There was an article today in the Second Life Herald that had an interesting graph of some data from the recent key metrics release. It showed that about 24% of sessions are ended "abnormally". I'm not sure exactly what this includes...obviously crashes or freezes. I find this number surprisingly high, as I would estimate 5% of my sessions, at most, end in freezes or crashes. Since my internet problems have started, however, I'm sure I've been contributing to this number from all the times my connection has gone out on me while I was logged in. This makes me wonder how many of these abnormally ended sessions are actually not the fault of Second Life, but something on the user's end.
How often do you freeze or crash? Are there other ways your sessions ends "abnormally"?
p.s.: I am writing this from my new computer which is now running Windows XP. I was finally successful in getting it installed, and I may write about it. Right now, though, it's something I just want to forget.
Saturday, 8 September 2007
Wednesday, 5 September 2007
The Fate of Second Life
Will Second Life still be here next year? In five years? To answer this question, let's first take a look at Linden Labs financial situation.
It's important to realize that revenue is not profit. Many people see that Second Life is generating millions of dollars a month, and then don't understand why LL doesn't upgrade their network infrastructure and servers to be able to support the growing population. The fact is that LL doesn't have millions of dollars to spend, because the cost of maintaining the grid at its current state is extremely high. An article published in the Second Life Herald this summer looked at the sources of income for LL and their costs, and found two important facts. First, that while LL is making a profit, it is a very small one; and second, that about 1/3 of their revenue comes from selling servers (private islands), and this doesn't include the monthly fees of those private islands, but just the setup fees. This means that unless LL significantly changes either their sources of revenue or their costs, Second Life is only profitable as long as it continues to grow at its current rate. This is basically what caused the dot-com bubble and burst. Internet companies assumed that the growth of the internet seen in the mid-late '90s would continue at that rate indefinitely. There are only so many people in the world, so obviously the growth is going to have to slow down at some point. For the dot-com companies, this happened around 2000. When will it happen for Second Life?
It was around the time I started SL that so many companies began experimenting with marketing in Second Life. If Second Life became an effective marketing tool, LL might be able to get support from companies to boost their profits. LL announced in their blog today the launch of a website focused on assisting companies and other organizations in creating a presence in Second Life. However, I began to see articles popping up a few weeks ago (I get email from Google whenever news stories with the term "Second Life" appear) reporting that many companies are now abandoning their Second Life projects. I think at this point it's too early to tell if and how companies will use Second Life.
There is an article I read earlier this summer I found intersting. Unfortunately, I haven't been able to find it again. The author speculated that LL might not care about Second Life continuing for years to come. Why? He saw the trend of LL releasing new features one after the other while the quality of the grid degraded, and recognized a strategy that has been used by companies before. The idea is basically that a company spends all their money creating features (which they patent) and don't care about the actual product failing. Other companies will create their own versions of the product, and now the original company will be able to collect royalties on their patents. So in the case of Second Life, it might fail but if anyone else creates a virtual world with, for example, integrated voice chat, LL will collect royalties. The caveat to this theory is that LL doesn't appear to be patenting like crazy. I personally don't think this is what's going on, but you never know.
From what I've seen, it doesn't look good for Linden Labs, but I don't think this means the end of Second Life. Even if Linden Labs loses all its money, declares bankruptcy, and the grid is shut down, there is still the open source viewer and OpenSim. People will start creating their own grids using OpenSim. It's even likely that the employees of Linden Labs would start contributing to the open source projects (I can only assume they are as obsessed with playing SL as all the normal residents are). But these new grids won't all be linked together in a monolithic "main grid" like the current SL. Instead, there will be numerous grids for different purposes: A C:SI grid, a Gorean grid, an underwater basket weaving grid, etc. I think this is a good thing. Smaller grids are cheaper and easier to maintain. Also, because both the client and server are open source, each grid can have specialized viewers and server code for whatever service they are providing. In fact, I think that once the OpenSim project becomes more mature, what I've just described will happen regardless of whether the main grid goes down or not. I think that people will want to specialize the software to fit their needs, as has happened with almost every open source project out there.
But it's expensive to buy server machines and rent enough bandwidth to maintain your own grid. It seems like these new grids would have to charge lots of money to stay running. If you look at Second Life now, though, there are tons of private sims that people have provided free of charge, and those sims are pretty expensive.
Still there are some other problems that will have to be addressed. For example, what happens to our accounts? Will we have separate accounts for each grid we want to log on to? Perhaps someone will maintain an account server that the grids can talk to, so that we can maintain one account across all the grids. What will happen with money? Will we revert back to a pure barter system (I'll give you this exploding egg for that dancing cow)?
Regardless of these problems, I think the future of Second Life is bright, even if LL goes out of business.
It's important to realize that revenue is not profit. Many people see that Second Life is generating millions of dollars a month, and then don't understand why LL doesn't upgrade their network infrastructure and servers to be able to support the growing population. The fact is that LL doesn't have millions of dollars to spend, because the cost of maintaining the grid at its current state is extremely high. An article published in the Second Life Herald this summer looked at the sources of income for LL and their costs, and found two important facts. First, that while LL is making a profit, it is a very small one; and second, that about 1/3 of their revenue comes from selling servers (private islands), and this doesn't include the monthly fees of those private islands, but just the setup fees. This means that unless LL significantly changes either their sources of revenue or their costs, Second Life is only profitable as long as it continues to grow at its current rate. This is basically what caused the dot-com bubble and burst. Internet companies assumed that the growth of the internet seen in the mid-late '90s would continue at that rate indefinitely. There are only so many people in the world, so obviously the growth is going to have to slow down at some point. For the dot-com companies, this happened around 2000. When will it happen for Second Life?
It was around the time I started SL that so many companies began experimenting with marketing in Second Life. If Second Life became an effective marketing tool, LL might be able to get support from companies to boost their profits. LL announced in their blog today the launch of a website focused on assisting companies and other organizations in creating a presence in Second Life. However, I began to see articles popping up a few weeks ago (I get email from Google whenever news stories with the term "Second Life" appear) reporting that many companies are now abandoning their Second Life projects. I think at this point it's too early to tell if and how companies will use Second Life.
There is an article I read earlier this summer I found intersting. Unfortunately, I haven't been able to find it again. The author speculated that LL might not care about Second Life continuing for years to come. Why? He saw the trend of LL releasing new features one after the other while the quality of the grid degraded, and recognized a strategy that has been used by companies before. The idea is basically that a company spends all their money creating features (which they patent) and don't care about the actual product failing. Other companies will create their own versions of the product, and now the original company will be able to collect royalties on their patents. So in the case of Second Life, it might fail but if anyone else creates a virtual world with, for example, integrated voice chat, LL will collect royalties. The caveat to this theory is that LL doesn't appear to be patenting like crazy. I personally don't think this is what's going on, but you never know.
From what I've seen, it doesn't look good for Linden Labs, but I don't think this means the end of Second Life. Even if Linden Labs loses all its money, declares bankruptcy, and the grid is shut down, there is still the open source viewer and OpenSim. People will start creating their own grids using OpenSim. It's even likely that the employees of Linden Labs would start contributing to the open source projects (I can only assume they are as obsessed with playing SL as all the normal residents are). But these new grids won't all be linked together in a monolithic "main grid" like the current SL. Instead, there will be numerous grids for different purposes: A C:SI grid, a Gorean grid, an underwater basket weaving grid, etc. I think this is a good thing. Smaller grids are cheaper and easier to maintain. Also, because both the client and server are open source, each grid can have specialized viewers and server code for whatever service they are providing. In fact, I think that once the OpenSim project becomes more mature, what I've just described will happen regardless of whether the main grid goes down or not. I think that people will want to specialize the software to fit their needs, as has happened with almost every open source project out there.
But it's expensive to buy server machines and rent enough bandwidth to maintain your own grid. It seems like these new grids would have to charge lots of money to stay running. If you look at Second Life now, though, there are tons of private sims that people have provided free of charge, and those sims are pretty expensive.
Still there are some other problems that will have to be addressed. For example, what happens to our accounts? Will we have separate accounts for each grid we want to log on to? Perhaps someone will maintain an account server that the grids can talk to, so that we can maintain one account across all the grids. What will happen with money? Will we revert back to a pure barter system (I'll give you this exploding egg for that dancing cow)?
Regardless of these problems, I think the future of Second Life is bright, even if LL goes out of business.
Tuesday, 4 September 2007
Monkey Wrench
I went to the "Technology Solutions Center" on campus to buy Windows XP today (see my previous post), but it turns out I have to order on their website and wait for shipping; they don't keep it in stock. Being impatient, I started looking for alternative methods of getting an install CD of Windows XP with SP2 (other than pirating it). I came across this, which I am going to try. Basically, it involves copying the contents of the install CD into a folder, downloading the SP2 update and applying it to the folder, then burning a new CD. First, I am going to try doing it on my new computer in Ubuntu, using Wine and other tools. I would love to be able to say I took an old Windows XP CD and created a (legal?) up-to-date CD using Linux. If I fail (which is likely), the backup plan is to use my laptop and create the CD in Windows. That will be more annoying because I use Ubuntu on my laptop and so I will have to backup files, install Windows, create the CD, then reinstall Ubuntu, i.e., to get Windows on my desktop, which has Ubuntu, I have to install Windows on my laptop, which I want Ubuntu on, so that I can install Windows on my desktop, and then install Ubuntu back on my laptop.
Monday, 3 September 2007
Worse than Dial-up?
It's story time again. This one is about the horrible internet connection I've been living with all summer.
For the past year, we had Verizon DSL, which was fine, although rather slow (downloads were about 90 KB/s on average). The problems started towards the beginning of the summer during a lightning storm. I was at my computer, on SL, and there was a popping noise accompied by my computer restarting. I don't remember there being a lightning strike or anything at that moment, but my only guess is that it was a power surge. When my computer finished booting (thank god it booted, the popping noise was not unlike that of a capacitor taking on too much charge), I noticed the internet was not working. I went down to the basement and saw that all the lights were off on the DSL modem, except for the power light which was flashing, and all the router's lights were off. I unplugged them and plugged them back in, tried other outlets around the house, still the same thing. I called Verizon, they sent a new modem, and I bought a new router.
With the new equipment, we could get back online, but the internet was unreliable. We kept losing our connection (dsl light on the modem would flash green, which means it isn't getting a signal from the ISP). It would come on and off, sometimes just for a few minutes, other times for hours at a time. Also, when it was on it was about half the speed as it used to be. I made multiple frustrating calls to Verizon tech support. Finally I convinced one of them that the problem couldn't be fixed by trivial methods such as turning off the modem and turning it back on, but that there was actually a more serious problem. They put in an order for a technician to come out and find the problem.
The technician came out and said that the line going out of our house needed replacing. You'd think he would then go ahead and replace it, but that would be wrong. Instead, he set up another date for someone to come out and fix it. Are these people really so specialized that one person is trained to diagnose a problem, but then someone else needs to actually fix it? So the line was replaced, and I hoped everything would finally work. Wrong again. The speed seems better, although still fluctuated, and the internet still goes on and off (notice I'm switched to present tense, as this is still the current situation). I didn't bother calling Verizon again, as it's like pulling teeth to get them to do anything other than the basic troubleshooting steps they always go through. Also, I'm now fairly sure the problem is with the internal wiring in the house. Fixing that is something Verizon probably wouldn't cover.
Obviously, this makes SL rather painful to play, which is why I haven't been around as much this summer. Let's take a look.
Now, it's not always like this, just when it's downloading something. So it's like this for maybe 5 - 10 minutes (remember, slow connection) after I teleport somewhere or log in, and maybe 30 seconds whenever a new person arrives. If you're not sure what a ping time of 6476 msec means, it means that if I push a command like 'w' to walk forward, I won't actually move forward until ~6.5 seconds later. The high packet loss rate ensures that a lot of my key presses never even make it to the server...but of course I don't know that until 6.5 seconds later. Add to this the sporadic total loss of internet connectivity, and you have a pretty hostile environment.
Things will finally be fixed when we switch to Comcast cable internet. They are scheduled to install it on September 13th.
For the past year, we had Verizon DSL, which was fine, although rather slow (downloads were about 90 KB/s on average). The problems started towards the beginning of the summer during a lightning storm. I was at my computer, on SL, and there was a popping noise accompied by my computer restarting. I don't remember there being a lightning strike or anything at that moment, but my only guess is that it was a power surge. When my computer finished booting (thank god it booted, the popping noise was not unlike that of a capacitor taking on too much charge), I noticed the internet was not working. I went down to the basement and saw that all the lights were off on the DSL modem, except for the power light which was flashing, and all the router's lights were off. I unplugged them and plugged them back in, tried other outlets around the house, still the same thing. I called Verizon, they sent a new modem, and I bought a new router.
With the new equipment, we could get back online, but the internet was unreliable. We kept losing our connection (dsl light on the modem would flash green, which means it isn't getting a signal from the ISP). It would come on and off, sometimes just for a few minutes, other times for hours at a time. Also, when it was on it was about half the speed as it used to be. I made multiple frustrating calls to Verizon tech support. Finally I convinced one of them that the problem couldn't be fixed by trivial methods such as turning off the modem and turning it back on, but that there was actually a more serious problem. They put in an order for a technician to come out and find the problem.
The technician came out and said that the line going out of our house needed replacing. You'd think he would then go ahead and replace it, but that would be wrong. Instead, he set up another date for someone to come out and fix it. Are these people really so specialized that one person is trained to diagnose a problem, but then someone else needs to actually fix it? So the line was replaced, and I hoped everything would finally work. Wrong again. The speed seems better, although still fluctuated, and the internet still goes on and off (notice I'm switched to present tense, as this is still the current situation). I didn't bother calling Verizon again, as it's like pulling teeth to get them to do anything other than the basic troubleshooting steps they always go through. Also, I'm now fairly sure the problem is with the internal wiring in the house. Fixing that is something Verizon probably wouldn't cover.
Obviously, this makes SL rather painful to play, which is why I haven't been around as much this summer. Let's take a look.
Now, it's not always like this, just when it's downloading something. So it's like this for maybe 5 - 10 minutes (remember, slow connection) after I teleport somewhere or log in, and maybe 30 seconds whenever a new person arrives. If you're not sure what a ping time of 6476 msec means, it means that if I push a command like 'w' to walk forward, I won't actually move forward until ~6.5 seconds later. The high packet loss rate ensures that a lot of my key presses never even make it to the server...but of course I don't know that until 6.5 seconds later. Add to this the sporadic total loss of internet connectivity, and you have a pretty hostile environment.
Things will finally be fixed when we switch to Comcast cable internet. They are scheduled to install it on September 13th.
How I found C:SI
In response to Robby's post, I'm going to tell the story of how I found C:SI, which is actually linked to how I found Second Life, even though I was on SL for a month before coming to C:SI.
The person who showed me SL was my (at the time) girlfriend. I had read a book called Snow Crash, which is a sci-fi book from the early '90s that describes the future of the internet as basically Second Life with virtual reality (in the book it's called the Metaverse, which happens to be the name of a newspaper distributed in SL). Anyway, I had lent it to my girlfriend who liked it a lot, and her Dad had mentioned SL to her (he doesn't play it, but he had read articles about it). She logged in to try it out, told me about it, and then the next day, January 25, 2007, I got my account. She logged in with me and we spent about an hour wandering around as newbies. After that, she never logged in again. She was disappointed that it wasn't exactly like the Metaverse from Snow Crash, and that she couldn't be exactly like the main character, who is the self-proclaimed greatest sword fighter in the world. Also, she recognized it was something she could get obsessed with to the point that it would consume her life. Unlike me and everyone else I know in SL, for which it HAS consumed their lives, she was able to get out while she still could.
I kept coming back, though. Until C:SI a month later, I spent all my time at a place called the Shelter, which is a dance club for newbies. I think what interested me about SL was the social aspect of it. I started graduate school a year ago in a place I had never lived before. All my friends from high school and college were hours away, so I didn't have any friends close by. SL was a place where I could make friends, which is hard when you aren't living in a dorm anymore and just going to campus for classes. One day at the Shelter, a girl showed up with a katana on her back. "Maybe this is more like Snow Crash than I thought," I thought. I asked her where she got it and she gave me a landmark to Samurai Island. I went there, was confused (I don't think I'd bought anything before in SL), and left.
I went back to the Shelter for a while, but I kept thinking I wanted to try the sword fighting. I don't remember exactly what prompted me to actually go out and try again, I think I was just bored one day. I ended up at Bare Rose (I don't remember how, maybe I searched for "katana"?), and asked one of the people working there where to buy a sword. She showed me where they sell them, and I ended up buying a Cursed Spirit. She also gave me a landmark to Shrouded.
I unpacked the sword, read the instructions, and tried swinging it around for a while. Then I went to Shrouded and watched some people fighting. I later realized one of them was Shidji (you can't forget a scary robotic death woman), but I don't remember who the others were. It didn't take long to realize there were clans, and that one should join a clan. I came back the next day and asked someone about joining a clan. That person happened to be Xiang Hifeng (or Kasumi as she is more recently known) who had just started Oda. I joined Oda and the rest, if you'll pardon the cliche, is history.
The person who showed me SL was my (at the time) girlfriend. I had read a book called Snow Crash, which is a sci-fi book from the early '90s that describes the future of the internet as basically Second Life with virtual reality (in the book it's called the Metaverse, which happens to be the name of a newspaper distributed in SL). Anyway, I had lent it to my girlfriend who liked it a lot, and her Dad had mentioned SL to her (he doesn't play it, but he had read articles about it). She logged in to try it out, told me about it, and then the next day, January 25, 2007, I got my account. She logged in with me and we spent about an hour wandering around as newbies. After that, she never logged in again. She was disappointed that it wasn't exactly like the Metaverse from Snow Crash, and that she couldn't be exactly like the main character, who is the self-proclaimed greatest sword fighter in the world. Also, she recognized it was something she could get obsessed with to the point that it would consume her life. Unlike me and everyone else I know in SL, for which it HAS consumed their lives, she was able to get out while she still could.
I kept coming back, though. Until C:SI a month later, I spent all my time at a place called the Shelter, which is a dance club for newbies. I think what interested me about SL was the social aspect of it. I started graduate school a year ago in a place I had never lived before. All my friends from high school and college were hours away, so I didn't have any friends close by. SL was a place where I could make friends, which is hard when you aren't living in a dorm anymore and just going to campus for classes. One day at the Shelter, a girl showed up with a katana on her back. "Maybe this is more like Snow Crash than I thought," I thought. I asked her where she got it and she gave me a landmark to Samurai Island. I went there, was confused (I don't think I'd bought anything before in SL), and left.
I went back to the Shelter for a while, but I kept thinking I wanted to try the sword fighting. I don't remember exactly what prompted me to actually go out and try again, I think I was just bored one day. I ended up at Bare Rose (I don't remember how, maybe I searched for "katana"?), and asked one of the people working there where to buy a sword. She showed me where they sell them, and I ended up buying a Cursed Spirit. She also gave me a landmark to Shrouded.
I unpacked the sword, read the instructions, and tried swinging it around for a while. Then I went to Shrouded and watched some people fighting. I later realized one of them was Shidji (you can't forget a scary robotic death woman), but I don't remember who the others were. It didn't take long to realize there were clans, and that one should join a clan. I came back the next day and asked someone about joining a clan. That person happened to be Xiang Hifeng (or Kasumi as she is more recently known) who had just started Oda. I joined Oda and the rest, if you'll pardon the cliche, is history.
Sunday, 2 September 2007
Fighting against a Naginata
I'm going to expand on my comment here about the naginata in C:SI.
First, I'll just post the comment here:
The two main strengths of the naginata are range and power. Its weakness is speed. A basic idea that can be applied to any fight is that you want to try to take advantage of weaknesses while trying to eliminate strengths. There really isn't a way to take away the power advantage of the naginata, but you can take away the range advantage easily. As long as you stay within katana range of your opponent, then it doesn't matter that they have a longer range. Once you are in range, you can easily use your speed to get in strikes whenever they lower their guard. Unfortunately, most people won't tank against a katana if they are using a naginata. Still, you can catch people off guard by trying to stay close. I think they expect the opponent to try to stay away from them, but doing that would allow the range advantage to come into play.
I think I'll stop here and try to fight against some nagis when I can get back on SL, then post a follow up here of what I learn. Most of my experience fighting naginatas was a while ago, when it was more a sort of novelty. I don't think at that time anyone used it as their main weapon or considered it their specialty. It was more that people were just trying it out for fun because it was different.
First, I'll just post the comment here:
Things have changed with the naginata since I started C:SI back in February. It used to be generally believed that someone with a sword fighting someone with a nagi would usually win, unless the person with the nagi was much more experienced. Now, though, people seem to think that the nagi is stronger in general. I’m not sure why this is. It could be that people such as Mayan and Jane actually advanced the art of the naginata, so people are better with it now. But it could also be that people just aren’t taught how to fight against it. It’s been awhile since I’ve fought a nagi, so I can’t say which I think is true, but I used to be able to fight pretty well against nagis. You definitely can’t win if you do the exact same thing you’d do against a katana, but if you adapt I don’t remember there being any advantage to having a nagi.
The two main strengths of the naginata are range and power. Its weakness is speed. A basic idea that can be applied to any fight is that you want to try to take advantage of weaknesses while trying to eliminate strengths. There really isn't a way to take away the power advantage of the naginata, but you can take away the range advantage easily. As long as you stay within katana range of your opponent, then it doesn't matter that they have a longer range. Once you are in range, you can easily use your speed to get in strikes whenever they lower their guard. Unfortunately, most people won't tank against a katana if they are using a naginata. Still, you can catch people off guard by trying to stay close. I think they expect the opponent to try to stay away from them, but doing that would allow the range advantage to come into play.
I think I'll stop here and try to fight against some nagis when I can get back on SL, then post a follow up here of what I learn. Most of my experience fighting naginatas was a while ago, when it was more a sort of novelty. I don't think at that time anyone used it as their main weapon or considered it their specialty. It was more that people were just trying it out for fun because it was different.
The Circle of Life
I broke my computer last week. I was moving my memory into another slot to see if it would stop making the annoying noise it has been making, but something went wrong. I built this computer myself, meaning that the inside is a massive jungle of IDE and power cables, mostly centered over my memory slots. As a result, I couldn't see the slot very well and was doing most of the work my feel, which turned out to be a bad idea. I thought the memory was in all the way, but it wasn't. This resulted in smoke when I turned the computer on and a few of the metal connectors on the RAM had burned. I'm pretty sure putting any other RAM in that slot afterwards would have had bad results. I could have bought some new memory and maybe a new motherboard, but I just decided it was time for a new computer. After this experience, however, I didn't feel like building one.
I came up with a plan that would allow me to get a better computer than I had before for relatively cheap. First of all, I bought a new LCD monitor last spring, so I didn't need to get a monitor. That's about $250 off the price tag. The second part of the plan was to buy a Dell preloaded with Ubuntu Linux. Linux is a free operating system, which means another few hundred dollars off the price of the computer because I don't have to buy a Windows license. This is my gaming machine, though, and so I actually want Windows on it. I have my old Windows XP CD that I can use to install Windows once I get the computer.
I got the computer yesterday, and discovered there was a flaw in the plan. I boot up the computer with the XP disc and after it loads the necessary drivers it attempts to boot into a minimal Windows XP to do the installation, but crashes with a blue screen of death. After trying some things and searching around on the internet, I finally figure out what the problem is. My Windows XP CD has the original release on it, without SP1 or SP2. It wasn't until SP2 that XP supported PCI-E based graphics cards (all new computers now use PCI-E instead of AGP). So now the plan is to buy a new Windows CD. My plan still saved me money, though, because I can get it pretty cheap through the University. I just wish it wasn't Labor Day weekend.
I came up with a plan that would allow me to get a better computer than I had before for relatively cheap. First of all, I bought a new LCD monitor last spring, so I didn't need to get a monitor. That's about $250 off the price tag. The second part of the plan was to buy a Dell preloaded with Ubuntu Linux. Linux is a free operating system, which means another few hundred dollars off the price of the computer because I don't have to buy a Windows license. This is my gaming machine, though, and so I actually want Windows on it. I have my old Windows XP CD that I can use to install Windows once I get the computer.
I got the computer yesterday, and discovered there was a flaw in the plan. I boot up the computer with the XP disc and after it loads the necessary drivers it attempts to boot into a minimal Windows XP to do the installation, but crashes with a blue screen of death. After trying some things and searching around on the internet, I finally figure out what the problem is. My Windows XP CD has the original release on it, without SP1 or SP2. It wasn't until SP2 that XP supported PCI-E based graphics cards (all new computers now use PCI-E instead of AGP). So now the plan is to buy a new Windows CD. My plan still saved me money, though, because I can get it pretty cheap through the University. I just wish it wasn't Labor Day weekend.
Subscribe to:
Posts (Atom)