happen in June if they raise prices again. The risks are too high and a wait and see strategy is prudent. My colleagues and I have even been talking about our exit strategy from Homesteads if it comes down to it and I’m sure many other managers are taking the same attitude and making the same plans.
This, of course, is the problem with dragging out the price increases like this. It has frozen the market. If they’d gone ahead with the $125 pricing immediately then sure, the suffering would have been greater but it also would have been briefer. Like tearing a plaster (band aid for my American readers (as if I have any readers at all!)) off quickly instead of slowly.
Of course it’s not just the Homestead market drying up. Tyche reports weekly on SLU about the number of sims coming online and going offline. There has been a slow and steady decrease in sim count every week this year. Not massive just a hundred or so every week. There is a lot of turmoil in the full prim estate market too. Every day there are lots of people desperate to sell their estate sims, even (barely) profitable ones filled with renters and at the same time plenty of people willing to buy them. At this point only a fool would buy a sim from Linden Lab when you can easily get a proper estate sim for $700 with 20-25 days worth of tier remaining. Turnaround times are quick as well, a transfer can be done in just 2 or 3 days now. Of course Linden Lab still get their pound of flesh in the form of $100 transfer fees but it can hardly be satisfying swapping sims around between all the different land barons instead of printing new ones.
Overall I think the problem is that most of the incautious, barely profitable estates run by estate managers who flunked mathematics 101 are shutting up shop and only the prudent guys are left. They’re not interested in wild and unrestrained expansion right now. And neither am I.
13 March 2009 by ElanthiusAdult Ghetto
In a flurry of publicity LL have announced in not one, not two but 5 forum threads and 1 massive blog post that they are going to be forcefully evicting all adult oriented businesses and herding them onto a new continent built specially for the purpose, presumably with a non-stick coating and plastic sheeting over all the couches.
Firstly though, I firmly believe that the decision is made, the plan decided. LL have shown on several very significant occasions that they are really only pretending to listen to user input on significant policy changes like this. Indeed, they’re a business they have to do what they can to maximise profit and the mob mentality of the forums is not a great place to get your business strategies from. Besides, they’ve already defined and described everything in the past tense in the knowledge base. Why put this in place if the whole thing wasn’t a done deal?
The implications of such a move are so mind boggling that it’s hard to get your head around. Certainly predicting the future is going to be an exercise in futility but I like to try anyway.
Land Market
Of course this is just a side issue to most people but to me it’s the most critical factor of all.
Short Term
Prices on the new continent will be absurdly high as speculators and businesses buy the land up hoping to flip it for a profit. The whole continent will be very well landscaped and so will have a high value like Nautilus or Bay City. Mainland prices will drop sharply in the first week or so since people will be abandoning their land and getting the hell out of dodge. Printing new sims will cause panic in the mainland market that will crush prices. Eventually we could see prices drop below L$1/sqm if there is significantly more land than people.
Medium Term
The collapse in mainland prices will start to negatively affect the estate market as mainland starts to look more attractive due to super cheap prices. Of course at the same time mainland prices will start to come back or at least stabilise. At this point it will be hard for mainland prices to rise but they can certainly stop going down. Prices on the new continent will begin to fall and equalize with the rest of the mainland.
Basically we’re going to be looking at a long period of extreme volatility plus two totally separate mainland markets where the prices won’t correlate well at all. There’s going to be a lot of gouging and speculating and all the other fun stuff that capitalism brings.
Adult Industry
It’s not just that all adult businesses will have to move, lose all their picks, adjust their classifieds, fix all their notecards and landmarks, rebuild their entire stores from scratch etc. It’s also that the entire porn ghetto and all adult estate sims will be unaccessible to anyone except Payment Info On File/Age verified accounts. Presumably people selling sex toys and clothes will be able to rely on XStreet to sell but clubs and hangouts will have to clean up their acts and stay on the old mainland or they will end up being almost entirely empty. What I suppose will happen is that only an elite few adult businesses will even be able to own land. Most will resort to selling through websites, either XStreet if it continues to allow adult goods or some other site that will be sure to spring up if they don’t. In my own case I suppose we’ll have to evict a few residents who engage in adult behaviour. I certainly can’t afford to lock any of my sims to payment-info-on-file residents only. I’ll resist it as long as possible of course but if LL are serious about enforcing this then I woneventually ‘t have a choice.
In The End
I suppose SL will survive. As has been stated elsewhere this is certainly the end of LLs dream to make SL the new internet. The internet has grown and thrived and expanded /because/ of the adult industry not despite it. I suppose this is all part of LLs plan to attract big business to SL. I’ve explained before why I think this plan is doomed. SL is still utterly worthless for all business uses and even cleaned up and Disney-fied it will still be just as useless but a whole lot less popular. Let us not weep for the adult industry though. Those guys are some of the most innovative people in the world. Porn will survive.
Posted in Adult, Land, SL Comments (1) 10:38
Small improvements to land
Jack asked here for some small improvements he could make with a weeks’ worth of coding. I posted my favourite but I thought I’d list the others here because then people would see the track back and come read my blog and maybe pay me money for stuff.
- Ability to move, rename. change estate on sims via the land.secondlife.com page
- Top Scripts on mainland sims, open to everyone who owns land there
- Ability for groups to hold more than 1,000,000sqm of tier (right now very bad things happen if you go over that number)
- Ability for group permissions to be linked to different parcels. So, OK this group has 3 parcels, people in this role can do this on parcel 1 but not on parcel 2 etc
- Ability to donate tier direct to a person instead of just to groups
- Objects deeded to groups are returned to group founder when auto returned
- Group transaction history. OMG group transaction history please. Currently if someone buys land from me I have no serious way of knowing what they paid or even if they bought the land at all
- Make parcel property lines easier to see, especially underwater
Really it’s a crying shame that Jack even needs to ask, there are so many things broken about SL Land that anyone should be able to see at least a couple of things to fix right away.
9 March 2009 by ElanthiusViburnum - our newest sim
NinjaLand is proud to welcome Viburnum our newest sim to join The Arboretum estate. Viburnum is very much a party sim with several large, popular clubs. Drop by if you’re looking for fun.
We’re very grateful to Drakon Cortes for selling the sim to us. We’ve relocated the sim to the south edge of Blackwood. Since Blackwood is full of shops and Viburnum is full of clubs there will hopefully be a very good relationship between the sims, trading customers and foot traffic.
Of course clubs and commercial operations bring performance difficulties and we’re not oblivious to that but these sims are far enough away from the residential sims that we’re confident there will be no problems for our existing renters. In order to get a better grip on performance issues in all our sims we are working on some new tools that may help our residents see the script usage for their parcels and the effect on the sim.
Univeral Network Abandoned
When I first started universal network I had this dream that I would get a 16 in as many sims as possible and then I’d open the group up so scripters could use the parcels for whatever they wanted. I imagined there were tons of uses people would find for a great big network of parcels like that. The collection of parcels grew but there were only ever a couple of people that used it and it didn’t seem like it was very popular at all. We did use it though for a few things like the w-hat name2key scanner which my bots heavily rely on even now that the new search techniques exist. Also Max case’s Sim neighbours project which now seems to be totally defunct.
After a while we changed the way we looked at the group. It started growing much more rapidly as Skye worked tirelessly to find more 16s but at the same time we started shutting other people out of the group. Eventually the ad farm policy made it impossible to trust anyone else with access to the group. We couldn’t risk someone putting up ads or prims that looked like ads since it would just get us AR’d.
In all this time we never really put the network to any serious use apart from a few pointless scanners. It always seemed to me like the thing would be extremely valuable if the right idea ever came along but I guess it never did. In the last few months we haven’t been adding any parcels to it at all. The number of parcel move requests seems to have grown considerably though. It was pretty obvious that the whole thing was a waste of money and has been for the whole two years we’ve kept it. On the other hand it was still pretty awesome. Like owning some sort of wonderfully landscaped mainland sim it’s awesomeness was fundamentally valuable even if there was no real use for it.
I pretty much knew Linden Lab would contact me about the group and talk about their new land cutting policy. Harmony did ask me what I planned to use the parcels for and I explained some of the uses we had for it. To be honest, I kinda BSed and represented that it was much more useful than it really is. In the end she didn’t leave me with the impression she was going to force me to do one thing or the other. Nonetheless I get a lot of ARs and a lot of people dislike various aspects of my business. I don’t want any more attention from Linden Lab than is absolutely necessary. Not because I’m doing anything particularly wrong it’s just better if LL don’t immediately associate me as that asshole with all the 16s whenever they get an AR for some bot related thing.
Anyway, with all that in mind I’ve decided to donate the whole thing to Arbor project. All 20,000sqm worth of 16sqm parcels. It’s perhaps a touch evil of me to just dump the whole mess on those guys. The trouble is, of course, that the whole thing is like a massive pile of toxic waste. If I dispose of it in the wrong way, i.e. by selling it for very low prices, then it will become a big problem for everyone. Who knows what crap will end up on these parcels and it’ll just make a lot of people unnecesarily angry. By donating it all to Arbor Project I avoid a lot of those problems. Those guys will run around handing the parcels out to whoever deserves them most. Bless their hearts. I’ve donated the tier to their group so there’s no need to worry about that. I’ll probably try to draw down the tier as I get the chance but I appreciate it might be a long time, maybe years, before I can take it all out.
All in all, it’s surely the right business decision for me as well as a great PR move. Plus it’s obviously a good thing for the grid as a whole. So I think everyone’s a winner.
Posted in Land Comments (0) 0:48